Memoirs of Museum Victoria 74:5-15 (2016) Published 2016

ISSN 1447-2546 (Print) 1447-2554 (On-line)

http://museumvictoria.com.au/about/books-and-journals/journals/memoirs-of-museum-victoria/

Quantifying scientific significance of a fossil site: the Gogo Fossil sites (Late

Devonian, Western Australia) as a case study

John A. Long

School of Biological Sciences, Flinders University, GPO Box 2100 Adelaide, South Australia 50901 and Geosciences,

Museum Victoria, GPO Box 666, Melbourne, Victoria 3001, Australia (john.long@flinders.edu.au)

Abstract Long, J.A. 2016. Quantifying scientific significance of a fossil site: the Gogo Fossil sites (Late Devonian, Western

Australia) as a case study. Memoirs of Museum Victoria 74: 5-15.

Assessing the scientific significance of fossil sites has up to now been largely a matter of subjective opinion with

few or no metrics being employed. By applying similar metrics used for assessing academic performance, both qualitative

and quantitative, to fossil sites we gain a real indication of their significance that enables direct comparison with other sites

both nationally and globally. Indices suggested are those using total pages published for both peer-reviewed and combined

peer-reviewed and popular publications, total citations from the papers, total impact points for site (citing palaeontology-

related papers only), total number of very high impact papers (VHIP; journal impact factor>30) and social media metrics.

These provide a measure of how much the published fossil data from a site has been utilised. The Late Devonian fossils of

the Gogo Formation of Western Australia are here used as an example of how these metrics can be applied. The Gogo sites

for example, have produced c.4384 pages of peer-reviewed papers (c.5458 total combined with popular works); generated

papers with a total impact point score of 611, including 10 VHI papers, and generated 4009 citations. The sites have an

overall h-index of 33. Combing these into a Scientific Site Significance Index (SSSI) will permit direct comparisons of site

significance to be made for initiating discussions about site protection, tourism, geopark status, local heritage listings or

potential future world heritage nominations.

Keywords natural heritage, palaeontology. Paleozoic.

Introduction

Tom Rich once said in an interview with Natalie O’Brien of The

Australian, that Gogo was “the most important fossil vertebrate

site in Australia” (The Australian, 2 nd August, 2001, p.4).

Whether or not one agrees with such a statement is completely

arbitrary, but such statements raise a number of questions

concerning how we routinely assess the scientific significance of

a fossil site. Fossil sites are non-renewable resources that may

hold values ranging from the tangible, in terms of scientific

significance and economic values to local tourism, but they may

hold other values in terms of cultural significance when fossil

sites coincide with places valued by indigenous peoples, or be

historically significant in their own right (e.g., Wellington Caves,

New South Wales; Lake Eyre Basin, South Australia; Rich &

Archbold, 1991). Fossils can also hold intangible natural heritage

values making them far more significant than their net scientific

worth as can geological sites (Pena do reis & Henriques, 2009;

Long, 2012a). An example of this is the Gogo fish Mcnamaraspis

(fig. 2A) which in December 1995 became historically significant

as the first state fossil emblem to be formally decreed in Australia

(Long, 2004; https://www.dpc.wa.gov.au/GuidelinesAndPolicies/

Symbol sOf WA/Pages/Fossil EmblemGogoFi sh. aspx).

Although there are several papers evaluating geoheritage

sites (e.g. Pena do reis & Henriques, 2009) there are none the

author knows of that specifically attempt to evaluate the

scientific significance of a fossil site using an applied approach.

To initiate discussion on potential methods to evaluate the

significance of any fossil site, this paper will analyse the Gogo

fossil sites in north Western Australia, of late Devonian age

(Frasnian), in terms of various metrics which allow a

comparative measure of scientific significance as is generally

applied to measuring academic performance. This approach is

based on the premise that a site rich in fossils of high diversity

and significance will have been targeted for primary research.

If the papers are indeed significant they will have been used in

other studies and thus be cited frequently. If the work has high

international significance then the work will most likely have

been published in high impact multidisciplinary science

journals, if the discoveries were made in relatively recent

times. Many highly significant sites lack extensive publications

in these journals as I acknowledge that the need to publish in

high impact journals is a relatively recent phenomenon tied to

increased grant success and even academic promotion. If the

fauna from a site is highly diverse it should also result in large

number of publications being developed, and possibly have

J.A. Long

Figure 1. Map showing area covering the Gogo Formation site localities (geology taken from Long and Trinajstic, 2010, figure 1).

produced large numbers of published pages. As scientific

papers can range from a one-page abstract to several hundred

page monographs, the number of scientific publications on the

fossil fauna and/or flora of a site is here considered of far less

importance than the total numbers of peer-reviewed pages

published. Finally, like routine academic performance, a high

citation rate from papers produced around a site can be

expressed as an h-index (numbers of publications having equal

or greater numbers of citations) for each site. In palaeontology,

unlike other mainstream science disciplines, there is a smaller

pool of active academics (compare with say medicine or

biology in general), so overall citation rates tend to be low.

A fossil site is used here is in the broader context to contain

all individual sites from similar formation defined as follows: a

set of sites within a contained area that is linked by having

fossils of similar preservation coming from the same geological

Quantifying scientific significance of a fossil site: the Gogo Fossil sites (Late Devonian, Western Australia) as a case study

Figure 2. Examples of Gogo fish preservation (WAM = Western Australian Museum; NMV P = Museum Victoria Palaeontology Collection). A,

eubrachythoracid placoderm Mcnamaraspis kaprios, the state fossil emblem of Western Australia (WAM 86.9.676, from Long, 1995). B, The

tetrapodomorph fish Gogonasus andrewsae (NMV P221807). C, the ptyctodontid placoderm Austroptyctodus gardineri which has 3 embryos preserved

inside it (WAM 86.9.662). D, palatal view of the lungfish ‘Chirodipterus’australis (most likely a new genus). WAM 90.10.8. Scale bars are 1 cm.

formation, of approximate same stratigraphic age, and same

palaeoenvironment. The Gogo sites comprise a large number

of individual sites across an area of about 200km 2 , so the term

“Gogo fossil sites” refers to a suite of sites containing well-

preserved fossils all found from surface limestone concretions

derived from exposures of the Gogo Formation (fig. 1). Most of

the better known sites are listed by Miles (1971). The sites

comprises one of only 6 recognised Lagerstatten (sites of

exceptional quality fossil preservation), designated for the

known fossil fish sites of this age globally, as listed in the

supplementary information (Anderson et al., 2011).

In this paper the measure of that significance is discussed

with respect to factors that apply to measuring academic

outputs in terms of both quantitative (numbers of significant

specimens as a corollary of numbers of significant scientific

papers) and qualitative measures (quality of preservation,

disparity of species variation also translating into a diversity

of scientific papers). The metrics applied are provided in

Supplementary Appendix 1. Natural History collections can

be evaluated in a similar way, as shown by Winkler and

Withrow (2013). By creating a Google Scholar profile for the

bird collections at the University of Alaska Fairbanks Museum,

they were able to show that the body of work supported by the

collection had a profile h-index of 42, equivalent to that of an

average Nobel Laureate in Physics. Similarly the Louisiana

Museum of Natural History bird collections have their own

Google scholar page that now records an h index of 69 (based

on 25,469 citation as of December 1 st , 2014). These measures

thus send a positive message of how significant these

collections are by virtue of how well used and cited is the work

on its specimens. A similar approach is proposed in this paper

to show how well utilised the scientific papers generated from

a particular fossil site can be applied to give an indication of

the site’s scientific significance.

J.A. Long

It is important to be able to quantifiably measure the

relative significance of a fossil site for varying applications:

assessing heritage status for local conservation planning, for

issuing permits for collecting for scientific work in national

parks (or when sites fall under the auspices of local shire land

management); and for potential future nominations for national

heritage registers and international recognition including

potential future world heritage nominations.

Previously fossil sites have been assessed for the old

how expert assessors with specialist knowledge of the site and

its fauna position them within an arbitrary framework relative

to other known sites. The problem with this is that the

individual specialist who knows a particular site well may not

necessarily be familiar with the full body of scientific literature

on other sites of different age and faunal composition. This

informal placement of sites as ‘highly significant’ or extremely

important lacks a quantitative approach that can be used to

argue that other sites that are potentially above the average

local or national significance, and thus help target future sites

in need of some form of legislative protection against site

damage by non-professional collectors or fossil dealers (see

Long, 2002, for case examples). The Gogo sites have been

chosen as the case study in this work mainly because of the

author’s familiarity with the site, having collected at the

localities for the past 30 years and worked on its vertebrate

fauna for the past 30 years, and from being familiar with its

extensive scientific and popular literature.

The Gogo Fossil Sites and Inferred Scientific Significance

The significance of the Gogo Fossil sites were recognised in

the 1960s when the acetic acid preparation technique was

refined at the Natural History Museum, London by Harry

Toombs (Toombs, 1948). Toombs was able to prepare 3-D

skeletons of Devonian fishes out of limestone, revealing

perfect 3-D shape and form. Furthermore, in recent years

preservation of fossil impressions of soft tissues including

muscle fibres (Trinajstic et al., 2007, 2013), nerve cells

(Trinasjtic et al., 2007), umbilical structures and embryos

(Long et al., 2008, 2009, Trinajstic et al., 2014; also see fig 2C)

and alimentary structures (Long and Trinajstic, 2010) have

been identified. Examples of the 3-D preservation of Gogo

fishes and embryos is shown in figure 2. Biomarkers have

recently been identified in Gogo crustaceans preserving

proteins only found in living crustaceans (Melendez et al.,

2013). The real utilisation of the Gogo fossil fish fauna though

is due to the clear unambiguous preservation of the bony

skeletons. It has been widely utilised for detailed histological

studies of early vertebrate tissues (e.g. Smith, 1977; Smith &

Campbell, 1987 etc.), as well as inclusion in major phylogenetic

papers analysing character distributions (Miles and Dennis,

1979; Miles and Young, 1977, Long et al., 2014). The landmark

paper on transformed cladistics by Rosen et al. (1981) utilised

Gogo lungfish material to press home certain points about the

homology of the tetrapod choana. Today this paper has been

cited nearly 400 times (Google Scholar) and remains a seminal

work on the topic.

In terms of non-quantifiable highlights of the Gogo fauna’s

significance, this includes a series of world first or unique

occurrences of species, genera and families. The Gogo fauna

contains a mostly endemic fauna of around 90% unique genera

and species. It contains the world’s only known record of

camuropiscid and inscisoscutid arthrodires. It also contains

the highest diversity of lungfishes (c. 12 spp.) and

actinopterygians (c. 5 spp.) for any site of similar age. In its

high diversity of vertebrates (c. 50 spp.) it is well above any

other site of similar age, including the World Heritage

Miguasha site in Canada (20 fish species; Cloutier, 2010).

From a purely intuitional perspective one can sense the value

and degree of scientific significance of such a uniquely well-

preserved, diverse and endemic fauna is obviously high. The

question is how do we measure this?

Gogo Sites: Actual Measures of Scientific Significance

The premise of this work implies that the diversity of a fauna

is reflected in the numbers of papers on the fauna, and the

numbers of pages of peer-reviewed publications. Works of

monographic stature indicate seminal papers that are widely

cited as the key reference for the study group (e.g., Gardiner’s

1984 monograph on Gogo actinopterygians has 250 citations).

Thus sites of high scientific significance would be expected to

generate not only a lot of papers, but large descriptive papers

and therefore highly cited papers. Large monographic works

are amongst some of the most highly cited works in

palaeontology, and on-line journals like Paleontologia

Electronica still publish such works without charge to authors.

A good measure for the total significance of a body of work

centred around a topic is the cumulative impact factor points

of the journals they have been published in. The academic

website ResearchGate uses this approach by adding up the

impact factor points for an individual’s body of work to give a

total tally. Using this approach we find that the Gogo sites have

an impact factor points tally of around 611. The method used

in the scoring was to assign low ranking journals (IF=0-1) as

a score of 1, and then round up or down any other impact

factors to the nearest whole numbers.

Assessing very significant fossil specimens. A fossil that solves

a major evolutionary problem or bridges a major morphological

boundary as a key transitional form will attract high numbers of

citations. For example the discovery of Tiktaalik, which was

found to be the immediate ancestor of all living tetrapods, was

published in the journal Nature as two back-to-back articles in

April 2006 (it was also the cover story). To date these two papers

on Tiktaalik , now one of the most well-known and iconic fossil

discoveries of the 20 th century, have received 255 citations

(Daeschler et al., 2006) and 188 citations (Shubin et al., 2006)

respectively. Yet a third paper on Tiktaalik, also published in

Nature (Downes et al., 2008), has only received 25 citations.

Another example is the discovery of Homo floresiensis, the so-

called ‘hobbit’. The initial paper announcing the discovery by

Brown et al. (2004) in Nature has now received around 609

citations. Using these as a reasonable basis for comparison, we

see that the Gogo fauna’s most highly cited descriptive papers

(Miles, 1977, 188 citations; Gardiner, 1984, 250 citations),

Quantifying scientific significance of a fossil site: the Gogo Fossil sites (Late Devonian, Western Australia) as a case study

though not published in high impact journals, still yield high

citations directly comparable to the original Tiktaalik papers.

Very High Impact Papers. One measure of the international

significance of fossil specimens is whether they are valuable in

solving a major phylogenetic or biogeographic problem, or

provide new information about evolutionary biology deemed

highly significant. In such cases the work is occasionally

accepted in the highest ranking interdisciplinary science

journals like Nature or Science. Of course, this is not always the

case as some very significant discoveries get routinely rejected

by such journals. Nonetheless, as these journals publish only a

handful of palaeontological papers each year (e.g., Nature

published about 10 palaeontological papers in 2014), each paper

is judged to be a highly significant breakthrough worthy of high

impact publication by both the board of editors and reviewers of

these papers. These are thus here given special attention as

‘very high impact papers’ (impact factor as of 2013 > 30; Nature,

IF=42; Science IF=31).

A measure of international significance for a site can

therefore be also gauged by the total number of very high impact

papers (VHIP) published on its fossil materials. For the Gogo

sites, this amounts to 10 such papers (9 on primary materials,

plus one review paper on the site’s significance, Ahlberg, 1989;

see Supplementary Appendix 1). This is more VHIPs than any

other fossil site in Australia (for comparison, Riversleigh World

Heritage fossil mammal site has 4 VHIPs, Victoria’s Early

Cretaceous vertebrate sites have 4 VHIPs, Ediacara c. 3 VHIPs).

Excluding African hominid sites, the only other fossil sites in

the world that immediately come to mind to have 10 or more

VHIPs published on their faunas would include the Jehol Biota

of China (Liaoning sites, covering a very wide range of sites and

stratigraphic horizons), and the Burgess Shale sites in Canada.

Numbers of papers/pages published. This gives an overall

estimate of the quantity, but not quality, of peer-reviewed work

published from a site. Naturally a large number of papers reflect

either a diverse fauna, or that continuous new data is being

described from a site. This suggests the site hasn’t yet peaked

in terms of yielding scientifically significant new specimens. In

some cases work might be published as a series of monographs,

which limits the overall numbers of papers published but

increases the total number of pages of published work. For

Gogo this is a large number: 4389 pages. This excludes any

books that are not specifically on the Gogo fossil fauna. This

doesn’t include papers on the geology of the site, but only

papers that primarily describe Gogo fossils (fishes plus

invertebrates and microfossils) or figure Gogo specimens in

elucidating the descriptions of other early vertebrate specimens.

Other measures of the cultural and scientific significance of

a site

Books and popular magazine articles published. Both technical

and popular books centred around the fossil biota of a site can

be used to gauge its significance as only books that publishers

see worthy of competing in the market will be published by

mainstream or academic publishers. In other words a site

yielding information that is a consumable product for the

general public rather than just for a specific scientific audience

will get published by major publishers (excluding self-published

projects). In such cases, we can also corroborate publication

about the site in a number of popular science magazine

publications (e.g., New Scientist, Scientific American, Discover,

Cosmos, Australasian Science etc.) as well as being subject

material for popular books for the adult lay audience. This has

been achieved for the Gogo fauna, with many articles appearing

in the international popular science journals (eg New Scientist,

April 1989; Scientific American, cover story for January 2011).

This includes books written about the evolution of fishes

featuring Gogo specimens (Long, 1995, 2011a), as well as a

history and significance of the Gogo fossil discoveries (Long,

2006), a book telling the story of the origins of copulation

(Long, 2011b, 2012b) and a children’s book about the story of

Gogo fish becoming a state fossil emblem (Long, 2004).

Media Focus on the Site. The international significance of a

fossil site can also be measured in terms of how much national

and international media coverage the site has generated through

inclusion in general documentary programs covering broad

topics like vertebrate evolution or the prehistory of a country.

The Gogo sites featured in David Attenborough’s 1979 series

Life on Earth (episode 4, fishes), and in recent years featured on

other local documentaries made about Australia’s prehistoric

past (e.g., Richard Smith’s Australia: Time Traveller series,

ABC TV, 2012). Gogo fish fossils have also been the topic of 3

features screened on the ABC TV’s science program Catalyst

(formerly called ‘ Quantum ’). Social media could be another

way to measure impact or visibility of a site, using popular web

blog sites like ‘The Conversation’ to highlight the significance

of certain sites. Altmetrics from such sites record numbers of

hits, tweets and media pick-ups on each article.

Tourism. The amount of tourism to a site (e.g., annual number of

visitors), funds raised by visitations, sales, and so on could also

be applied if the site was open to the public (e.g., Naracoorte

Caves world heritage site). These metrics really only apply to the

sites that can capture such visitations and their associated funding.

A Proposed Site Significance Index (SSSI)

One way of generating a usable index of scientific significance

is to arbitrarily combine all the metrics discussed above into a

formula that smooths out the very high numbers with the low

but significance indices.

Ppr=pages published (peer-reviewed) is a large number for

Gogo, so Ppr/100 gives 4389/100 =43.89, rounded up to 44.

Ip=Impact points 611 is also a large number so Ip/10 gives

611/10=61.

Cn=Total citations is also a large number, so Cn/100 gives

4009/100=40 (rounded).

VHIP=very high impact papers is always a relative low number

so needs to be multiplied by 10 for a reasonable comparison

with other sites where VHIP might regularly be less than 5.

For Gogo this gives 10x10=100. This measure is seen to be

highly significant as an index of globally significant papers.

J.A. Long

The suggested formula for assessing site significance that

brings these factors into account would be: SSSI=(Ppr/100)+(

Cn/100)+(Ip/10)+(VHIP xlO).

For the Gogo sites as assessed at the time of submission of this

work, this score is

SSSI=44+40+61+100=245.

The meaning of such a score can only be assessed when

further work is completed on other significant fossil sites to

measure and compare the same metrics. As this paper was

intended primarily to be a generator of discussion, I hope that

this will incite other researchers to score other fossil sites they

have worked on, particularly those who have focussed on a

special fossil site for most of their working careers so that the

body of published information on the site is captured. This

will enable other fossil sites to be eventually compared with

one another using the metric approach here outlined. It is also

hoped that further discussion on this topic will be generated

by this paper to determine if the parameters chosen herein are

suitable enough or if other factors need be scored and added

into the mix to give a more meaningful assessment of a fossil

site’s scientific significance.

Acknowledgements

This paper is dedicated to Dr Thomas H. Rich who has been

one of the most inspirational mentors towards my early career

development. Tom dedicated many years of his life working

the Victorian Cretaceous fossil sites in search of our country’s

oldest mammals, and he eventually found them after many

years of hard searching. He has taught me and many of my

colleagues the value of persistence in working a site year after

year until the very significant fossils are eventually found. The

MS has benefited from discussions with Erich Fitzgerald,

Gavin Prideaux, Gavin Young and Kate Trinajstic.

References

Ahlberg PE. 1989. Fossil fishes from Gogo. Nature 337: 511-512.

Anderson, P., Friedman, M., Brazeau, M. and Rayfield, E.J. 2011.

Initial radiation of jaws demonstrated stability despite faunal and

environmental change. Nature 476: 206-209.

Brown, P., Sutikna, T., Morwood, M.J., Soejono, R.P., Jatmiko, Saptomo

E.W., and Due, R.A. 2004. A new small-bodied hominin from the

Fate Pleistocene of Flores, Indonesia. Nature 431: 1055-1061.

Cloutier, R. 2010. The late Devonian Biota of the Miguasha national

park UNESCO world heritage site. AAAP Search and Discovery

Article #90172, GeoConvention, Calgary, Alberta, May 10-14.

Ppl-4 (on-line).

Daeschler, E.B., Shubin, N.H., and Jenkins, F.A. 2006. A Devonian

tetrapod-like fish and the evolution of the tetrapod body plan.

Nature 440: 757-763.

Downes, J., Daeschler, E.B., Jenkins, F.A. and Shubin, N.H. 2008. The

cranial endoskeleton of Tiktaalik rosae. Nature 455: 925-929.

Gardiner, B.G. 1984. Relationships of the palaeoniscoid fishes, a

review based on new specimens of Mimia and Moythomasia from

the Upper Devonian of Western Australia. Bulletin of the British

Museum of Natural History (Geology) 37: 173-428.

Fong, J.A. 1995. The Rise of Fishes, 500 Million years of Evolution.

UNSW Press, Johns Hopkins University Press, 224 pp.

Fong, J.A. 2002. The Dinosaur Dealers: mission, to uncover

international fossil smuggling. Allen & Unwin, Sydney, 220 pp.

Fong, J.A. 2004. Gogo Fish! The story of the state fossil emblem of

Western Australia. The Western Australian Museum, Perth. 48 pp.

Fong, J.A. 2011a. The Rise of Fishes, 500 Million years of Evolution.

2 nd ed. UNSW Press, Johns Hopkins University Press, 288 pp.

Fong, J.A. 2011b. Hung like an Argentine duck: a journey back in

time to the origins of sexual intimacy. Fourth Estate,

HarperCollins, Sydney. 278 pp.

Fong, J.A. 2012a. Case Studies of Intangible Natural Heritage from

Museum Collections. Pp. 43-55 in Intangible Natural Heritage:

New Perspectives on Natural Objects, E. Dorfman (ed.),

Routledge Press, New York.

Fong, J,A, 2012b. Dawn of the Deed: the prehistoric origins of sex.

University of Chicago Press, Illinois. 278 pp.

Fong, J.A. and Trinajstic, K. 2010. The Late Devonian Gogo Formation

Lagerstatte -Exceptional preservation and Diversity in early

Vertebrates. Annual Reviews of Earth and Planetary Sciences 38:

665-680.

Long, J.A., Trinajstic, K.M. and Johanson, Z. 2009. Devonian

arthrodire embryos and the origin of internal fertilization in

vertebrates. Nature 457: 1124-112.

Long, J.A., Trinajstic, K.M., Young, G.C. and Senden, T. 2008. Live

birth in the Devonian period. Nature 453: 650-652.

Long, J.A., Mark-Kurik, E., Johanson, Z., Lee, M.S.Y., Young, G.C.,

Zhu, M., Ahlberg, P.E., Newman, M., Jones, R., den Blaauwen, J.,

Choo, B., and Trinajstic, K. 2015. Copulation in antiarch

placoderms and the origin of gnathostome internal fertilization.

Nature 517: 196-199.

Melendez, I., Grice, K., Trinajstic, K., Ladjavardi, M., Greenwood, P.,

and Thompson, K. 2013. Biomarkers reveal the role of photic zone

euxiniain exceptional fossil ptreservation: an organic geochemical

perspective. Geology 41: 123-126.

Miles, R.S. 1971. The Holonematidae (placoderm fishes): a review

based on new specimens of Holonema from the Upper Devonian

of Western Australia. Philosophical Transactions of the Royal

Society of London 263B: 101-234.

Miles, R.S. 1977. Dipnoan (lungfish) skulls from the Upper Devonian

of Western Australia. Zoological Journal of the Linnean Society

61: 1-328.

Miles, R.S. and Dennis, K. 1979. A primitive eubrachythoracid

arthrodire from Gogo, Western Australia. Zoological Journal of

the Linnean Society 66: 31-62

Miles, R.S. and Young, G.C. 1977. Placoderm interrelationships

reconsidered in the light of new ptyctodontids from Gogo Western

Australia. Zoological Journal of the Linnean Society, Symposium

Series 4: 123-198.

Pena dos Reis, R. and Henriques, M.H. 2009. Approaching an

integrated qualification and evaluation system for geological

heritage. Geoheritage 1: 1-10.

Rich, PV. and Archbold, N.W. 1991. Squatters, priests and professors:

a brief history of vertebrate palaeontology in Terra Australis.

Ch.l in: The Vertebrate Palaeontology of Australasia, PV. Rich,

J. Monaghan, R.F. Baird & T. Rich, (eds). Pioneer Design Studios.

Lilydale: pp. 1-44.

Rosen, D.E., Forey, P.L., Gardiner, B.G. and Patterson, C. 1981.

Lungfishes, tetrapods, palaeontology and plesiomorphy. Bulletin

of the American Museum of Natural History 167: 159-276.

Shubin, N.H., Daeschler, E.B. and Jenkins, F.A. 2006. The pectoral

fin of Tiktaalik rosae and the origin of the tetrapod limb. Nature

440: 764-771.

Quantifying scientific significance of a fossil site: the Gogo Fossil sites (Late Devonian, Western Australia) as a case study

Smith, M.M. 1977. The microstructure of the dentition and dermal

ornament of three dipnoans from the Devonian of Western Australia:

a contribution towards dipnoan interrelations, and morphogenesis,

growth and adaptation of the skeletal tissues. Philosophical

Transactions of the Royal Society of London 28 IB: 29-72.

Smith, M.M., and Campbell, K.S.W. 1987. Comparative morphology,

histology and growth of dental plates of the Devonian dipnoan

Chirodipterus. Philosophical Transactions of the Royal Society

of London 317B: 329-363.

Toombs, H. A. 1948. The use of acetic acid in the development of

vertebrate fossils. Museums Journal 48: 54-55.

Trinajstic, K., Boisvert, C., Long, J., Masimenko, A. and Johanson, Z.

2014. Pelvic and reproductive structures in placoderms (stem

gnathostomes). Biological Reviews DOI: 10.111 l/brv.12118. 35 pp.

Trinajstic, K., Marshall, C., Long, J.A.and Bifield, K. 2007.

Exceptional preservation of nerve and muscle tissues in Late

Devonian placoderm fish and their evolutionary implications.

Biology Letters 3: 197-200.

Winkler, K. and Withrow, J.J. 2013. Natural history: small collections

make a big impact. Nature 493. doi:10.1038/493480b (published

online 23 Jan, 2013).

J.A. Long

Supplementary Appendix 1.

Gogo Fossil publications (palaeontology only, papers primarily

concerned with palaeontology or fossil preservation/

diagenesis, not primarily geology or background information).

Includes page numbers, impact factors, and citations. Impact

factors (IF) based on 2013 scores rounded up or down to

nearest full number (eg 2.34=2, 2.56 =3), journals with no

recorded impact factor or those lower than 1 are here allocated

an arbitrary score of T; citations from google scholar (cn, as

of Nov 20-21, 2014). Incudes primary references on Gogo

fossils, major review papers on Gogo fossils, plus peer-

reviewed papers illustrating Gogo specimens to assist with

morphological interpretation of other materials. Only

publications which have citation and or impact factor metrics

are recorded.

VERY HIGH IMPACT PAPERS (IF>30)

(full references cited below).

1. Rolfel966 (Nature)

2. Ahlberg 1989 (Nature)

3. Smith & Johanson 2003 (Science)

4. Long 2006 (Nature)

5. Long et al 2008 (Nature)

6. Long et al 2009 (Nature)

7. Ahlberg et al 2009 (Nature)

8. Rucklin et al 2012 (Nature)

9. Trinajstic et al 2013 (Science)

10. Long et al 2014 (Nature)

TOTAL Impact factor points added as scored by system

defined above= 611.

TOTAL pages published (peer-reviewed papers only) -c (less

popular books) = 4389pp.

^Including popular books featuring a lot of Gogo material:

Rise of Fishes 2nd ed. 288pp, Dawn of the Deed, 278pp. Rise

of Fishes 1 (188pp), Swimming in Stone (320pp) = 4384 plus

1074 = 5458 pp.

TOTAL citations, google scholar = 4009 citations.

Gogo Site h factor =33

Papers with citations=>32

1. Andrews et al 2006 (cn=34); 2. Campbell Barwick 1983

(cn=41); 3. Campbell, Barwick 1987 (cn=87); 4. Campbell,

Barwick 1988b (cn=37); 5. Dennis Miles 1979a (cn=36);

6. Dennis Miles 1981 (cn=43); 7. Dennis-Bryan 1987

(cn=64); 8. Druce 1976 (cn=77); 9. Gardiner 1973 (cn=57);

10. Gardiner 1984 (cn=250); 11. Gardiner & Bartram

19977 (cn=39); 12. Glenister 1958 (cn=51); 13. Glenister,

Klapper 1966 (cn=124); 14. Long 1995 (cn= 121); 15.

Long 1997 (cn=36); 16. Long et al. 1997 (cn=64); 17. Long

et al 2009 (cn=40); 18. Long et al. 2008 (cn=52); 19. Long

et al. 2006 (cn=66); 20. Marshall et al.1986 (cn=39); 21.

Miles 1971 (cn=80); 22. Miles 1977 (cn=188); 23. Miles

Dennis 1979 (cn=45); 24. Miles & Young 1977 (cn=55);

25. Nazarov et al. 1982 (cn=48); 26. Nicoll 1977 (cn=42);

27. Rosen et al.1981 (cn^392); 28. Smith 1977(cn=48); 29.

Smith 1979 (cn=48); 30. Young 1984 (cn=43); 31. Young

1986- 94; 32. Sanchez et al 2012 -33; 33. Trinajstic et al

2007- 34

BIBLIOGRAPHY OF GOGO FOSSIL PAPERS

Not inclusive of primarily geological papers. As mentioned

above, the list below contains primary references on Gogo

fossils, major review papers on Gogo fossils, plus peer-

reviewed papers illustrating Gogo specimens to assist with

morphological interpretation of other materials.

Ahlberg PE. 1989. Fossil fishes from Gogo. Nature 337, 511-512.

(IF=42, cn=6, pp=3).

Ahlberg PE, Smith MM, Johanson Z. 2006. Developmental plasticity

and disparity in early dipnoan (lungfish) dentitions. Evol.

Devel. 8:331-349 (IF= 3, cn=32, pp=19)

Ahlberg PE, Trinajstic K, Johnason Z, Long, JA. 2009. Pelvic claspers

confirm chondrichthyan-like internal fertilisation in arthrodires.

Nature 459: 888-889 (IF =42; cn=21, pp=2).

Anderson P. 2008. Shape variation between arthrodire morphotypes

indicates possible feeding niches. J. Vert. Paleontol. 28:961-969

(IF=2, cn=8, pp=9).

Andrews SM, Long J, Ahlberg P, Barwick R, Campbell K. 2006. The

structure of the sarcopterygian Onychodus jandemarrai n.sp.

from Gogo, Western Australia: with a functional interpretation of

the skeleton. Trans. R. Soc. Edinb. (Earth Sci.) 96: 197-307 (IF=1;

cn=34, pp=110).

Barwick R, Campbell KSW. 1996. A Late Devonian dipnoan,

Pillararhynchus, from Gogo, Western Australia, and its

relationships. Palaeontograph. 239A: 1-42 (IF=1, cn=18, pp=42).

Cn= 129

Briggs DEG, Rolfe 1.1983. New Concavicarida (new order:?Crustacea)

from the Upper Devonian of Gogo, Western Australia, and the

palaeoecology and affinities of the group. Spec. Pap. Palaeont.

30: 249-276 (IF=2; cn=26, pp=28).

Brunton CHC, Miles RS, Rolfe WDI. 1969. Gogo Expedition 1967.

Proc. Geol. Soc. Lond. 1655: 79-83 (IF=1, cn=l, pp =5).

Burrow, C. J., Trinajstic, K. & Long, J.A. 2012. First acanthodian from

the Upper Devonian (Frasnian) Gogo Formation of Western Australia.

Historical Biology, ifirsDOI: 10.1080/08912963.2012.660150, 1-9.

(IF=1, cn=2, pp=9)

Campbell KSW. 1981. Lungfishes - alive and extinct. Field Mus. Nat.

Hist. Bull. Sept. 1981: 3-5. (no IF, cn=5, pp=2).

Campbell KSW, Barwick RE. 1982. The neurocranium of the

primitive dipnoan Dipnorhynchus sussmilchi (Etheridge). J. Vert.

Paleontol. 2: 286-327 (IF=2, cn=30, pp=32).

Campbell KSW, Barwick RE. 1983. Early evolution of dipnoan

dentitions and a new species Speonesydrion. Mems. Ass.

Australas. Palaeontols 1: 17-49 (IF=2, cn=41, pp=33). 230

Campbell KSW, Barwick RE. 1984a. The choana, maxillae,

premaxillae and anterior bones of early dipnoans. Proc. Linn.

Soc.N.S.W. 107: 147-170 (IF=1, cn=25, pp=24).

Campbell KSW, Barwick RE. 1987. Palaeozoic lungfishes - a review.

J. Morph. Suppl. 1: 93-131. (IF=2, cn=87, pp=38).

Campbell KSW, Barwick RE. 1988a. Geological and palaeontological

information and phylogenetic hypotheses. Geol. Mag. 125: 207-

227 (IF=2, cn=27, pp=20).

Campbell KSW, Barwick RE. 1988b. Uranolophus: a reappraisal of a

primitive dipnoan. Mems. Ass. Australas. Palaeontols 7: 87-144

(IF=1, cn=37, pp=61).

Quantifying scientific significance of a fossil site: the Gogo Fossil sites (Late Devonian, Western Australia) as a case study

Campbell KSW, Barwick RE. 1990. Palaeozoic dipnoan phylogeny:

functional complexes and evolution without parsimony. Paleobiol.

16: 143-169 (IF=3. cn=55, pp=27).

Campbell KSW, Barwick RE. 1991. Teeth and tooth plates in primitive

lungfish and a new species of Holodipterus. In Early vertebrates

and related problems of evolutionary biology, ed. MM Chang,

Liu, YH, Zhang GR, pp. 429-440. Beijing: Science Press, (no IF;

cn=15, pp=12).

Campbell KSW, Barwick RE. 1995. The primitive dipnoan dental

plate. J. Vert. Palaeontol. 15: 13-27 (IF=2, cn=14, pp=14).

Campbell KSW, Barwick RE. 1998. A new tooth-plated dipnoan from

the Upper Devonian Gogo Formation and its relationships. Mems.

Queensl. Mus. 42: 403-437 (no IF, cn=23, pp=33). 513

Campbell KSW, Barwick RE. 1999. Dipnoan fishes from the Late

Devonian Gogo Formation of Western Australia. Rees. West.

Aust. Mus., Suppl. 57: 107- 138 (IF=1, cn=18, pp=32).

Campbell KSW, Barwick RE. 2002. The axial postcranial structure of

Griphognathus whitei from Gogo; comparisons with other Devonian

dipnoans. Rees. West. Aust. Mus. 21:167-201 (IF=1, cn=6, pp=35

Campbell, K. S. W. & Barwick, R. E. 2006. Morphological innovation

through gene regulation: an example from Devonian

Onychodontiform fish. International Journal of Developmental

Biology 50: 371-375 (IF=3, cn =9, pp=5).

Campbell KSW, Barwick RE, JL den Blaauwen. 2006. Structure and

function of the shoulder girdle in dipnoans: new material from

Dipterus valenciennesi. Senckenberg. Leth. 86:77-91 (IF=1,

cn=4, pp=15).

Campbell KSW, Smith MM. 1987. The Devonian dipnoan

Holodipterus: dental variation and remodelling growth

mechanisms.J'tecs. Aust. Mus. 38: 131-67 (IF=1; cn=20, pp=37).

Cheng H. 1989. On the tubuli in Devonian lungfishes. Alcheringa 13:

153-166 (IF=1; cn=6, pp=14).

Choo, B. 2011. Revision of the actinopterygian genus Mimipiscis

(=Mimia) from the Upper Devonian Gogo Formation of Western

Australia and the interrelationships of the early Actinopterygii.

Earth and Environmental Science Transactions of the Royal

Society of Edinburgh 102: 1-28 (IF=1, cn=10, pp=28).

Choo, B. 2015., A new species of the Devonian actinopterygian

Moythomasia from Bergisch-Gladbach, Germany and fresh

observations on M. durgaringa from the Gogo Formation of

Western Australia. Journal of Vertebrate Paleontology35: 1-20

(IF=2.0, cn =0, pp=20)

Choo B, Long J, Trinajstic K. 2009. A new genus and species of basal

actinopterygian fish from the Upper Devonian Gogo Formation of

Western Australia Act. Zool. 90: 194-210 (IF=1; cn=12, pp=17). 592

Clement, A., Long, J.A. 2010. Xeradipterus hatcheri, a new dipnoan

from the late Devonian (Frasnian) Gogo Formation, Western

Australia, and other new holodontid material. Journal of

Vertebrate Paleontology 30: 681-695. (IF=2, cn=2, pp=15).

Clement, A. 2012. A new species of long-snouted lungfish from the

Late Devonian of Australia, and its functional and biogeographical

implications. Palaeontology 55, 51-71 (IF=2, cn=0, pp=20).

Clement, A. & Long, J.A. 2010. Air-breathing adaptation in a marine

Devonian lungfish. Biology Letters 6:509-512 (IF=3, cn=15,

PP=4).

Dennis KD, Miles RS. 1979a. A second eubrachythoracid arthrodire

from Gogo, Western Australia. Zool. J. Linn. Soc. 67: 1-29 (IF=3,

cn=36, pp=29).

Dennis KD, Miles RS. 1979b. Eubrachythoracid arthrodires with

tubular rostral plates from Gogo, Western Australia. Zool. J. Linn.

Soc. 67: 297-328 (IF=3, cn=29, pp=32).

Dennis KD, Miles RS. 1980. New durophagous arthrodires from

Gogo, Western Australia. Zool. J. Linn. Soc. 69: 43-85 (IF=3;

cn=28, pp=43).

Dennis KD, Miles RS 1981. A pachyosteomorph arthrodire from

Gogo, Western Australia. Zool. J. Linn. Soc. 73: 213-258 (IF=3,

cn=43, pp=46).

Dennis KD, Miles RS. 1982. A eubrachythoracid arthrodire with a

snub-nose from Gogo, Western Australia. Zool. J. Linn. Soc. 75:

153-166 (I IF=3, cn=15, pp=14).

Dennis-Bryan K. 1987. A new species of eastmanosteid arthrodire

(Pisces: Placodermi) from Gogo, Western Australia Zool. J. Linn.

Soc. Society 90: 1-64 (IF=3, cn=40, cn=64, pp=64).

Dennis-Bryan, K. Miles RS. 1983. Further eubrachythoracid

arthrodires from Gogo, Western Australia. Zool. J. Linn. Soc. 77:

145-173 (IF=3, cn=23, pp=28). 947

Druce EC. 1976. Conodont biostratigraphy of the Upper Devonian

reef complexes of the canning basin. Western Australia. Bur. Min.

Res. Aust. Bull. 158: 1-303 (no IF, cn=77, pp=303). 1024

Forey PL, Gardiner BG.1986. Observations on Ctenurella

(Ptyctodontida) and the classification of placoderm fishes. Zool. J.

Linn. Soc. 86: 43-74 (IF=3, cn=29, cn=26, pp=34).

Freidman M. 2007a. The interrelationships of Devonian lungfishes

(Sarcopterygii; Dipnoi) as inferred from neurocranial evidence

and new data from the genus Soederberghia Lehman 1959. Zool.

J. Linn. Soc. 151:115-171 (IF=3, cn=26, pp= 56).

Friedman M. 2007c. Cranial structure in the Devonian lungfish

Soederberghia groenlandica and its implications for the

interrelationships of ‘rhynchodipterids”. Trans. R. Soc. Edinb.

(Earth Sci.) 98: 179-198 (IF=1, cn=4, pp=20).

Gardiner BG. 1973. Interrelationships of teleostomes. In

Interrelationships of Fishes, ed. PH Greenwood, RS Miles, C

Patterson. Zool. J. Linn. Soc. 52 (supplement): 105-135 (IF=3,

cn=57, pp=31).

Gardiner BG. 1984. Relationships of the palaeoniscoid fishes, a review

based on new specimens of Mimia and Moythomasia from the

Upper Devonian of Western Australia. Bull. Brit. Mus.. Nat. Hist.

(Geol.) 37: 173-428 (no IF, cn=250, pp=256).

Gardiner BG, Bartram AWH.1977. The homologies of ventral cranial

fissures in osteichthyans. In Problems in early vertebrate

evolution ed. SM Andrews, RS Miles, AD Walker, pp. 227-245.

London: Academic Press (no IF, cn=39, pp=19).

Gardiner BG, Miles RS. 1975. Devonian fishes of the Gogo Formation,

Western Australia. Coll, internat. C.N.R.S. 218: 73-79 (no IF,

cn=25, pp=7). 1454

Gardiner BG, Miles RS. 1990. A new genus of eubrachythoracid

arthrodire from Gogo,Western Australia. Zool. J. Linn. Soc 99:

159-204 (IF=3, cn=27, pp=46).

Gardiner BG, Miles RS. 1994. Eubrachythoracid arthrodires from

Gogo, Western Australia. Zool. J. Linn. Soc.112: 443-477 (IF=3,

cn=24, pp=35).

Glenister BF. 1958. Upper Devonian ammonoids from the

Manticoceras zone. Fitzroy Basin, Western Australia. J. Paleont

32: 58-96 (IF=1, cn=51, pp=39).

Glenister BF, Klapper G. 1966. Upper Devonian conodonts from the

Canning Basin, Western Australia. J. Paleont. 40: 777-842 (IF=1,

cn=124, pp=66). 1680

Grey K. 1992. Miospore assemblages from the Devonian reef

complexes. Canning Basin, Western Australia. Geol. Surv. West.

Aust. Bull. 140: 1-139 (no IF, cn=15, pp=139).

Holland, T. 2014. The endocranial anatomy of Gogonasus andrewsae

Long 195 revaled through micro-CT scanning. Earth and

Environmental Science Transactions of the Royal Society of

Edinburgh 105, 9-34 (IF=1, cn=0, pp=25)

Holland T, Long .A. 2009. On the phylogenetic position of Gogonasus

andrewsae Long 1985, within the Tetrapodamorpha. Acta Zool.

90: 285-296 (IF=1, cn=7, pp=12). 1702

J.A. Long

Johanson Z. 2003. Placoderm branchial and hypobranchial muscles

and origins in jawed vertebrates. J. Vert. Paleont. 23: 735-749

(IF=2, cn=20, pp=15).

Johanson, Z, Trinasjtic, K. 2014. Fosisl ontogenies: the contribution of

placoderm ontogeny to our understanding of the evlution of eraly

gnathostomes. Palaeontology 57: 505-516. (IF=2, cn=0, pp=12).

Long JA. 1985b. A new osteolepidid fish from the Upper Devonian

Gogo Formation, Western Australia. Rec. West. Aust. Mus. 12:

361-367 (IF= 1, cn=18, pp=7).

Long JA. 1987b. Late Devonian fishes from the Gogo Formation,

Western Australia, new discoveries. Search 18: 203-205 (no IF,

cn=4, pp=3) 1744

Long JA. 1988a. New information on the Late Devonian arthrodire

Tubonasus from Gogo, Western Australia. Mems. Ass. Australas.

Palaeontols. 7:81-85 (IF=1, cn=5, pp=5). 1749

Long JA. 1988b. A new camuropiscid arthrodire (Pisces: Placodermi)

from Gogo, Western Australia. Tool. J. Linn. Soc. 94: 233-258

(IF=3; cn=16, pp=26).

Long JA. 1988c. Late Devonian fishes from the Gogo Formation,

Western Australia. Nat. Geog. Res. 4: 436-450 (no IF; cn=24,

pp=15). 1789

Long JA. 1988d. The extraordinary fishes of Gogo. New Scientist 1639:

41-45 (IF=1; cn=4, pp=5).

Long JA. 1988e. 360 million-year-old Gogo fishes. Geo, Austral.

Geograph. Mag. 11 (3):102-113 (no IF, cn=0, pp=12).

Long JA. 1990b. Two new arthrodires (placoderm fishes) from the

Upper Devonian Gogo Formation, Western Australia. Mems. Qld.

Mus. 28: 51-63 (IF=1, cn=9, pp=13).

Long JA. 1991. Arthrodire predation by Onychodus (Pisces,

Crossopterygii) from the Late Devonian Gogo Formation, Western

Australia. Rec. West. Aust. Mus. 15: 479-482 (IF=1, cn=6, pp=4).

Long JA. 1992a. Gogodipterus paddyensis gen. nov., a new

chirodipterid lungfish from the Late Devonian Gogo Formation,

Western Australia. The Beagle, N.T. Mus. 9: 11-20 (no IF, cn=12,

pp=10). 1820

Long JA. 1994. A second incisoscutid arthrodire from Gogo, Western

Australia. Alcheringa 18: 59-69 (IF=1, cn=7, pp=ll).

Long JA. 1995. A new plourdosteid arthrodire from the Upper

Devonian Gogo Formation of Western Australia. Palaeontology

38: 39-62 (IF=2, cn=20, pp=33). 1847

Long JA. 1995. The Rise of Fishes-500 million years of evolution.

Baltimore: Johns Hopkins University Press. 188pp (no IF, cn =121,

pp=188).

Long JA. 1997. Ptyctodontid fishes from the Late Devonian Gogo

Formation, Western Australia, with a revision of the European

genus Ctenurella 0rvig 1960. Geodiversitas 19: 515-555 (IF =1,

cn=36, pp=41).

Long JA. 2001. On the relationships of Onychodus and Psarolepis. J.

Vert. Paleontol. 21: 815-820 (IF=2, cn=23, pp=6).

Long JA. 2006. Swimming in Stone -The amazing Gogo fossils of the

Kimberley. Perth: Fremantle Press. 320pp (no IF, cn=16, p=320).

Long JA. 2010.Holodontid lungfishes from the Late Devonian Gogo

Formation of Western Australia. In Fossil fishes and related Biota:

Morphology, Phylogeny and Paleobiogeography -In honour of

Chang Meeman, ed.Yu X, Maisey J, Miao D. Berlin:Verlag Pfeil,

pp. 277-299. (no IF, cn=3, pp=23)

Long JA, BarwickRE, Campbell KSW. 1997. Osteology and functional

morphology of the osteolepiform fish, Gogonasus andrewsae

Long, 1985, from the Upper Devonian Gogo Formation, Western

Australia. Rec. West. Aust. Mus. Suppl. 53: 1-90 (IF=1, cn=64,

pp=90).

Long JA, Trinajstic KM. 2000. Devonian microvertebrate faunas from

Western Australia. Courier-Forsch. Senckenberg 223: 471-486

(IF=1, cn=14, pp=15).

Long, J.A. & Trinajstic, K. 2010. The Late Devonian Gogo Formation

Lagerstatte -Exceptional preservation and Diversity in early

Vertebrates. Annual Reviews of Earth and Planetary Sciences 38:

665-680 (IF=9, cn =31, pp=26). 2356

Long JA, Trinajstic KM, Johanson Z. 2009. Devonian arthrodire

embryos and the origin of internal fertilization in vertebrates.

Nature 457:1124-1127 (IF=42, cn=40, pp=4 +SI).

Long JA, Trinajstic KM, Young GC, Senden T. 2008. Live birth in the

Devonian period. Nature 453: 650-652 (IF=42, cn=52, pp=3 +SI).

Long JA, Young GC, Holland T, Senden T, Fitzgerald EMG. 2006. An

exceptional Devonian fish shed slight on tetrapod evolution. Nature

444: 199-202 (IF=42, cn=66, pp=4 + SI).

Marshall CR. 1986. Lungfish: Phylogeny and Parsimony. J. Morph.

Suppl. 1:151-162 (IF=2, cn= 39, pp=12). 2553

McNamara KJ, Long JA, Brimmell K. 1991. Catalogue of type fossils in

the Western Australian Museum. Rees. West. Aust. Mus. Suppl. 39:

1-106 (IF=1, cn=0, pp=106).

Melendez, I., Grice, K., Trinajstict, K., Ladjavardi, M., Greenwood, P. &

Thompson, K. 2013. Biomarkers reveal the role of photic zone

euxinia in exceptional fossil ptreservation: an organic geochemical

perspective. Geology 41: 123-126 (IF=5, cn=12, pp=4).

Miles RS. 1971. The Holonematidae (placoderm fishes): a review based

on new specimens of Holonema from the Upper Devonian of

Western Australia. Phil. Trans. R. Soc. Lond. 263B: 101-234 (IF= 3,

cn=80, pp=133).

Miles RS. 1977. Dipnoan (lungfish) skulls from the Upper Devonian of

Western Australia. Zool. J.. Linn. Soc. 61: 1-328 (IF=3, cn=188,

pp=328).

Miles RS, Dennis K. 1979. A primitive eubrachythoracid arthrodire

from Gogo, Western Australia. Zool. J. Linn. Soc 66: 31-62 (IF=3

cn=45, pp=32).

Miles RS, Young GC. 1977. Placoderm interrelationships reconsidered

in the light of new ptyctodontids from Gogo Western Australia.

Linn. Soc. Symp. Ser. 4: 123-198 (IF=3, cn=55, pp=7748

Nazarov BB, Cockbain AE, Playford PE. 1982. Late Devonian

Radiolaria from the Gogo Formation, canning basin. Western

Australia. Alcheringa 6: 161-174 (IF=1, cn=40, pp=14). 2973

Nazarov BB, Ormiston AR. 1984. Upper Devonian (Frasnian)

radiolarian fauna from the Gogo Formation, Western Australia.

Micropal. 29: 454-466 (IF=1, cn=48, pp=13).

NicollRS. 1977. Conodont apparatuses in Upper Devonian palaeoniscoid

fish from the Canning basin. Western Australia. Bur. Min. Res. J.

Aust. Geol. Geophys. 2: 217-228 (no IF, cn=42, pp=12).

Pridmore PA, Barwick RE. 1993. Post-cranial morphologies of the Late

Devonian dipnoans Griphognathus and Chirodipterus and

locomotor implications. Mems. Australas. Ass. Palaeontols. 15:

161-182 (IF=1, cn=15, pp=21). 3078

Pridmore P A, Campbell KSW, Barwick RE. 1994. Morphology and

phylogenetic position of the holodipteran dipnoans of the Upper

Devonian Gogo Formation of northeastern Australia. Phil. Trans.

R. Soc. Lond. B. 344: 105-164 (IF=3, cn=18, pp=60).

Rolfe, WDI. 1966. Phyllocarid crustaceans of European aspect from the

Devonian of Western Australia. Nature 209: 192 (IF=42, cn=12,

pp=l).

Rolfe, WDI 1995. Form and function in Thylacocpehala,

Conchyliocarida and Concavicarida (?Crustacea): aproblem of

interpretation. Trans. Royal Society of Edinburgh 76: 391-399. (IF

=1, cn=19, pp=9).

Rosen DE, Forey PL, Gardiner BG, Patterson C. 1981. Lungfishes,

tetrapods, palaeontology and plesiomorphy. Bull. Amer. Mus.. Nat.

Hist. 167: 159-276 (IF =1, cn=392, pp=117). 3519

Rucklin, M., Donoghue, P.C. J., Trinajstic, K., Marone, F. & Stampanoni.

M. 2012. Development of teeth and jaws in the the earliest jawed

vertebrates. Nature 491: 888-892. (IF=42, cn=21, pp=5).

Quantifying scientific significance of a fossil site: the Gogo Fossil sites (Late Devonian, Western Australia) as a case study

Sanchez, S., Ahlberg, P.E., Trinasjtic, K.M., Mirone, A., Tafforeau, P.

2012. Three-dimesnional synchrotron virtual paleohistology: a

new insight into the world of fossil bone microstructures.

Microscopy and Microanalysis 18: 1095-1105. (IF=2, cn=33,

pp=ll). 3573

Smith MM. 1977. The microstructure of the dentition and dermal

ornament of three dipnoans from the Devonian of Western

Australia: a contribution towards dipnoan interrelations, and

morphogenesis, growth and adaptation of the skeletal tissues.

Phil. Trans. R. Soc. Lond. B 281: 29-72 (IF =3, cn=48, pp=43).

Smith MM. 1979. SEM of the enamel layer in oral teeth of fossil and

extant crossopterygian and dipnoan fishes. Scan. Elec. Micros. 2:

483-90 (IF=3, cn=48, pp=8).

Smith MM. 1984. Petrodentine in extant and fossil dipnoan dentitions:

microstructure, histogenesis and growth. Proc. Linn. Soc. N.S.W.

107: 367-407 (IF=1, cn=33, pp=40).

Smith MM, Campbell KSW. 1987. Comparative morphology,

histology and growth of dental plates of the Devonian dipnoan

Chirodipterus. Phil. Trans. R. Soc. Lond. B 317: 329-363 (IF=3;

cn=27, pp=34).

Tetlie OE, Braddy SJ, Butler PD, Briggs DE. 2004. A new eurypterid

(Chelicerata: Eurypterida) from the Upper Devonian Gogo

Formation of Western Australia, with a review of the

Rhenopteridae. Palaeontology 47: 801-809 (IF=2, cn=20, pp=9).

3701

Trinajstic K. 1995. The role of heterochrony in the evolution of the

eubrachythoracid arthrodires with special reference to

Compagospiscis croucheri and Incisoscutum ritchiei from the

Late Devonian Gogo Formation, Western Australia. Geobios,

Mem. Spec. 19: 125-128 (IF=1, cn=4, pp=4).

Trinajstic K. 1999. New anatomical information on Holonema

(Placodermi) based on material from the Frasnian Gogo

Formation and the Givetian-Frasnian Gneudna Formation,

Western Australia. Geodiversitas 21: 69-84 (IF=1; cn=4, pp=16).

Trinajstic K. 1999b. Scales of palaeoniscoid fishes (Osteichthyes:

Actinopterygii) from the Late Devonian of Western Australia.

Rec. West. Aust. Mus. Suppl. 57: 93-106 (IF=1, cn=7, pp=14).

Trinajstic K. 1999c. Scale morphology of the Late Devonian

palaeoniscoid Moythomasia durgaringa Gardiner and Bartram

1977. Alcheringa 23: 9-19 (IF=1, cn=13, pp=ll).

Trinajstic, K., Boisvert, C., Long, J., Masimenko, A. Johanson, Z.

2014. Pelvic and reproductive structures in placoderms (stem

gnathostomes). Biological reviews DOI: 10.1111/brv.l2118. (If=10,

cn=2, pp=35)

Trinajstic K, Dennis-Bryan K. 2009. Phenotypic plasticity,

polymorphism and phylogeny within placoderms. Acta Zoologica.

90: 83-102 (IF=1; cn=10, pp=20).

Trinajstic KM, George AD. 2009. Microvertebrate biostratigraphy of

Upper Devonian (Frasnian) carbonate rocks in the Canning and

Carnarvon Basins of Western Australia. Palaeontology 52: 642-

659 (IF=2; cn=7, pp=17).

Trinajstic K, Hazelton M. 2007. Ontogeny, phenotypic variation and

phylogenetic implications of arthrodires from the Gogo

Formation, Western Australia. J. Vert. Paleontol.21 : 571-583

(IF=2; cn=ll, pp=13). 3759

Trinajstic K, Long JA. 2009. A new genus and species of Ptyctodont

(Placodermi) from the Late Devonian Gneudna Formation,

Western Australia, and an analysis of ptyctodont phylogeny Geol.

Mag. 146: 743-760 (IF=2; cn=4, pp=18).

Trinajstic K, Marshall C, Long JA, Bifield,, K. 2007. Exceptional

preservation of nerve and muscle tissues in Late Devonian

placoderm fish and their evolutionary implications. Biol. Lett. 3:

197-200 (IF=3; cn=34, pp=14)

Trinajstic K, McNamara KJ. 1999. Heterochrony in the Late Devonian

arthrodiran fishes Compagopiscis and Incisoscutum. Rec. West.

Aust. Mus. Suppl. 57: 77-92 (IF=1, cn=8, pp=15). 3805

Trinajstic, K., Sanchez, S., Dupret, V., Tafforeau, P.Long, J., Young,

G., Senden, T., Boisvert, C., Power, N. & Ahlberg, PE. 2013.

Fossil musculature of the most primitive jawed vertebrates.

Science 341: 160-164 (IF=31, cn=5, pp=5).

Young GC. 1984c. Reconstruction of the jaws and braincase in the

Devonian placoderm fish Bothriolepis. Palaeontology IT. 625-

661 (IF=2; cn=43, pp=38).

Young GC. 1986b. The relationships of placoderm fishes. Zool. J.

Linn. Soc. 88: 1-57 (IF=3, cn=.94, pp=57). 3942

Young GC, Barwick RE, Campbell KSW. 1990. Pelvic girdles of

lungfishes (Dipnoi). In Pathways in Geology- Essays in honour of

E.S. Hills, ed. RW Le Maitre, pp.59-75. Melbourne: Blackwell

Press. (Book Chapter, no IF, cn=8, pp=17

PART B: POPULAR PUBLICATIONS CITING GOGO

FISHES

Long, J.A. 2004. Gogo Fish! The story of the Western Australian State

fossil emblem. The Western Australian Museum, Perth. 48pp.

Long, J.A. 2011. Hung Like an Argentine Duck-the prehistoric origins

of intimate sex. Sydney: Fourth Estate, Harper Collins. 278pp.

Long, J.A. 2011. The Rise of Fishes -500 Million years of Evolution*

2nd ed. Johns Hopkins University Press, Baltimore, and University

of New South Wales Press, UNSW, +288pp.

Long, J.A. 2012. The Dawn of the Deed -The Prehistoric Origins of

Sex. Chicago: University of Chicago Press. 288pp.

Long, J.A. 2014.The Fossil Files -The placoderm renaissance

Australasian Science, 35(7): p.42

Long, J.A., Trinajstic, K. 2014. The first vertebrate sexual organs

evolved as an extra piar of legs. The Conversation,

https://theconversation.com/the-first-vertebrate-sexual-organs-

evolved-as-an-extra-pair-of-legs-27578

Trinajstic, K. 2013. From bone to brawn: ancient fish show off their

muscles. The Conversation, https://theconversation.com/from-bone-

to-brawn-ancient-fish-show-off-their-muscles-15098.

Memoirs of Museum Victoria 74:17-28 (2016) Published 2016

ISSN 1447-2546 (Print) 1447-2554 (On-line)

http://museumvictoria.com.au/about/books-and-journals/journals/memoirs-of-museum-victoria/

Cretaceous marine amniotes of Australia: perspectives on a decade of new research

Benjamin P. Kear

Museum of Evolution, Uppsala University, Norbyvagen 18, SE-752 36 Uppsala, Sweden (banjamin.kear@em.uu.se)

Abstract Kear, B.P. 2016. Cretaceous marine amniotes of Australia: perspectives on a decade of new research. Memoirs of

Museum Victoria 74: 17-28.

Cretaceous marine amniote fossils have been documented from Australia for more than 150 years, however, their

global significance has only come to the fore in the last decade. This recognition is a product of accelerated research

coupled with spectacular new discoveries from the Aptian-Albian epeiric sequences of the Eromanga Basin - especially

the opal-bearing deposits of South Australia and vast lagerstdtten exposures of central-northern Queensland. Novel

fragmentary records have also surfaced in Cenomanian and Maastrichtian strata from Western Australia. The most

notable advances include a proliferation of plesiosaurian taxa, as well as detailed characterization of the ‘last surviving’

ichthyosaurian Platypterygius, and some of the stratigraphically oldest protostegid sea turtles based on exceptionally

preserved remains. Compositionally, the Australian assemblages provide a unique window into the otherwise poorly

known Early Cretaceous marine amniote faunas of Gondwana. Their association with freezing high latitude

palaeoenvironments is also extremely unusual, and evinces a climate change coincident diversity turnover incorporating

the nascent radiation of lineages that went on to dominate later Mesozoic seas.

Keywords Plesiosauria, Platypterygius, Protostegidae, Mosasauroidea, Aptian-Albian, Cenomanian, Maastrichtian.

Introduction

Although Australia has anecdotal Triassic (Kear, 2004; Kear

and Hamilton-Bruce, 2011) and rare Jurassic (Kear, 2012)

marine amniote fossil occurrences, virtually all of its currently

documented Mesozoic record is Cretaceous in age. The

earliest historical publications date from the late 19 th century

(McCoy, 1867a, 1867b, 1869; Owen, 1882; Etheridge 1888,

1897), with only sporadic reports appearing between 1900-

1940 (Etheridge, 1904; Longman, 1915, 1922, 1924, 1930,

1932, 1935, 1943; White, 1935; Teichert and Matheson, 1944),

and in the last decades of the 20 th Century (Romer and Lewis,

1959; Lundelius and Warne, 1960; Persson, 1960, 1982;

McGowan, 1972; Gaffney, 1981; Murray, 1985, 1987; Wade,

1984, 1985, 1990; Molnar, 1991; Thulborn and Turner, 1993;

Cruickshank and Long, 1997; Long and Cruickshank, 1998;

Cruickshank et al., 1999; Choo, 1999). Kear (2003) provided

the first comprehensive overview confirming the presence of

elasmosaurid, possible cryptoclidid, polycotylid,

rhomaleosaurid, and pliosaurid plesiosaurians, the ubiquitous

ophthalmosaurian ichthyosaurian Platypterygius, protostegid

sea turtles, and enigmatic mosasaurids (fig. 1). Since then, a

decade of intensive study has anatomically clarified and

phylogenetically redefined many of these taxa, and added a

plethora of new discoveries that emphasize the global

palaeoecological and palaeobiological significance of the

Australian Cretaceous assemblages. This paper provides both

an updated summary and bibliography of these finds, with the

purpose of stimulating further investigation into this dynamic

field of antipodean vertebrate palaeontology over the decade

to come.

Institutional abbreviations

AM, Australian Museum, Sydney, Australia; AOD, Australian

Age of Dinosaurs Museum, Winton, Australia; NMV, Museum

Victoria, Melbourne, Australia; NTM, Northern Territory

Museum and Art Gallery, Darwin, Australia; QM, Queensland

Museum, Brisbane, Australia; SAM, South Australian

Museum, Adelaide, Australia; UWA, University of Western

Australia, Perth, Australia; WAM, Western Australian

Museum, Perth, Australia.

Plesiosaurians

Australian Cretaceous plesiosaurian fossils are abundant, and

in several rock units, exceptionally well preserved. Those

from the Aptian opal-bearing strata of the Bulldog Shale at

Coober Pedy and Andamooka in South Australia, and

Wallumbilla Formation at White Cliffs, as well as the Albian

Griman Creek Formation in New South Wales (fig. 2), are

perhaps the most unusual because they manifest diagenetic

replacement of the bony tissue by opaline hydrated silica. Kear

(2005a, 2006a, 2006b) revised the existing specimens and

BP. Kear

Figure 1. Stratigraphical distribution of Australian Cretaceous marine amniote taxa updated from Kear (2003). Australian standard microplankton

(dinoflagellate) zonation is modified from Partridge (2006) to accommodate the emended geological timescale of Gradstein et al. (2012). Taxon

ranges indicate named species (black bars) or indeterminate occurrences assigned to higher-level taxa (open bars).

recognized a diversity of taxa. These included indeterminate

elasmosaurids, some of which were osteologically immature

and of surprisingly small body-size - the most diminutive

being less than 2 m in estimated maximum length (AM

F9639-F9928: see Kear, 2002a). Their vertebral morphology

showed characteristics of other Gondwanan elasmosaurids.

For example, the lack of a pronounced ventral notch on the

articular faces of the cervical centra (Kear, 2002a, p. 672, fig.

1) is typical of the Early Cretaceous Callawayasaurus

colombiensis (Welles, 1962) from Colombia, as well as

Eromangasaurus australis (Sachs, 2005) and the nomen

dubium Woolungasaurus glendowerensis Persson, 1960

(Sachs, 2004) from Australia. Furthermore, the centrum

proportions of these ‘juvenile’ elasmosaurids (Kear, 2002a, p.

673, table 1) are reminiscent of the austral high-latitude

Aristonectes parvidens Cabrera, 1941 and Kaiwhekia katiki

Cruickshank and Fordyce, 2002 (see O’Gorman et al., 2014)

from Patagonia-Antarctica and New Zealand respectively.

Note, though, that elasmosaurid cervical centrum proportions

are intraspecifically variable and of uncertain taxonomic

significance (O’Keefe and Hiller, 2006; Sachs et al., 2013).

Kear (2005b, 2007a) and Sachs (2005) reported on cranial

material of E. australis from the middle-upper Albian

Toolebuc Formation near Maxwelton in Queensland (fig. 2).

Cretaceous marine amniotes of Australia: perspectives on a decade of new research

0 500 1000 1500

Figure 2. Diagrammatic map of Cretaceous rock outcrops on the Australian continent with state borders and specific locality references for fossil

occurrences discussed in the text (developed from Kear and Hamilton-Bruce, 2011).

The Toolebuc Formation and overlying Allaru Mudstone are

lagerstatten sequences that have yielded some of the most

spectacular Early Cretaceous marine amniote fossils found

worldwide. The holotype skull of E. australis (QM FI 1050:

Kear, 2005b, p. 794, fig. 2A-C) is a classic example. It bears a

series of depression fractures and crushing attributed to a bite

from a gigantic predatory pliosauroid (Thulborn and Turner,

1993). Moreover, its phylogenetic character states include

circular (non-compressed) tooth cross-sections and the

possible presence of a pineal foramen, which suggest a basal

position within Elasmosauridae (Benson and Druckenmiller,

2014), although, its topology within this clade is ambiguous

(Sachs and Kear, 2015). Several other fragmentary

elasmosaurid skulls have also been found in the Toolebuc

Formation (e.g. AM F87826: Kear, 2001a), one of which has a

well-preserved premaxillary palate exposing the intracranial

sinuses (SAM P40510: fig. 3A). A number of articulated

elasmosaurid postcranial skeletons await adequate preparation

and study. In addition, indeterminate isolated vertebrae occur

in the Aptian Birdrong Sandstone (WAM 94.7.6), lower Albian

WindaliaRadiolarite (WAM 05.2.1), as well as the Cenomanian

upper Gearle Siltstone (WAM 15.2.1) in Western Australia.

McHenry et al. (2005) reported on a remarkable elasmosaurid

specimen (QM F33037) from the Wallumbilla Formation

(referred to as the Blackdown or Doncaster Formation in the

Walsh Creek region of Queensland) that preserved an

BP. Kear

Figure 3. Marine amniote fossils from Cretaceous strata in Australia. A, elasmosaurid premaxillary palate (SAM P40510) exposing the vomerine

contact and intracranial sinus. B, spectacular mounted skeleton (QM F18041) of the new polycotylid popularly dubbed the ‘Richmond pliosaur’.

C, partially disarticulated ‘juvenile’ postcranium referred to Umoonasaurus demoscyllus. Both scapulae (outlined) and an in situ gastrolith mass

are indicated. D, ‘ Umoonasaurus- like’ propodial from the late Aptian Darwin Formation, Northern Territory. E, CT rendering of an exceptionally

preserved ‘juvenile’ Platypterygius australis cranium and mandible (AM F98273). Image compilation: Ben Hill (Adelaide). F, articulated

humerus and distal forelimb elements (AM F107444) of a ‘juvenile’ Platypterygius australis. G, ophthalmosaurian phalanx (WAM 99.1.4) from

the late Cenomanian Geale Siltstone, Western Australia. Image: Mikael Siversson (Western Australian Museum). H, mosasaurid ulna (UWA

37092) with antebrachial foramen and intermedium contact indicated. I, cranium of Bouliachelys suteri (SAM P41106) in lateral view. J,

articulated cranium and carapace of Bouliachelys suteri (SAM P40525) in dorsal view. Scale bars represent 20 mm in A, G, H; 500 mm in B;

100 mm in C, J; and 50 mm in D-F, I. Abbreviations: abf - antebrachial foramen; dfi - distal facet for the intermedium; ics - intracranial sinus;

gst - gastrolith mass; hpx - hooked premaxillae; lea - lateral exposure of angular; pmj - premaxillary, maxillary, and jugal contacts; rbe -

reduced basioccipital extracondylar area; rze - position of radial zeugopodial element; scp - scapulae.

Cretaceous marine amniotes of Australia: perspectives on a decade of new research

associated bromalite comprising belemnites and a high

proportion of benthic bivalves and gastropods. This implied a

propensity for bottom feeding, a habit advocated elsewhere

from bite marks on bivalve shells (Kear and Godthelp, 2008).

Perhaps the most enigmatic Australian plesiosaurian named

in recent years is Opallionectes andamookaensis Kear, 2006a

from the Aptian Bulldog Shale of Andamooka. This taxon was

identified from a single skeleton (SAM P24560), but an isolated

tooth (privately owned) has since been found in coeval strata at

Coober Pedy. Kear (2006a) noted features of the vertebrae,

including craniocaudally short/broad cervical centra, the lack

of a longitudinal ridge, and platycoelous to shallowly

amphocoelous articular faces that might be homologous with

those of the Late Cretaceous aristonectine elasmosaurids

Aristonectes spp. and K. katiki (O’Gorman et al., 2013;

O’Gorman et al., 2014). The absence of an intercoracoid

embayment on the pectoral girdle (see Kear, 2006a, p. 843, fig.

3B), however, clearly differentiates O. andamookaensis from

remains referred to Aristonectes (O’Gorman et al., 2013; Otero

et al., 2014). Furthermore, a combination of its prominent

caudolateral coracoid cornuae, distally expanded propodials,

pre- and post-xial accessory ossicles in the epipodial row, and

small labiolingually compressed teeth that lack enamel ridges

is alternatively consistent with Late Jurassic - Early Cretaceous

cryptoclidids - e.g. Kimmerosaurus langhami Brown, 1981,

Tatenectes laramiensis (Knight, 1900) (O’Keefe and Street,

2009), and Abyssosaurus nataliae Berezin, 2011.

Australian polycotylids are contentious and might

incorporate the oldest stratigraphical exemplar of the group: a

fragmentary opalized skeleton (AM F6266-F6298) from the

Aptian Wallumbilla Formation at White Cliffs. Persson (1960)

first mooted the polycotylid affinity of this specimen, a

conclusion supported by Kear (2005a) based on its

characteristically slender, homodont teeth and cervical centra

that were shorter than high with sharp mid-ventral keels and

constricted lateral sides (see O’Keefe, 2004; Druckenmiller

and Russell, 2008a; Schmeisser, McKean, 2012). Other features

such as laterally flared dorsal apices on the neural spines (see

Kear, 2005a, p. 775, fig. 4C) compare well with Dolichorynchops

herschelensis Sato, 2005. Craniad swelling of the median

ventral edge of the articular facet rim is diagnostic for

Polycotylidae (Sato and Storrs, 2000) but also resembles the

prominent ‘lip’ described in the leptocleidian Hastanectes

valdensis Benson, Ketchum, Naish and Turner, 2013a.

Polycotylid-like elements have been recovered in the

Aptian Bulldog Shale (SAM P36356: Kear, 2006a), middle-

late Albian Toolebuc Formation (SAM P41967), latest Albian-

Cenomanian Mackunda Formation in Queensland (e.g. AOD

F336), and late Cenomanian upper Gearle Siltstone (WAM

14.10.3.1-11). However, the only unequivocally diagnostic

skeleton (QM F18041: fig. 3B) is the ‘Richmond pliosaur’ from

the Allaru Mudstone of Richmond, Queensland (fig. 2). This

sensational specimen represents one of the most complete

Gondwanan plesiosaurian fossils yet discovered, and

comprises a skull with characteristically elongate maxillary

rostrum and symphyseal region of the mandible, incorporating

a caudal extension of the splenial to the eighth tooth position.

The splenial alternatively projects beyond the 10 th tooth

position in most Late Cretaceous polycotylids (Carpenter,

1996; Arkhangelsky et al., 2007). The palate of QM F18041

displays distinctively convex (= “dished” sensu O’Keefe, 2001)

lateral palatal pterygoid surfaces bordering the posterior

interpterygoid vacuities. Loss of the pineal foramen serves to

differentiate QM F18041 from the only other named Early

Cretaceous polycotylid Edgarosaurus muddi Druckenmiller,

2002. Conspicuous ornamentation of ridges and grooves along

the snout and mandible is further reminiscent of the Patagonian

Campanian-Maastrichtian Sulcusuchus erraini Gasparini and

Spalletti, 1990, and might have housed a dermal sensory

system (O’Gorman and Gasparini, 2013; Foffa et al., 2014).

A second osteologically immature polycotylid skeleton

(QM F12719) from the Toolebuc Formation near Hughenden in

Queensland (fig. 2) was considered a new leptocleidid by Glen

and McHenry (2007) but is morphologically indistinguishable

from QM F18041 and thus probably conspecific.

Cruickshank and Long (1997) named the first Australian

leptocleidid plesiosaurian Leptocleidus clemai Cruickshank

and Long, 1997 based on several partial skeletons (WAM 92.8.1,

WAM 94.1.6) from the Aptian Birdrong Sandstone of Kalbarri

in Western Australia (fig. 2). These were phylogenetically

re-evaluated by Kear and Barrett (2011), who failed to resolve L.

clemai with other Leptocleidus spp., and noted that the only

discrete character state diagnosing the species - epipodials

broader than long - was ubiquitous amongst polycotylids and

other Cretaceous plesiosaurians. Cruickshank and Long (1997)

listed a 30% size increase relative to the type species

Leptocleidus superstes Andrews, 1922 as another specifically

differential feature, but failed to offer an explicit case for their

generic referral to Leptocleidus (Kear and Barrett, 2011). Little

else remains to distinguish Leptocleidus clemai except perhaps

its robust propodials, which have a noticeably sigmoidal profile

like polycotylids (e.g. Albright et al., 2007; O’Keefe, 2008) and

L. superstes (Kear and Barrett, 2011); although, the articular

surface on the humeral head exhibits pronounced lateral flaring

unlike the more cylindrical capitulum of L. superstes (compare

Kear, 2003, p. 294, fig. 6A, B with Kear and Barrett, 2011, p.

673, fig. 4F-I).

Kear (2006b) documented isolated small pliosauroid bones

and teeth that were similar to those of leptocleidids but occurred

in non-marine strata of the Aptian-Albian Eumeralla Formation

from Cape Otway to Inverloch in Victoria, and in the early-

middle Albian Griman Creek Formation of Lightning Ridge,

New South Wales and Surat in Queensland (fig. 2). New

plesiosaurian elements from the Eumeralla Formation were

figured by Benson et al. (2013b, p. 3, fig. 2), including NMV

P198945, a large broken tooth (45 mm high) missing most of its

enamel surface (Benson et al., 2013b, p. 2, fig. 1). Benson et al.

(2013b) argued for plesiosaurian affinity based on remnants of

three incomplete and irregularly spaced enamel ridges (the

remaining intact surface was otherwise smooth), conical tooth

shape, and the apparent absence of carinae. While these traits are

certainly compatible with plesiosaurians, they are likewise

similar to spinosaurid theropods (also recovered from the

Eumeralla Formation: Barrett et al., 2011), which can express

smooth or fluted enamel, conical tooth form, and reduced carinae

(e.g. Dal Sasso et al., 2005; Medeiros, 2006; Richter et al., 2013).

BP. Kear

The best-known leptocleidan taxon from Australia is

Umoonasaurus demoscyllus Kear, Schroeder and Lee, 2006a

from the Aptian Bulldog Shale of Coober Pedy. A number of

skeletons have been discovered, including multiple small-bodied

‘juveniles’, one of which (SAM P15980; originally identified as

cf. Leptocleidus sp. by Kear, 2007b) was proportionately scaled

to the 2.5 m long holotype (AM F99374) and thereby estimated

to be only 700 mm in maximum body length (Kear, 2007b).

Another ‘juvenile’ specimen (SAM P33915) was also probably

less than 1 m long and includes an in situ gastrolith accumulation,

together with complete scapulae that display lateral shelves - a

key leptocleidid synapomorphy (fig. 3C). Originally classified

with Jurassic rhomaleosaurids (Kear et al., 2006a) or

polycotylids (Druckenmiller and Russell, 2008a), the

leptocleidid affinity of U. demoscyllus has been iterated by

recent phylogenies (Ketchum and Benson, 2010; Benson and

Druckenmiller, 2014), but relies upon few states including the

presence of a lateral shelf on the scapula (disparately occurring

in Jurassic taxa: Sato et ah, 2003; Sachs et ah, 2014), and a

triangular fossa extending from the pineal foramen to the

sagittal crest on the dorsal surface of the parietal; evident

elsewhere in the disputed leptocleidids Nichollssaura borealis

(Druckenmiller and Russell, 2008b) and Brancasaurus brancai

Wegner, 1914 (Benson et ah, 2013a). The skull of U. demoscyllus

is, however, unique in its development of thin, high crests along

the midline of the snout and above the orbits on the frentals (see

Kear et ah, 2006a, p. 617, fig. lb). The function of these is

unclear but they potentially represent display structures that

might have been sexually dimorphic (Kear et ah, 2006a).

Kear (2002b) reported on leptocleidid remains from the late

Aptian Darwin Formation near Darwin in the Northern

Territory (fig. 2). Further assessment of this material has

revealed propodials (e.g. NTM P998-6; fig. 3D) and associated

vertebrae that are indistinguishable from those of U. demoscyllus

(see Kear, 2006a, supplemental fig. S7f, g) and might evidence

this, or another closely related species inhabiting the Australian

continental margin during the Early Cretaceous.

The gigantic pliosauroid Kronosaurs queenslandicus

Longman, 1924 is the largest and most stratigraphically

widespread plesiosaurian taxon thus far documented from

Australia. Its conspicuous remains have been recovered from

Aptian units throughout the Eromanga Basin, including the

Bulldog Shale at Coober Pedy (Kear, 2006a), and the Wallumbilla

Formation at both White Cliffs (Kear, 2005a) and near

Richmond; this was the source of the famous Harvard University

skeleton (MCZ 1285: Romer and Lewis, 1959). The holotype

(QM F1609), however, derived from the Albian Toolebuc

Formation at Hughenden, with a second flattened skull (QM

F2446) that is thought to represent a separate species (Molnar,

1991). In contrast, McHenry (2009) considered all of the

Australian Kronosaurus remains coherent with a monospecific

morphotype, and thus presented a composite reconstruction of

the cranium and mandible (see McHenry, 2009, p. 349, fig. 5-35)

incorporating parts of a 10 m long skeleton (QM F10113) from

the Toolebuc Formation near Hughenden.

Phylogenetic determinations have placed K. queenslandicus

as a derived member of the Brachaucheninae (Benson and

Druckenmiller, 2014), a Cretaceous pliosauroid radiation

notably characterised by loss of the subcentral foramina on the

cervical vertebrae. Cranial modelling and inferred gastric

residues also suggest that K. queenslandicus might have

favoured smaller prey particularly marine turtles, elasmosaurid

plesiosaurians, and possibly sharks (McHenry, 2009).

Ichthyosaurians

Australian Cretaceous ichthyosaurian fossils have been

intensively studied. Wade (1984) compiled a seminal review,

recognizing a single species Platypterygius australis (McCoy,

1867a). Subsequent uncertainty over the holotype led to

taxonomic conflict (Wade, 1990). However, Zammit (2010)

resolved these issues with a reassessment of the original

specimens described by McCoy (1869). These included a partial

skull (MV P12989: Zammit, 2010, p. 6, fig. 2A), probably

associated with the type vertebrae (Wade, 1985), that supported

referral to the genus Platypterygius (via a reduced extracondylar

area on the basioccipital: McGowan and Motani, 2003), and

conformed with other exemplars of P. australis (which exhibit

exclusion of the lacrimal from the external bony nasal aperture

and the presence of accessory caudodorsal nasal foramina: Kear,

2005c). Fossils of P. australis are otherwise prolific and often

excellently preserved, especially in the middle-late Albian

Toolebuc Formation and Allaru Mudstone of Queensland. This

unprecedented quantity and quality of material has facilitated

comprehensive appraisals of craniodental (Kear, 2005c; Maxwell

et al., 2011) and postcranial anatomy (Zammit et al., 2010) that

are now a comparative standard for Cretaceous ichthyosaurian

remains worldwide (e.g. Maxwell and Kear, 2010; Maxwell et

al., 2012; Fischer et al., 2014). Moreover, functional analyses and

feeding traces have permitted reconstruction of locomotory

modes (Zammit et al., 2014), jaw musculature and sense organs

(Kear, 2005c), and dietary specialisation towards small-bodied

prey including bony fish (Wretman and Kear, 2014) and aquatic

amniotes (e.g. hatchling turtles: Kear et al., 2003). Pathological

elements have further afforded evocative glimpses into

ichthyosaurian disease (dental caries: Kear, 2001b) and

intraspecific behavioural interactions (Zammit and Kear, 2011).

Exceptionally preserved foetal remains also infer a K-selection

reproductive strategy favouring large young (around one meter

long at parturition) that were born ‘tail first’ and probably

precocial (Kear and Zammit, 2014).

The precise phylogenetic relationships of P. australis are

ambiguous, but the taxon is undoubtedly an advanced

ophthalmosaurian because of the restricted basicoccipital

extracondylar area, extensive lateral exposure of the angular,

and extra pre-radial zeugopodial element/digit articulating

with the humerus (McGowan and Motani, 2003: fig. 3E, F).

Kear and Zammit (2014) identified additional ontogenetically

stable autapomorphies that were consistently expressed

through an in-utero to osteologically mature ‘adult’ growth

trajectory: premaxillary processus supranarialis having

minimal contact with the bony nasal aperture; premaxillary

processus subnarialis of subequal length to the processus

supranarialis and extending across the external face of the

maxilla; a well sutured jugal-maxilla contact; and absence of

a squamosal (fig. 3E).

Cretaceous marine amniotes of Australia: perspectives on a decade of new research

Other Australian Platypterygius occurrences are known

from the Aptian Bulldog Shale, Wallumbilla Formation,

Darwin Formation, and Birdrong Sandstone (Kear, 2002b,

2003, 2005a, 2006a), as well as the late Albian-Cenomanian

Alinga Formation and Molecap Greensand of Western

Australia (Choo, 1999; Kear 2003). Novel discoveries have

extended this range into the mid-late Cenomanian upper-most

Gearle Siltstone. This was based on an isolated phalanx

(WAM 99.1.4: fig. 3G) from the Murchison River region west

of Kalbarri, which is important because it could represent the

stratigraphically youngest ichthyosaurian fossil thus far

documented from the southern hemisphere (see Sachs and

Grant-Mackie, 2003; Zammit, 2012).

Aquatic squamates

Kearetal. (2005) summarized the Australian marine squamate

record noting the presence of various indeterminate

mosasaurids. These incorporated an ulna and phalanx (UWA

37092) which Lundelius and Warne (1960, p. 1216) thought

similar to either Platecarpus Cope, 1869 or Clidastes Cope,

1868, but “perhaps closer to Platecarpus”. The ulna (fig. 3H)

certainly has a compact shaft with shallowly concave edges

suggesting an oval antebrachial foramen. However, the distal

extremity appears to be offset for contact with the intermedium,

which is more like Clidastes (Russell, 1967). UWA 37092

derived from the Molecap Greensand near Gingin (fig. 2), a

slumped sequence of Cenomanian-Coniacian strata associated

with a buried Cretaceous impact crater (Mory et al., 2005).

Some mosasaurid dorsal vertebrae have been documented

from this unit (WAM 98.7.1-10), and other mosasaurid

vertebrae are known from the early Maastrichtian Korojon

Calcarenite (UWA 133937), and late Maastrichtian Miria

Formation (WAM 91.8.16) of Western Australia (see Kear et

ah, 2005, p. 309, fig. 2G-0).

Scanlon and Hocknull (2008) mentioned the surprising

occurrence of a ‘dolichosaur-like’ presacral vertebra (QM

F52673) in the non-marine latest Albian-Turonian Winton

Formation at Winton in Queensland (fig. 2). The tapered

centrum shape, “moderately prominent” synapophyses, and

apparent absence of pachyostosis (evident in various aquatic

varanoids: Houssaye et al., 2008) were advocated to support

this assignment. The oval outline and cranial inclination of the

cotyle (see Scanlon and Hocknull, 2008, p. 133, fig. IF) are

otherwise similar to varanids as well as basal mosasauroids

(Carroll and De Braga, 1992; Makadi et ah, 2012), yet the

badly eroded condyle seems to lack a varanid-like precondylar

constriction (Scanlon and Hocknull, 2008). The affinities of

QM F52673 are thus unclear and the specimen is probably best

interpreted as an indeterminate varanoid.

Marine Turtles

Marine turtle fossils are frequently discovered in Australia,

but are at present stratigraphically restricted to only a few

units including the middle-late Albian Toolebuc Formation

and Allaru Mudstone (Kear, 2003), together with the late

Albian-Cenomanian Mackunda Formation (e.g. AOD F795),

and late Maastrichtian Miria Formation (Kear and Siverson,

2010). Molnar (1991, p. 618) reported a possible marine, “or at

least aquatic” turtle (listed as a “tortoise” by Molnar, 1991, p.

612) from an unspecified locality northwest of Winton (fig. 2)

that was mapped within the non-marine Winton Formation.

This specimen (QM FI2413) consists of an internal cast of the

carapace (Molnar, 1991, p. 690, pi. 1), but seems to show

reduced costal plates and fontanellization consistent with

Chelonioidea (Wood et ah, 1996; Lehman and Tomlinson,

2004). All other identifiable Winton Formation turtle remains

pertain to chelids (Hocknull et ah, 2009). However, the lower¬

most Albian section of the Winton Formation does produce

typically marine vertebrates (e.g. ichthyodectiform teleosts:

Berrell et ah, 2014), and was deposited by a tidal fluvial system

that could have accommodated euryhaline organisms.

Most Australian Cretaceous marine turtles are referred to

the cosmopolitan clade Protostegidae, which might represent

either a basal chelonioid lineage (Hirayama, 1998; Hooks, 1998;

Kear and Lee, 2006; Bardet et ah, 2013), or a more archaic

radiation of marine cryptodires (Joyce, 2007). Three endemic

Australian taxa have been named from the middle-late Albian

Toolebuc Formation, and are amongst the oldest protostegids

documented worldwide. Owen (1882) described the partial

carapace and plastron (AM F67326) of Notochelys costata

Owen, 1882, from an unknown location on the Thomson River

in Queensland (see De Vis, 1911). Lydekker (1889) subsequently

replaced the epithet ‘ Notochelys ’ with Notochelone Lydekker,

1889 because of synonymy, and De Vis (1911) referred additional

elements that Kear (2003) used to compile an emended

diagnosis. Kear and Lee (2006) listed the jugal-quadrate

contact, extension of the pterygoid onto the articular condyle of

the quadrate, and incorporation of the vomer into the upper

triturating surface as states distinguishing N. costata from the

larger-bodied (around 50%) coeval taxon Bouliachelys suteri

Kear and Lee, 2006. This is known from several spectacular

skulls (e.g. SAM P41106: fig. 31) found near Boulia in

Queensland (fig. 2). Kear and Lee (2006) phylogenetically

placed both N. costata and B. suteri as basal protostegids.

Parham and Pyenson (2010) further correlated the distinctive

hooked premaxillae and poorly developed secondary palate of

B. suteri with ‘shear’ feeding (as opposed to durophagy), and

proposed protostegid convergence upon extant herbivorous

cheloniids. Interestingly, Kear (2006c) identified both

gastrolites/cololites andcoprolites within multiple ‘ Notochelone-

like’ specimens that contained dense accumulations of

Inoceramus bivalve shell. These were processed orally, implying

benthic ‘grazing’ and an invertebrate-based diet in the earliest

protostegids (Kear, 2006c).

Despite obvious character state differentiation, Myers

(2007) suggested that N. costata and B. suteri might be

synonymous because of proportional similarities in their

crania. This warrants further exploration especially relative to

their postcranial elements, which now include several

articulated skeletons under preparation and study (e.g. SAM

P40525: fig 3J).

The holotype (QM F14550: Kear, 2006d, p. 781, fig. 2A-R)

and only specimen of the colossal (around four meters in length)

Toolebuc Formation protostegid Cratochelone berneyi

Longman, 1915 discovered near Hughenden, was re-evaluated

BP. Kear

by Kear (2006d), and found to possess highly vascular limb bone

surfaces compatible with the advanced protostegids Archelon

Wieland, 1896 and Protostega Cope, 1872. Comparable

microstructures occur in Australia’s only Late Cretaceous

chelonioid fossil, a dermochelyoid scapula (WAM 03.3.37: Kear

and Siverson, 2010, p. 4, fig. 2E) from the late Maastrichtian

Miria Formation southwest of Exmouth in Western Australia

(fig. 2), and suggest that complex metabolic physiology repeatedly

coupled with body size (see Rhodin, 1985) and perhaps pelagic

lifestyles throughout marine turtle evolution.

Conclusions and Future Research

Australian Cretaceous marine amniote fossils represent a

significant resource for exploring the enigmatic vertebrate

biodiversity of Gondwana. The astounding taxonomic

richness, preservation quality, and sheer productivity of the

documented source units, particularly those from the Aptian-

Albian strata of the Eromanga Basin in Queensland and South

Australia (Kear, 2003), mark them as some of the most

important (although as yet under-popularized) Mesozoic

marine vertebrate lagerstdtten known worldwide. Their

association with Early Cretaceous freezing high latitude

palaeoenvironments is also unique, and has been linked with

climate change coincident dispersals and cladogenic events

that shaped later Cretaceous faunas (Kear et al., 2006b).

Clearly this offers an exciting focus for on-going research,

especially given the recent advances in quantitative modeling

(e.g. Benson et al., 2010a), geochemical analyses (e.g. Bernard

et al., 2010), bone microstructure visualization (e.g. Houssaye,

2013), and soft tissue reconstruction (e.g. Lindgren et al., 2014)

that are changing the perspectives on Mesozoic marine

amniote evolution. Nevertheless, such innovative approaches

would not be possible without the pioneering studies

undertaken over the last decade. These have provided not only

a fundamental systematic framework but also posed an

intriguing experimental question that can be used to frame

future investigations - did Australia’s Cretaceous marine

amniotes experience high-latitude thermal isolation and low-

temperature adaptation concomitant with the coeval

continental record of inimitable relics (e.g. Thulborn and

Turner, 2003; Smith and Kear, 2013), regional immigrants

(e.g. Rich and Vickers-Rich, 2003; Smith et al., 2008; Hocknull

et al., 2009; Rich et al., 2009; Agnolin et al., 2010; Benson et

al., 2010b; Kellner et al., 2010; Barrett et al., 2011; Benson et

al., 2012; Fitzgerald et al., 2012; Poropat et al., 2014; Poropat

et al., 2015; Rich et al., 2014), and endemic progenitors (e.g.

Pridmore et al., 2005; Salisbury et al., 2006; Rowe et al., 2008;

Smith, 2010) whose descendants are still extant on the

Australian landmass today?

Acknowledgements

This article is dedicated to Dr Thomas H. Rich, whose

outstanding contributions to Australian vertebrate palaeontology

have inspired generations of young scientists to follow in his

footsteps. Thanks to Erich Fitzgerald for the invitation to

contribute to this festschrift volume. Mary-Anne Binnie (South

Australian Museum), Gavin Dally (Museum & Art Gallery of

the Northern Territory), David Pickering (Museum Victoria), R.

John Reeve (University of Western Australia), Mikael Siversson

(Western Australian Museum), Kristen Spring and Andrew

Rozefelds (Queensland Museum), and Yong Yi Zhen (Australian

Museum) assisted with access to specimens and collection

information. The Australian Research Council, Swedish

Research Council, National Geographic Society, Sir Mark

Mitchell Research Fund, South Australian Museum, Umoona

Opal Mine and Museum, Uppsala University and other private

patrons provided financial support for the research.

References

Agnolin, F.L., Ezcurra, M.D., Pais, D.F., and Salisbury, S.W. 2010. A

reappraisal of the Cretaceous non-avian dinosaur faunas from

Australia and New Zealand: Evidence for their Gondwanan

affinities. Journal of Systematic Palaeontology 8: 257-300.

Albright, L.B. Ill, Gillette, D.D., and Titus, A.L. 2007. Plesiosaurs

from the Upper Cretaceous (Cenomanian-Turonian) tropic shale

of southern Utah, part 2: polycotylidae. Journal of Vertebrate

Paleontology 27: 41-58.

Arkhangelsky, M.S., Averianov, A.O., and Pervushov, E.M. 2007.

Short-necked plesiosaurs of the Family Polycotylidae from the

Campanian of the Saratov Region. Paleontological Journal 41:

656-660.

Bardet, N., Jalil, N.E., de Broin, F.D.L., Germain, D., Lambert, O.,

and Amaghzaz, M. 2013. A giant chelonioid turtle from the Late

Cretaceous of Morocco with a suction feeding apparatus unique

among tetrapods. PloS ONE 8(7), e63586.

Barrett, P.M., Benson, R.B.J., Rich, T.H., and Vickers-Rich, P. 2011.

First spinosaurid dinosaur from Australia and the cosmopolitanism

of Cretaceous dinosaur faunas. Biology Letters 7: 933-936.

Benson, R.B.J., and Druckenmiller, P.S. 2014. Faunal turnover of

marine tetrapods during the Jurassic-Cretaceous transition.

Biological Reviews 89: 1-23.

Benson, R.B.J., Butler, R.J., Lindgren, J., and Smith, A.S. 2010a.

Mesozoic marine tetrapod diversity: mass extinctions and

temporal heterogeneity in geological megabiases affecting

vertebrates. Proceedings of the Royal Society of London, Series B

277: 829-34.

Benson, R.B.J., Barrett, P,M., Rich, T.H., and Vickers-Rich, P. 2010b.

A southern tyrant reptile. Science 327: 1613.

Benson, R.B.J., Rich, T.H., Vickers-Rich, P, and Hall, M. 2012.

Theropod fauna from southern Australia indicates high polar

diversity and climate-driven dinosaur provinciality. PloS ONE

7(5): e37122.

Benson, R.B.J., Ketchum, H.F., Naish, D., and Turner, L.E. 2013a. A

new leptocleidid (Sauropterygia, Plesiosauria) from the Vectis

Formation (Early Barremian-early Aptian; Early Cretaceous) of

the Isle of Wight and the evolution of Leptocleididae, a

controversial clade. Journal of Systematic Palaeontology 11:

233-250.

Benson, R.B.J., Fitzgerald, E.M.G., Rich, T.H., and Vickers-Rich, P.

2013b. Large freshwater plesiosaurian from the Cretaceous

(Aptian) of Australia. Alcheringa 37: 1-6.

Berezin, A.Y. 2011. A new plesiosaur of the Family Aristonectidae

from the Early Cretaceous of the center of the Russian platform.

Paleontological Journal 45: 648-660.

Bernard, A., Lecuyer, C., Vincent, P, Amiot, R., Bardet, N., Buffetaut,

E., Cuny, G., Fourel, F., Martineau, F., Mazin, J.-M., and Prieur,

A. 2010. Regulation of body temperature by some Mesozoic

marine reptiles. Science 328: 1379-1382.

Cretaceous marine amniotes of Australia: perspectives on a decade of new research

Berrell, R.W., Alvarado-Ortega, J., Yabumoto, Y., and Salisbury, S.W.

2014. First record of the ichthyodectiform fish Cladocyclus from

eastern Gondwana: An articulated skeleton from the Early

Cretaceous of Queensland, Australia. Acta Palaeontologica

Polonica. 59: 903-920.

Brown, D.S. 1981. The English Upper Jurassic Plesiosauroidea

(Reptilia) and a review of the phylogeny and classification of the

Plesiosauria. Bulletin of the British Museum (Natural History)

Geological Series 35: 253-347.

Cabrera, A. 1941. Un plesiosaurio nuevo del Cretaceo del Chubut.

Revista del Museo de La Plata 2: 113-130.

Carpenter, K. 1996. A review of the short-necked plesiosaurs from the

Cretaceous of the Western Interior, North America. Nues Jahrbuch

fiir Geologie und Palaontologie, Abhandlungen 201: 259-287.

Choo, B. 1999. Cretaceous ichthyosaurs from Western Australia.

Records of the Western Australian Museum , Supplement 57:

207-218.

Cope, E.D. 1868. On new species of extinct reptiles. Proceedings of

the Academy of Natural Sciences of Philadelphia 20: 181.

Cope, E.D. 1869. On the reptilian orders Pythonomorpha and

Streptosauria. Boston Society of Natural History Proceedings 12:

250-266

Cope, E.D. 1872. A description of the genus Protostega, a form of

extinct Testudinata. Proceedings of the American Philosophical

Society 12: 422-433.

Cruickshank, A.R.I., and Long, J.A. 1997. A new species of pliosaurid

reptile from the Early Cretaceous Birdrong Sandstone of Western

Australia. Records of the Western Australian Museum 18: 263-276.

Cruickshank, A.R.I., and Fordyce, R.E. 2002. A new marine reptile

(Sauropterygia) from New Zealand: further evidence for a Late

Cretaceous austral radiation of cryptoclidid plesiosaurs.

Palaeontology 45: 557-575.

Cruickshank, A.R.I., Fordyce, R.E., and Long, J.A. 1999. Recent

developments in Australasian sauropterygian palaeontology

(Reptilia: Sauropterygia). Records of the Western Australian

Museum, Supplement 57: 201-205.

Carroll, R.L., and Debraga, M. 1992. Aigialosaurs: mid-Cretaceous

varanoid lizards. Journal of Vertebrate Paleontology 12: 66-86.

Dal Sasso, C., Maganuco, S., Buffetaut, E., and Mendez, M.A. 2005.

New information on the skull of the enigmatic theropod

Spinosaurus, with remarks on its size and affinities. Journal of

Vertebrate Paleontology 25: 888-896.

De Vis, C.W. 1911. On some Mesozoic fossils. Memoirs of the

Queensland Museum 10: 1-11.

Druckenmiller, P.S. 2002. Osteology of a new plesiosaur from the

Lower Cretaceous (Albian) Thermopolis Shale of Montana.

Journal of Vertebrate Paleontology 22: 29-42.

Druckenmiller, P.S., and Russell, A.P. 2008a. A phylogeny of

Plesiosauria (Sauropterygia) and its bearing on the systematic

status of Leptocleidus Andrews, 1922. Zootaxa, 1863: 1-120.

Druckenmiller, P.S., and Russell, A.P. 2008b. Skeletal anatomy of an

exceptionally complete specimen of a new genus of plesiosaur

from the Early Cretaceous (Early Albian) of northeastern Alberta,

Canada. Palaeontographica A, 283: 1-33.

Etheridge, R. Jr 1888. On additional evidence of the genus

Ichthyosaurus in the Mesozoic rocks (‘Rolling Downs Formation’)

of north-eastern Australia. Proceedings of the Linnean Society of

New South Wales 2: 405-409.

Etheridge, R. Jr 1897. An Australian sauropterygian ( Cimoliasaurus ),

converted into precious opal. Records of the Australian Museum

3: 21-29.

Etheridge, R. Jr 1904. A second sauropterygian converted to opal,

from the Upper Cretaceous of White Cliffs, New South Wales.

Records of the Australian Museum 5: 306-316.

Fischer, V., Arkhangelsky, M.S., Naish, D., Stenshin, I.M., Uspensky,

G.N., and Godefroit, P. 2014. Simbirskiasaurus and

Pervushovisaurus reassessed: implications for the taxonomy and

cranial osteology of Cretaceous platypterygiine ichthyosaurs.

Zoological Journal of the Linnean Society 171: 822-841.

Fitzgerald, E.M.G., Carrano, M.T., Holland, T., Wagstaff, B.E.,

Pickering, D., Rich, T.H., and Vickers-Rich, P. 2012. First

ceratosaurian dinosaur from Australia. Naturwis sens chaf ten 99:

397-405.

Foffa, D., Sassoon, J., Cuff, A.R., Mavrogordato, M.N., and Benton,

M. J. 2014. Complex rostral neurovascular system in a giant

pliosaur. Naturwis sens chaf ten 101: 453-456.

Gaffney, E.S. 1981. A review of the fossil turtles of Australia.

American Museum Novitates 2720: 1-38.

Gasparini, Z., and Spalletti, L.A. 1990. Un nuevo crocodilo en los

depositos mareales maastrichtianos de la Patagonia noroccidental.

Ameghiniana 27: 141-150.

Glen, C., and McHenry, C. 2007. Preliminary report on a plesiosaur

from the early Cretaceous of central Queensland, Australia.

Journal of Vertebrate Paleontology 27: 82A-82A.

Gradstein, F.J., Ogg, J.G., Schmitz, M.D., and Ogg, G.M. 2012.

The Geologic Time Scale 2012. Volume 2. Elsevier: Amsterdam,

1144 pp.

Hirayama, R. 1998. Oldest known sea turtle. Nature 392: 705-708.

Hocknull, S.A., White, M.A., Tischler, T.R., Cook, A.G., Calleja,

N. D., Sloan, T., and Elliott, D.A. 2009. New mid-Cretaceous

(latest Albian) dinosaurs from Winton, Queensland, Australia.

PLoS ONE 4(T)\ e6190.

Hooks, G.E. Ill 1998. Systematic revision of the Protostegidae, with a

redescription of Calcarichelys gemma Zangerl, 1953. Journal of

Vertebrate Palaeontology 18: 85-98.

Houssaye, A. 2013. Bone histology of aquatic reptiles: what does it tell

us about secondary adaptation to an aquatic life? Biological

Journal of the Linnean Society 108: 3-21.

Houssaye, A., De Buffrenil, V., Rage, J.C., and Bardet, N. 2008. An

analysis of vertebral ‘pachyostosis’ in Carentonosaurus mineaui

(Mosasauroidea, Squamata) from the Cenomanian (early Late

Cretaceous) of France, with comments on its phylogenetic and

functional significance. Journal of Vertebrate Paleontology 28,

685-691.

Joyce, W.G. 2007. Phylogenetic relationships of Mesozoic turtles.

Bulletin of the Peabody Museum of Natural History 48: 3-102.

Kear, B.P 2001a. Elasmosaur (Reptilia: Plesiosauria) basicranial

remains from the Early Cretaceous of Queensland. Records of the

South Australian Museum 34: 127-133.

Kear, B.P. 2001b: Dental caries in an Early Cretaceous ichthyosaur.

Alcheringa 25: 387-390.

Kear, B.P. 2002a. Reassessment of the Early Cretaceous plesiosaur

Cimoliasaurus maccoyi Etheridge, 1904 (Reptilia: Sauropterygia)

from White Cliffs, New South Wales. Australian Journal of

Zoology 50: 671-685.

Kear, B.P. 2002b. Darwin Formation (Early Cretaceous, Northern

Territory) marine reptile remains in the South Australian Museum.

Records of the South Australian Museum 35: 33-47.

Kear, B.P. 2003. Cretaceous marine reptiles of Australia: a review of

taxonomy and distribution. Cretaceous Research 24: 277-303.

Kear, B.P. 2004. Biogeographic and biostratigraphic implications of

Australian Mesozoic marine reptiles. Australian Biologist 17,

4-22.

Kear, B.P. 2005a. Marine reptiles from the Lower Cretaceous (Aptian)

deposits of White Cliffs, southeastern Australia: implications of a

high-latitude cold water assemblage. Cretaceous Research 26:

769-782.

BP. Kear

Kear, B.P. 2005b. A new elasmosaurid plesiosaur from the Lower

Cretaceous of Queensland, Australia. Journal of Vertebrate

Paleontology 25: 792-805.

Kear, B.P. 2005c. Cranial morphology of Platypterygius longmani

Wade, 1990 (Reptilia: Ichthyosauria) from the Lower Cretaceous of

Australia. Zoological Journal oftheLinnean Society 145: 583-622.

Kear, B. P. 2006a. Marine reptiles from the Lower Cretaceous of

South Australia: elements of a high-latitude cold water assemblage.

Palaeontology 49: 837-856.

Kear, B.P. 2006b. Plesiosaur remains from Cretaceous high-latitude

non-marine deposits in southeastern Australia. Journal of

Vertebrate Paleontology 26: 196-199.

Kear, B.P. 2006c. First gut contents in a Cretaceous sea turtle. Biology

Letters 2: 113-115.

Kear, B.P. 2006d. Reassessment of Cratochelone berneyi Longman,

1915, a giant Early Cretaceous sea turtle from Australia. Journal

of Vertebrate Paleontology 26: 779-783.

Kear, B.P. 2007a. Taxonomic clarification of the Australian

elasmosaurid Eromangasaurus, with reference to other austral

elasmosaur taxa. Journal of Vertebrate Paleontology 27: 241-246.

Kear, B.P. 2007b. A juvenile pliosauroid plesiosaur from the Lower

Cretaceous of South Australia. Journal of Paleontology 81:

154-162.

Kear, B.P. 2012. A revision of Australia’s Jurassic plesiosaurs.

Palaeontology 55: 1125-1138.

Kear, B.P, and Lee, M.S.Y. 2006. A primitive protostegid from

Australia and early sea turtle evolution. Biology Letters 2: 116-119.

Kear, B.P., and Siverson, M. 2010. First evidence of a Late Cretaceous

sea turtle from Australia. Alcheringa 34: 265-272.

Kear, B.P., and Barrett, P.M. 2011. Reassessment of the Lower

Cretaceous (Barremian) pliosauroid Leptocleidus superstes

Andrews, 1922 and other plesiosaur remains from the non-marine

Wealden succession of southern England. Zoological Journal of

the Linnean Society 161: 663-691.

Kear, B.P, and Godthelp, H. 2008. Inferred vertebrate bite marks on

an Early Cretaceous unionoid bivalve from Lightning Ridge, New

South Wales, Australia. Alcheringa 32: 65-71.

Kear, B.P., and Hamilton-Bruce, R.J. 2011. Dinosaurs in Australia.

Mesozoic Life from the Southern Continent. CSIRO Publishing:

Melbourne, 190 pp.

Kear, B.P., and Zammit, M. 2014. In utero foetal remains of the

Cretaceous ichthyosaurian Platypterygius: ontogenetic implications

for character state efficacy. Geological Magazine 151: 71-86.

Kear, B.P, Boles, W.E., and Smith, E.T. 2003. Unusual gut contents in

a Cretaceous ichthyosaur. Proceedings of the Royal Society of

London, Series B 270: S206-S208.

Kear, B.P., Long, J.A., and Martin, J.E. 2005. A review of Australian

mosasaur occurrences. Netherlands Journal of Geosciences 84:

307-313.

Kear, B.P., Schroeder, N.I., and Lee, M.S.Y. 2006a. An archaic crested

plesiosaur in opal from the Lower Cretaceous high latitude

deposits of Australia. Biology Letters 2: 615-619.

Kear, B.P., Schroeder, N.I., Vickers-Rich, P, and Rich, T.H. 2006b.

Early Cretaceous high latitude marine reptile assemblages from

southern Australia. Paludicola 5: 200-205.

Kellner, A.W., Rich, T.H., Costa, F.R., Vickers-Rich, P., Kear, B.P,

Walters, M., and Kool, L. 2010. New isolated pterodactyloid

bones from the Albian Toolebuc Formation (western Queensland,

Australia) with comments on the Australian pterosaur fauna.

Alcheringa 34: 219-230.

Ketchum, H.F., and Benson, R.B.J., 2010. Global interrelationships of

Plesiosauria (Reptilia, Sauropterygia) and the pivotal role of taxon

sampling in determining the outcome of phylogenetic analyses.

Biological Reviews 85: 361-392.

Knight, W. C. 1900. Some new Jurassic vertebrates. American Journal

of Science, Fourth Series 10 (160): 115-119.

Lehman, T.M., and Tomlinson, S.L., 2004. Terlinguachelysfischbecki,

anew genus and species of sea turtle (Chelonioidea: Protostegidae)

from the Upper Cretaceous of Texas. Journal of Paleontology 78:

1163-1178.

Lindgren, J., Sjovall, P., Carney, R.M., Uvdal, P., Gren, J.A., Dyke, G.,

Schultz, B.P., Shawkey, M.D., Barnes, K.R., and Polcyn, M.J.

2014. Skin pigmentation provides evidence of convergent

melanism in extinct marine reptiles. Nature 506: 484-488.

Long, J.A., and Cruickshank, A.R.I. 1998. Further records of

plesiosaurian reptiles of Jurassic and Cretaceous age from

Western Australia. Records of the Western Australian Museum

19: 47-55.

Longman, H.A. 1915. On a giant turtle from the Queensland Lower

Cretaceous. Memoirs of the Queensland Museum 3: 24-29.

Longman, H.A. 1922. An ichthyosaurian skull from Queensland.

Memoirs of the Queensland Museum 7: 246-256.

Longman, H.A. 1924. A new gigantic marine reptile from the

Queensland Cretaceous, Kronosaurus queenslandicus new genus

and species. Memoirs of the Queensland Museum 8: 26-28.

Longman, H.A. 1930. Kronosaurus queenslandicus. A gigantic

Cretaceous pliosaur. Memoirs of the Queensland Museum 10:

1-7.

Longman, H.A. 1932. Restoration of Kronosaurus queenslandicus.

Memoirs of the Queensland Museum 10: 98.

Longman, H.A. 1935. Palaeontological notes. Memoirs of the

Queensland Museum 10: 236-239.

Longman, H.A. 1943. Further notes on Australian ichthyosaurs.

Memoirs of the Queensland Museum 12: 101-104.

Lundelius, E., and Warne, S.St.J. 1960. Mosasaur remains from the

Upper Cretaceous of Western Australia. Journal of Paleontology

34: 1215-1217.

Lydekker, R. 1889. Catalogue of the Fossil Reptilia and Amphibia in

the British Museum (Natural History). Part III. Order Chelonia.

British Museum (Natural History): London, 309 pp.

Makadi, L., Caldwell, M.W., and Osi, A. 2012. The first freshwater

mosasauroid (Upper Cretaceous, Hungary) and a new clade of

basal mosasauroids. PloS ONE 7(12): e51781.

Maxwell, E.E., and Kear, B.P. 2010. Postcranial anatomy of

Platypterygius americanus (Reptilia: Ichthyosauria) from the

Cretaceous of Wyoming. Journal of Vertebrate Paleontology 30:

1059-1068.

Maxwell, E.E., Caldwell, M.W., and Lamoureux, D.O. 2011. Tooth

histology in the Cretaceous ichthyosaur Platypterygius australis,

and its significance for the conservation and divergence of

mineralized tooth tissues in amniotes. Journal of Morphology

272: 129-135.

Maxwell, E.E., Zammit, M., and Druckenmiller, PS. 2012.

Morphology and orientation of the ichthyosaurian femur. Journal

of Vertebrate Paleontology 32: 1207-1211.

McCoy, F. 1867a. On the occurrence of Ichthyosaurus and

Plesiosaurus in Australia. Annals and Magazine of Natural

History, Third Series 19: 355-356.

McCoy, F. 1867b. On the discovery of the Enaliosauria and other

Cretaceous fossils in Australia. Transactions and Proceedings of

the Royal Society of Victoria 8: 41-42.

McCoy, F. 1869. On the fossil eye and teeth of the Ichthyosaurus

australis (M’Coy), from the Cretaceous formations of the source

of the Flinder’s River; and on the palate of Diprotodon, from the

Tertiary limestone of Limeburner’s Point, near Geelong.

Transactions and Proceedings of the Royal Society of Victoria 2:

77-78.

Cretaceous marine amniotes of Australia: perspectives on a decade of new research

McGowan, C. 1972. The systematics of Cretaceous ichthyosaurs with

particular reference to the material from North America.

Contributions to Geology, University of Wyoming 11: 9-29.

McGowan, C., and Motani, R. 2003. Ichthyopterygia. Handbuch der

Paleoherpetologie Part 8. Verlag Dr. Friedrich Pfeil: Munchen.

175 pp.

McHenry, C.R. 2009. ‘Devourer of gods’. The palaeoecology of the

Cretaceous pliosaur Kronosaurus queenslandicus. Ph.D. Thesis,

University of Newcastle: Newcastle. 616 pp.

McHenry, C.R., Cook, A.G., and Wroe, S. 2005. Bottom-feeding

plesiosaurs. Science 310: 75.

Medeiros, M.A. 2006. Large theropod teeth from the Eocenomanian

of northeastern Brazil and the occurrence of Spinosauridae.

Revista brasileira de Paleontologia 9: 333-338.

Molnar, R.E. 1991. Fossil reptiles in Australia. Pp. 605-702 in:

Vickers-Rich, P, Monaghan, J.M., Baird, R.F., and Rich, T. H.

(eds). Vertebrate Palaeontology of Australasia. Pioneer Design

Studio, Monash University: Melbourne, 1437 pp.

Mory, A.J., Haig, D.W., McLoughlin, S., and Hocking, R.M. 2005.

Geology of the northern Perth Basin, Western Australia—a field

guide. Geological Survey of Western Australia Record 9: 1-71.

Murray, P.F., 1985. Ichthyosaurs from Cretaceous Mullaman Beds

near Darwin, Northern Territory. The Beagle 2: 39-55.

Murray, PF. 1987. Plesiosaurs from Albian aged Darwin Formation

siltstones near Darwin, Northern Territory, Australia. The Beagle

4: 95-102.

Myers, T. 2007. Osteological morphometries of Australian chelonioid

sea turtles. Zoological Science 24: 1012-1027.

O’Gorman, J.P., and Gasparini, Z., 2013. Revision of Sulcusuchus

erraini (Sauropterygia, Polycotylidae) from the Upper Cretaceous

of Patagonia, Argentina. Alcheringa 37: 161-174.

O’Gorman, J. P., Otero, R. A., and Hiller, N. 2014. A new record of an

aristonectine elasmosaurid (Sauropterygia, Plesiosauria) from the

Upper Cretaceous of New Zealand: implications for the Mauisaurus

haasti Hector, 1874 hypodigm. Alcheringa 38: 504-512.

O’Gorman, J. P., Gasparini, Z. B., and Salgado, L. 2013. Postcranial

morphology of Aristonectes (Plesiosauria, Elasmosauridae) from

the Upper Cretaceous of Patagonia and Antarctica. Antarctic

Science 25: 71-82.

O’Keefe, F.R. 2001. A cladistic analysis and taxonomic revision of the

Plesiosauria (Reptilia, Sauropterygia). Acta Zoologica Fennica

213: 1-63.

O’Keefe, F.R. 2004. On the cranial anatomy of the polycotylid

plesiosaurs, including new material of Polycotylus latipinnis

Cope, from Alabama. Journal of Vertebrate Paleontology 24:

326-340.

O’Keefe, F.R. 2008. Cranial anatomy and taxonomy of Dolichorhynchops

bonneri new combination, a polycotylid (Sauropterygia,

Plesiosauria) from the Pierre Shale of Wyoming and South Dakota.

Journal of Vertebrate Paleontology 28: 664-676.

O’Keefe, F. R., and Hiller, N. 2006. Morphologic and ontogenetic

patterns in elasmosaur neck length, with comments on the

taxonomic utility of neck length variables. Paludicola 5: 206-229.

O’Keefe, F. R., and Street, H. P. 2009. Osteology of the cryptocleidoid

plesiosaur Tatenectes laramiensis, with comments on the

taxonomic status of the Cimoliasauridae. Journal of Vertebrate

Paleontology 29: 48-57.

Otero, R. A., Soto-Acuna, S., O’Keefe, F. R., O’Gorman, J. P.,

Stinnesbeck, W., Suarez, M. E., Rubilar-Rogers, D., Salazar, C.,

and Quinzio-Sinn, L. A. 2014. Aristonectes quiriquinensis, sp.

nov., a new highly derived elasmosaurid from the upper

Maastrichtian of central Chile. Journal of Vertebrate Paleontology

34: 100-125.

Owen, R. 1882. On an extinct chelonian reptile ( Notochelys costata,

Owen) from Australia. Quarterly Journal of the Geological

Society of London 38: 178-183.

Parham, J.F., and Pyenson, N.D. 2010. New sea turtle from the

Miocene of Peru and the iterative evolution of feeding

ecomorphologies since the Cretaceous. Journal of Paleontology

84:231-247.

Persson, P.O. 1960. Early Cretaceous plesiosaurians (Reptilia) from

Australia. Lunds Universitets Arsskrift 56: 1-23.

Persson, P.O. 1982. Elasmosaurid skull from the Lower Cretaceous of

Queensland (Reptilia, Sauropterygia). Memoirs of the Queensland

Museum 20: 647-655.

Poropat, S.F., Mannion, P.D., Upchurch, P., Hocknull, S.A., Kear, B.P,

& Elliott, D.A. 2014. Reassessment of the non-titanosaurian

somphospondylan Wintonotitan wattsi (Dinosauria: Sauropoda:

Titanosauriformes) from the mid-Cretaceous Winton Formation,

Queensland, Australia. Papers in Palaeontology 1: 59-106.

Poropat, S.F., Upchurch, P., Mannion, P.D., Hocknull, S.A., Kear, B.P,

Sloan, T., Sinapius, G.H.K., and Elliott, D.A. 2015. Revision of the

sauropod dinosaur Diamantinasaurus matildae Hocknull et al.

2009 from the mid-Cretaceous of Australia: Implications for

Gondwanan titanosauriform dispersal. Gondwana Research 27:

995-1033.

Pridmore, P.A., Rich, T.H., Vickers-Rich, P, and Gambaryan, PP.

2005. A tachyglossid-like humerus from the Early Cretaceous of

south-eastern Australia. Journal of Mammalian Evolution 12:

359-378.

Rhodin, A.G. 1985. Comparative chondro-osseous development and

growth of marine turtles. Copeia 1985: 752-771.

Rich, T.H., and Vickers-Rich, P. 2003. Diversity of Early Cretaceous

mammals from Victoria, Australia. Bulletin of the American

Museum of Natural History 285: 36-53.

Rich, T.H., Vickers-Rich, P., Flannery, T.F., Kear, B.P., Cantrill, D.,

Komarower, P, Kool, L., Pickering, D., Trussler, P., Morton, S.,

Van Klaveren, N., and Fitzgerald, E.M.G. 2009. An Australian

multituberculate and its palaeobiogeographical implications. Acta

Palaeontologica Polonica 54: 1-6.

Rich, T.H., Kear, B.P., Sinclair, R., Chinnery, B., Carpenter, K.,

McHugh, M.L. and Vickers-Rich, P. 2014. Serendipaceratops

arthurcclarkei Rich & Vickers-Rich, 2003 is an Australian Early

Cretaceous ceratopsian. 38: 456-479.

Richter, U., Mudroch, A., and Buckley, L.G. 2013. Isolated theropod

teeth from the Kem Kem beds (early Cenomanian) near Taouz,

Morocco. Palciontologische Zeitschrift 87: 291-309.

Romer, A.S., and Lewis, A.D. 1959. A mounted skeleton of the giant

plesiosaur Kronosaurus. Breviora 112: 1-15.

Rowe, T., Rich, T.H., Vickers-Rich, P, Springer, M., and Woodburne,

M.O. 2008. The oldest platypus and its bearing on divergence

timing of the platypus and echidna clades. PNAS 1054: 1238-1242.

Russell, D.A. 1967. Systematics and morphology of American

mosasaurs (Reptilia: Sauria). Bulletin of the Peabody Museum of

Natural History, Yale University 23: 1-240.

Sachs, S. 2004. Redescription of Woolungasaurus glendowerensis

(Plesiosauria: Elasmosauridae) from the Lower Cretaceous of

Northeast Queensland. Memoirs of the Queensland Museum 49:

713-731.

Sachs, S. 2005. Tuarangisaurus australis sp. nov. (Plesiosauria:

Elasmosauridae) from the Lower Cretaceous of Northeastern

Queensland, with addional notes on the phylogeny of

the Elasmosauridae. Memoirs of the Queensland Museum 50:

425-440.

Sachs, S., and Grant-Mackie, J.A. 2003. An ichthyosaur fragment

from the Cretaceous of Northland, New Zealand. Journal of the

Royal Society of New Zealand 33: 307-314.

BP. Kear

Sachs, S. and Kear, B.P. 2015. Postcranium of the paradigm

elasmosaurid plesiosaurian Libonectes morgani (Welles, 1949).

Geological Magazine 152: 649-710.

Sachs, S., Kear, B.P., and Everhart M.J. 2013. Revised vertebral count

in the “longest-necked vertebrate” Elasmosaurus platyurus Cope

1868, and clarification of the cervical-dorsal transition in

Plesiosauria. PLoS ONE 8(8): e70877.

Sachs, S., Schubert, S., and Kear, B.P. 2014. Note on anew plesiosaur

(Reptilia: Sauropterygia) skeleton from the upper Pliensbachian

(Lower Jurassic) of Bielefeld, northwest Germany. Berichte

Naturwissenschaftlicher Vereinfiir Bielefeld und Umgegend 52:

26-35.

Salisbury, S.W., Molnar, R.E., Frey, E., and Willis, P.M. 2006. The

origin of modern crocodyliforms: new evidence from the

Cretaceous of Australia. Proceedings of the Royal Society, Series

B 273: 2439-2448.

Sato, T. 2005. A new polycotylid plesiosaur (Reptilia: Sauropterygia)

from the Upper Cretaceous Bearpaw Formation of Saskatchewan,

Canada. Journal of Palaeontology 79: 969-980.

Sato, T., and Storrs, G.W. 2000. An early polycotylid plesiosaur

(Reptilia, Sauropterygia) from the Cretaceous of Hokkaido,

Japan. Journal of Paleontology 74: 907-914.

Sato, T., Li, C., and Wu, X.C. 2003. Restudy of Bishanopliosaurus

youngi Dong 1980, a freshwater plesiosaurian from the Jurassic of

Chongqing. Vertebrata PalAsiatica 41: 18-33.

Scanlon, J.D., and Hocknull, S.A. 2008. A dolichosaurid lizard from

the latest Albian (mid-Cretaceous) Winton Formation,

Queensland, Australia. Pp. 131-136 in: Everhart, M.J. (ed).

Proceedings of the Second Mosasaur Meeting. Transactions of

the Kansas Academy of Science, Fort Hays Studies Special Issue

3, 172 pp.

Schmeisser McKean, R. 2012. A new species of polycotylid plesiosaur

(Reptilia: Sauropterygia) from the Lower Turonian of Utah:

extending the stratigraphic range of Dolichorhynchops.

Cretaceous Research 34: 184-199.

Smith, E.T. 2010. Early Cretaceous chelids from Lightning Ridge,

New South Wales. Alcheringa 34: 375-384.

Smith, E.T., and Kear, B.P. 2013. Spoochelys ormondea gen. et sp.

nov., an archaic meiolaniid-like turtle from the Lower Cretaceous

of Lightning Ridge, Australia. Pp. 277-287 in: Brinkman, D.,

Holroyd, P., and Gardner, J. (eds). Morphology and Evolution of

Turtles. Springer, Dordrecht: The Netherlands, 577 pp.

Smith, N.D., Makovicky, P.J., Agnolin, F.L., Ezcurra, M.D., Pais, D.F.,

and Salisbury, S.W. 2008. A Megaraptor -like theropod

(Dinosauria: Tetanurae) in Australia: support for faunal exchange

across eastern and western Gondwana in the Mid-Cretaceous.

Proceedings of the Royal Society B 275: 2085-2093.

Teichert, C., and Matheson, R. 1944. Upper Cretaceous ichthyosaurian

and plesiosaurian remains from Western Australia. Australian

Journal of Science 6: 167-178.

Thulborn, T., and Turner, S. 1993. An elasmosaur bitten by a pliosaur.

Modern Geology 18: 489-501.

Thulborn, T., and Turner, S. 2003. The last dicynodont: an Australian

Cretaceous relict. Proceedings of the Royal Society of London,

Series B 270: 985-993.

Wade, M. 1984. Platypterygius australis, an Australian Cretaceous

ichthyosaur. Lethaia 17: 93-113.

Wade, M. 1985. Platypterygius australis (McCoy) 1867: a Cretaceous

marine reptile. Pp. 137-142 in: Rich, P., and Van Tets, G. (eds),

Kadimakara. Princeton University Press: Princeton, 284 pp.

Wade, M. 1990. A review of the Australian Cretaceous longipinnate

ichthyosaur Platypterygius (Ichthyosauria, Ichthyopterygia).

Memoirs of the Queensland Museum 28: 115-137.

Welles, S.P. 1962. A new species of elasmosaur from the Aptian of

Colombia, and a review of the Cretaceous plesiosaurs. University

of California Publications in Geological Sciences 44: 1-96.

White, T.E. 1935. On the skull of Kronosaurus queenslandicus

Longman. Occasional Papers of the Boston Society of Natural

History 8:219-228.

Wieland, G.R. 1896. Archelon ischyros: a new gigantic cryptodire

testudinate from the Pierre Cretaceous of South Dakota. American

Journal of Science 2: 399-412.

Wood, R., Grove, C.J., Gaffnet, E.S., and Maley, K.F. 1996. Evolution

and phylogeny of leatherback turtles (Dermochelydae), with

descriptions of new fossil taxa. Chelonian Conservation and

Biology 2: 266-286.

Wretman, L., and Kear, B.P. 2014. Bite pathologies on an

ichthyodectiform fish skull from Australia: evidence of trophic

interaction in an Early Cretaceous marine ecosystem. Alcheringa

38: 170-176.

Zammit, M. 2010. A review of Australasian ichthyosaurs. Alcheringa

34: 281-292.

Zammit, M. 2012. Cretaceous ichthyosaurs: dwindling diversity, or

the empire strikes back? Geosciences 2: 11-24.

Zammit, M., and Kear, B.P. 2011. Healed bite marks on a Cretaceous

ichthyosaur. Acta Palaeontologica Polonica 56: 859-863.

Zammit, M., Norris, R., and Kear, B.P. 2010. The Australian

Cretaceous ichthyosaur Platypterygius australis: a description

and review of postcranial remains. Journal of Vertebrate

Paleontology 30: 1726-1735.

Zammit, M., Kear, B.P., and Norris, R. 2014. Locomotory capabilities

in the Early Cretaceous ichthyosaur Platypterygius australis

based on osteological comparisons with extant marine mammals.

Geological Magazine 151: 87-99.

Memoirs of Museum Victoria 74:29-48 (2016) Published 2016

ISSN 1447-2546 (Print) 1447-2554 (On-line)

http://museumvictoria.com.au/about/books-and-journals/journals/memoirs-of-museum-victoria/

A new specimen of Valdosaurus canaliculatus (Ornithopoda: Dryosauridae) from

the Lower Cretaceous of the Isle of Wight, England

Paul M. Barrett

Department of Earth Sciences, The Natural History Museum, Cromwell Road, London SW7 5BD, UK

(p.barrett@nhm.ac.uk)

Abstract Barrett, RM. 2016. A new specimen of Valdosaurus canaliculatus (Ornithopoda: Dryosauridae) from the Lower Cretaceous

of the Isle of Wight, England. Memoirs of Museum Victoria 74: 29-48.

The anatomy of Valdosaurus canaliculatus is incompletely known and until recently was based exclusively upon

the holotype femora. Additional discoveries from the Wessex Formation (Barremian) of the Isle of Wight during the

past decade have considerably expanded the amount of material available and offered insights into the morphology of

the vertebral column and pelvis. However, all of these specimens consist primarily of hind limb material. Here, I

describe a newly discovered individual of this taxon, the most complete yet found, which was found in articulation

and includes a partial dorsal series, an almost complete tail, pelvic material, and both hind limbs. Although the

specimen is partially crushed it offers new information on the anatomy of Valdosaurus, facilitating comparisons with

other dryosaurid taxa.

Keywords Wealden, Dinosauria, Wessex Formation, Barremian

Introduction

The Wessex Formation of the Isle of Wight (UK) has yielded

a rich Early Cretaceous (Barremian) dinosaur fauna (Martill

and Naish, 2001) that includes four well-established ornithopod

taxa: the non-iguanodontian Hypsilophodon foxii (Huxley,

1869), the dryosaurid Valdosaurus canaliculatus (Galton,

1975) and the ankylopollexians Mantellisaurus atherfieldensis

(Hooley, 1925) and Iguanodon bernissartensis Boulenger in

Beneden, 1881. It is possible that other taxa, represented by

isolated but potentially distinctive material, were also present

(Galton, 2009) and other names have been proposed for

Wessex Formation iguanodontian taxa (e.g. Hulke, 1879,1882;

Lydekker, 1888; Paul, 2008; Carpenter and Ishida, 2010), but

these are currently regarded either as nomina dubia or as

junior synonyms of the aforementioned taxa (Norman, 2004,

2012; McDonald, 2012).

Hypsilophodon, Mantellisaurus and Iguanodon are

known on the basis of complete or near-complete specimens

from the Isle of Wight and other Early Cretaceous strata in

the UK and elsewhere in western Europe: each has been the

focus of numerous studies, including comprehensive

monographs (e.g. Hulke, 1882; Hooley, 1925; Galton, 1974;

Norman, 1980, 1986). By contrast, Valdosaurus is relatively

poorly known. This taxon was established on the basis of an

associated pair of small femora (Galton, 1975), but the

collection of new specimens and reinterpretation of historical

collections has expanded both the amount and quality of

material available (Galton, 2009; Barrett et al., 2011) to

comprise four previously reported partial skeletons. Other

dryosaurid material is also known from the Wessex Formation

of the Isle of Wight and the Tunbridge Wells Sands Formation

(Valanginian) and Weald Clay Subgroup (Barremian) of

southern England, some of which may be referable to

Valdosaurus (Blows, 1998; Naish and Martill, 2001; Galton,

2009; Barrett et al., 2011; pers. obs.). Unusually, all previously

described specimens consist almost exclusively of hind limb

elements, with sporadic preservation of material from the

axial column and pelvic girdle (Barrett et al., 2011: BELUM

K17051; IWCMS 2007.4; MIWG.6438, MIWG.6879;

NHMUK PV R184, R185). Here, I describe a new specimen

of Valdosaurus from the Wessex Formation of the Isle of

Wight, the most complete yet found, and discuss its

implications.

Institutional abbreviations

BELUM, Ulster Museum (National Museums of Northern

Ireland), Belfast, UK; BMB, Booth Museum, Brighton, UK;

IWCMS (formerly MIWG, note that both sets of abbreviations

still in use), Dinosaur Isle, Sandown, UK; MUCPv, Universidad

Nacional del Comahue, Paleovertebrate Collection, Lago

Barreales, Argentina; NHMUK, Natural History Museum,

London, UK; NMV, Museum Victoria, Melbourne, Australia.

P.M. Barrett

Material

The specimen (IWCMS 2013.175) was discovered by Mr Nick

Chase in October 2012, in a plant debris bed (sensu Sweetman

and Insole, 2010) cropping out from a cliff on a National Trust

owned property in Compton Bay, on the southwest coast of the

Isle of Wight. As other dinosaur material was discovered

nearby, the exact locality details are withheld to prevent theft

from the site, but are held on file at IWCMS. The skeleton was

excavated from the Wessex Formation (Lower Cretaceous:

Barremian; Allen and Wimbledon, 1991; Rawson, 2006), a

sequence of terrestrial mudstones and sandstones deposited by

rivers and lakes on a seasonally arid floodplain that was

subject to seasonal wildfires (Martill and Naish, 2001).

IWCMS 2013.175 comprises a partial postcranial skeleton

representing the hindquarters of a single individual. The

specimen was found in situ within the plant debris bed and in

articulation (fig. 1). It consists of six middle-posterior dorsal

vertebrae (the only part of the specimen found weathered out

at the surface), the sacrum, an articulated series of 45 caudal

vertebrae (including chevrons in the anterior part of the tail),

both pelvic girdles, and both hind limbs (lacking many of the

pedal phalanges, but otherwise complete). Ossified tendons

are also present in the sacral and caudal regions. Although

surface collected, the proximity and characteristics of the

dorsal vertebrae clearly demonstrate that they pertain to the

same individual. Most parts of the skeleton have been fully

prepared, but matrix has been retained around the sacroiliac

block, obscuring many details of the sacrum and the medial

surfaces of the ilia. In addition, as the pelvic girdles and

proximal parts of the femora are articulated, these elements

often overlie each other preventing full description.

As the hind limb anatomy of Valdosaurus has already

been described in detail on the basis of several other specimens

(Galton, 1975, 2009; Barrett et al., 2011: BELUM K17051;

IWCMS 2007.4; MIWG.6438, MIWG.6879; NHMUK PV

R184, R185), and as those of IWCMS 2013.175 are rather

poorly preserved in spite of their completeness, the following

description focuses on those areas of the anatomy that were

previously unknown or poorly documented for Valdosaurus,

primarily the axial column and pelvic girdle.

The specimen can be referred to Valdosaurus canaliculatus

on the basis of the following femoral characters, which are

currently regarded as either autapomorphic or part of a unique

character combination for the taxon (Barrett et al., 2011): the

anterior trochanter extends dorsally to almost the same level

as the greater trochanter; and the left femur possesses an

anterior intercondylar groove that is broad, parallel-sided and

'U’-shaped (although crushing has partially closed the groove).

Moreover, the combined presence of a proximally placed

fourth trochanter and of a deep anterior intercondylar groove

is unique for Valdosaurus among the ornithopods currently

known from the Wealden Group of the UK (Galton, 1975,

2009; Barrett et al., 2011).

All measurements for IWCMS 2013.175 can be found in

Tables 1-5.

Description

Dorsal Vertebrae. A series of six middle-posterior dorsal (D)

vertebrae is present, all of which are completely prepared and

visible in all views. With the exception of two conjoined

vertebrae, they were not found in articulation, but can be

arranged into a series on the basis of size that suggests they did

originally form a natural sequence. The following numbering

scheme labels each of the vertebrae from 1 (anterior-most) to 6

(posterior-most), as arranged in the exhibition mount. There

are no indications of any tendons along the dorsal series, but

this observation is tentative due to the fragmentary nature of

the neural arches and the exposure of the vertebrae at the

surface prior to collection.

None of the vertebrae are complete, although each centrum

is present (see table 1 for measurements). All six dorsals have

been sheared and somewhat distorted, with the neural arch

(where present) rotated relative to the centrum. The left-hand

sides of all six centra are damaged: those of D1 and 2 bear

indentations on their left lateral surfaces, whereas in D3-6

this entire surface is either crushed inward or poorly preserved.

Some parts of the centra in D4-6 are missing. All of the

dorsals preserve small parts of the neural arch. The position of

the neurocentral junction can be seen in several of the

vertebrae, but it is unclear whether it was open or partially

fused. As D1 is the best preserved of the series most of the

following description is derived from this vertebra, but this

has been augmented with details from the other dorsals where

possible (fig. 2).

D1 consists of a complete centrum, with a partial neural

arch that is lacking all of the processes (fig. 2A-F). The arch

has been rotated relative to the centrum, so that it is displaced

to the right. The course of the neurocentral suture is clearly

visible on the right-hand side though it is not clear if it is still

open or partially fused. In anterior view, the articular surface

has a subcircular outline, with a straight dorsal margin and

rounded lateral and ventral margins that meet along a

continuous smooth curve. The articular surface is shallowly

concave and bordered by a raised rim on all sides. In lateral

view, the centrum is elongate, approximately 1.9 times as long

as it is high and has a ventral margin that is gently concave

dorsally. The lateral surface of the centrum is anteroposteriorly

concave and dorsoventrally convex and is pierced by a small

Table 1. Measurements (in mm) of the dorsal (D) vertebrae of

Valdosaurus canaliculatus (IWCMS 2013.175). Abbreviations: ACH,

anterior centrum height ACW, anterior centrum width; CL, centrum

length; PCH, posterior centrum height; PCW, posterior centrum width,

indicates that the measurement as been affected by deformation.

ACW

ACH

PCW

PCH

38*

41*

47*

42*

44*

58*

49*

56*

New Valdosaurus

Figure 1. New specimen of Valdosaurus canaliculatus (IWCMS 2013.175). A, field map of the discovery showing the position of the skeleton in

situ and the blocks in which the elements were originally excavated. Drawn by N. Chase, original on file at IWCMS. B, skeleton following

preparation, as displayed in Dinosaur Isle, in left lateral view. Scale bars = 200 mm.

elliptical nutrient foramen at midlength. The lateral surfaces

merge smoothly with the ventral surface and there is no distinct

break of slope or ridge to separate them. In ventral view, the

centrum is hourglass-shaped: there is some suggestion of a

ventral keel, but this has probably been accentuated by mild

crushing. Prominent longitudinally oriented striations extend

along the lateral and ventral surfaces of the centrum from their

junctions with the anterior and posterior surfaces. The posterior

articular surface is also subcircular in outline, though the

articular surface is flat, rather than concave.

The neural arch extends for almost the full length of the

centrum. Crushing and breakage obscure much of the

morphology of the processes and of the neural canal outline.

However, it can be determined that a thin lamina extended

posteriorly from the lateral margin of the prezygapophysis

towards the diapophysis (prezygodiapophyseal lamina: PRDL;

lamina terminology follows Wilson, 1999). The PRDL and

anterior centrodiapophyseal lamina (ACDL) together frame a

deep, anterolaterally facing prezygopophyseal

centrodiapophyseal fossa (prcdf; fossa terminology follows

Wilson et al., 2011) with a subelliptical outline in lateral view.

There is no sign of a parapophysis on the centrum or on the

ACDL, so this structure must have merged with the (missing)

diapophysis. A thick, buttress-like posterior centrodiapophyseal

lamina (PCDL) is also present, which with the ACDL frames a

shallow, laterally facing and extensive centrodiapophyseal fossa

(cdf). The PCDL merges with the base of the diapophysis to give

the latter a triangular cross-section, whose apex points ventrally.

A very small, elliptical spinoprezygapophyseal fossa (sprf) is

present at the base of the anterior surface of the neural spine. The

broken base of the neural spine indicates that it was mediolaterally

compressed and plate-like. A postzygodiapophyseal lamina

P.M. Barrett

Figure 2. Dorsal vertebrae of Valdosaurus canaliculatus (IWCMS 2013.175). A-F, dorsal 1 in A, anterior, B, left lateral, C, right lateral, D,

posterior, E, ventral, and F, dorsal views. G, dorsals 1-6 in articulation in left lateral view. Abbreviations: ACDL, anterior centrodiapophyseal

lamina; cdf, centrodiapophyseal fossa; NC, neural canal; PCDL, posterior centrodiapophyseal lamina; pocdf, postzygopophyseal

centrodiapophyseal fossa; PODL, postzygodiapophyseal lamina; prcdf, prezygopophyseal centrodiapophyseal fossa; PRDL, prezygodiapophyseal

lamina; sprf, spinoprezygapophyseal fossa. Scale bars = 25 mm (A-F) and 50 mm (G).

New Valdosaurus

Table 2. Measurements (in mm) of the caudal (Cd) vertebrae of

Valdosaurus canaliculatus (IWCMS 2013.175). Abbreviations: ACH,

anterior centrum height ACW, anterior centrum width; CL, centrum

length; PCH, posterior centrum height; PCW, posterior centrum

width. ‘*’ indicates that the measurement as been affected by

deformation. Many measurements are missing as obscured by

articulation or matrix or too deformed to be useful (all measurements

in the cases of Cdl3-14 and Cdl8).

ACW

ACH

PCW

PCH

Cdl

40*

45*

48*

49*

Cd2

54*

Cd3

49*

53*

50*

52*

50*

Cd4

49*

51*

54*

Cd5

36*

46*

Cd6

Cd7

60*

32*

42*

Cd8

Cd9

CdlO

Cdll

Cdl2

Cdl3

Cdl4

Cdl5

Cdl6

Cdl7

Cdl8

Cdl9

Cd20

Cd21

Cd22

Cd23

Cd24

Cd25

37*

33*

Cd26

Cd27

Cd28

Cd29

Cd30

Cd31

Cd32

Cd33

Cd34

Cd35

Cd36

Cd37

Cd38

Cd39

Cd40

Cd41

Cd42

Cd43

Cd44

Cd45

(PODL) also appears to have been present, framing a

posterolaterally opening postzygopophyseal centrodiapophyseal

fossa (pocdf) with the PCDL. Although broken, the posterior

part of the arch dorsal to the neural canal bears a small inverted

‘Y’-shaped process that extends ventral to the broken bases of

the postzygapophyses.

D2 consists of the centrum and a poorly preserved,

incomplete neural arch that lacks any indications of processes or

laminae other than the right PCDL (fig. 2G). As far as can be

determined, D2 is identical to D1 in almost every respect,

although the anterior articular surface is flat rather than concave.

D3 consists of a complete centrum and partial neural arch.

Deformation of the centrum is severe and the entire left-hand

side is crushed inward, which has created a pseudo-keel along

the ventral surface. As far as can be determined, the centrum and

neural arch is identical to that of D1 and 2. As in Dl, the ACDL

and PCDL are present and frame a shallow cdf. There is some

evidence for a PODL that frames a posteriorly opening and deep

pocdf with the PCDL. The bases of both diapophyses are present,

which are angled dorsolaterally with respect to the centrum, but

the angle between them has been exaggerated by deformation.

D4 is almost identical to D3 in terms of the parts preserved and

the features that can be assessed.

D5 and 6 each consists of a centrum and partial neural arch

(fig. 2G). The vertebrae are larger than the preceding ones and

are less elongate (length to height ratio of -1.3). The longitudinal

striations on the lateral surfaces of the centra are less prominent

than in the earlier dorsals. In all other respects, however, the

centra are similar to those of the other dorsals (with flat articular

surfaces, no keel, an hourglass-shaped outline in ventral view, no

distinct separation between the lateral and ventral surfaces,

single small nutrient foramen, etc.). D5 and 6 possess a buttress¬

like ACDL that is less distinct that those of earlier dorsals,

whereas the PCDL remains a clear and distinct lamina in all

preserved dorsals. Their prominent PCDL forms the posterior

margin of a deep subtriangular cdf and the anterior margin of a

smaller, but deep pocdf posteriorly. The dorsal margin of the

pocdf is provided by a prominent PODL. The postzygapophyses

of D5 are partially preserved but are largely obscured by the

overlapping prezygapophyses of D6. They appear to be very

steeply inclined forming an angle of almost 80° with the

horizontal plane in anterior view. The prezygapophyses of D6

only overlap the postzygapophyses of D5 for a short distance and

have narrow triangular transverse cross-sections.

Sacrum. Although the sacrum is present and likely complete, it

is largely obscured by matrix and the adjacent ilia. In ventral

view, a dorsosacral is exposed on the anterior part of the pelvic

block and the posterior margin of a posterior sacral and the

anterior part of the caudosacral (and associated sacral rib) can

be seen posterior to the ischia. The posterior part of the

caudosacral is articulated with the first caudal vertebra and is

thus preserved separately from the pelvic block. The number of

sacral vertebrae cannot be determined, but on the basis of the

lengths of those that are visible it seems likely that at least five

and potentially six were present (including those vertebrae

identified as the dorso- and caudosacrals). As each vertebra is

only partially visible in ventral view, the amount of anatomical

P.M. Barrett

information is limited. All three vertebrae have transversely

convex ventral surfaces that lack keels and longitudinal

grooves. The dorsosacral is identified as such as it bears a

short, anteroposteriorly narrow diapophysis that differs from

the expanded morphology that would be expected in ‘true’

sacral ribs. This diapophysis extends laterally to contact the

medial surface of the preacetabular process of the ilium. By

contrast, a large, fan-shaped sacral rib is borne by the

posterolateral corner of the exposed posterior sacral and the

anterolateral corner of the caudosacral. The medial part of the

rib is anteroposteriorly narrow, but flares within a short distance

to form a broad, subtriangular and dorsally convex sheet whose

scalloped lateral margin articulates with the ilium along almost

the full length of the brevis shelf.

Caudal vertebrae. Forty-five caudal (Cd) vertebrae are present

and it seems likely that almost the whole tail was preserved,

with the loss of only a few (if any) very small distal-most

vertebrae (see table 2 for measurements). Some of the vertebrae

have been fully prepared (Cdl-4), while others are partially

embedded in blocks of matrix and exposed in one view only.

The proximal caudals (Cdl-4) have suffered the most

deformation, mainly mild crushing and mediolateral shearing,

and have crazed bone surfaces. Nevertheless, they are largely

complete and the middle and posterior caudals are very well

preserved. Many of the vertebrae are still in articulation:

consequently, some individual details are obscured by the

presence of other vertebrae in the series.

Cdl and 2 are in the same block as the posterior part of the

caudosacral (fig. 3A, B). As a result the posterior surface of the

caudosacral and the anterior and posterior articular surfaces of

Cdl and the anterior surface of Cd2 are obscured. The right-

hand sides of the vertebrae are obscured by the overlying

posterior process of the left ilium, which is adhered to this block

and separated from the rest of the left ilium.

The centra have ventrally concave margins in lateral view

and the lateral surfaces are anteroposteriorly concave and

dorsoventrally convex. The longitudinal striations present on the

lateral surfaces of the dorsal vertebrae are absent, but the crazed

bone surface prevents determination of the presence/absence of

nutrient foramina. The lateral surfaces are distinctly separated

from the ventral surfaces by a break in slope and there is some

indication of a ventral groove on both Cdl and 2. The visible

parts of the articular surfaces suggest that they were very

shallowly convex. Poor preservation prevents identification of

the neurocentral sutures. There is no evidence for a chevron facet

on Cdl, but a posterior facet is present on Cd2. This chevron

facet is a flat, subtrapezoidal surface, which is indented on the

ventral midline.

The neural arches are reasonably complete, but some details

are obscured by matrix, deformation and the apposition of the

ilium and caudosacral. The prezygapophyses of Cd2 (missing in

Cdl) are short, stout triangular processes that only extend a short

distance beyond the anterior margin of the centrum. In both Cdl

and 2, the caudal ribs are broken, with only the remnants of the

bases present. The neural spines are inclined posteriorly and

dorsally, so that their posterior margins form an angle of

approximately 45° with the horizontal. They are mediolaterally

compressed, maintain the same mediolateral thickness along

their entire length, are anteroposteriorly short, and are taller (as

measured from the level of postzygapophyses to the tip of the

spine) than their respective centra are long. They are slightly

constricted in lateral view just above the level of the

postzygapophyses and expand again slightly anteroposteriorly

above this point. In lateral view, the neural spine of Cdl extends

to terminate posterior to the anterior margin of Cd3: the overlap

between the neural spine and the next two vertebrae in the series

continues throughout the proximal and middle sections of the

tail. The spine of Cdl is more expanded anteroposteriorly in its

distal part than that of Cd2. The neural spine summits are gently

rounded in lateral view. In contrast to the dorsal series, neural

arch laminae are absent. Postzygapophyses are present in Cdl

(missing from Cd2) and are short, robust and have squat

subtriangular transverse cross-sections.

Cd3 and 4 are in articulation, but their neural arches are

slightly detached from their respective centra (fig. 3C).

Deformation makes it difficult to assess the morphology, but

there is some evidence that the lateral surfaces met along a

grooved midline keel and there is no evidence for a distinct

ventral surface. Both vertebrae possess a large posterior chevron

facet (and the chevrons associated with Cd3 and 4 are preserved

on the next block in the sequence). No obvious anterior chevron

facets are present due to damage. In all other respects, the centra

of Cd3 and 4 are identical to those of Cdl and 2.

The neural arches are also identical to those of Cdl and 2,

but are better preserved. In lateral view, the neural spines expand

slightly anteroposteriorly towards their apices and are more

strongly posteriorly inclined than in Cdl and 2 (at around 30°

from the horizontal), but this orientation may have been

accentuated by crushing. Both the pre- and postzygapophyses

are small, subtriangular in cross-section, almost vertically

inclined, and short so that neither extends far beyond the margins

of the centrum. The caudal ribs are preserved but deformed: they

appear to have been inclined laterally, dorsally and slightly

posteriorly. All are incomplete distally, but the caudal ribs were

dorsoventrally flattened plates that were anteroposteriorly short

and tapered slightly as they extended laterally. The anterior

margin of the caudal rib is thickened relative to its posterior

margin. At the base of the caudal rib this thickening is more

marked and helps to define a shallow excavation on its

proximoventral surface. This thickened leading edge could be

regarded as an ACDL as it is continuous with the centrum.

Cd5-7 are visible in left lateral view only as they are

embedded in a block, although the anterior and posterior surfaces

of Cd5 and Cd7, respectively, are exposed (fig. 3C). The same

block also includes the chevrons for Cd3-6. As in Cdl-4, the

caudals are deformed due to mediolateral shear and torsion and

their surfaces are crazed, but the shearing is less marked than in

Cdl-4, so more details are visible. The centra are essentially

identical to those in the preceding vertebrae, but are better

preserved and only minor differences are noted here. The visible

articular surfaces of the centra are subelliptical (but deformed)

and shallowly concave. In contrast to earlier caudals, the

neurocentral sutures are visible and appear to be open. A small

anterior and large posterior chevron facet is present in Cd6 and

7. The very narrow ventral surface is separated from the lateral

surfaces by a distinct break in slope and bears a marked

New Valdosaurus

Figure 3. Proximal caudal vertebrae of Valdosaurus canaliculatus (IWCMS 2013.175). A-B, caudosacral plus caudals 1 and 2 in right (A) and

left (B) lateral views. Note that the postacetabular process of the left ilium is attached to caudals 1 and 2. C, caudal vertebrae 3-7 in right lateral

view. D, caudal vertebrae 8-13 in right lateral view. Abbreviations: Cd, caudal vertebra; Ch, chevron; Cs, caudosacral vertebra; l.il, left ilium;

ten, ossified tendons. Scale bars = 50 mm.

Cd13

CdS

P.M. Barrett

Figure 4. Middle caudal vertebrae of Valdosaurus canaliculatus (IWCMS 2013.175). A, caudal vertebrae 14-20 in right lateral view. B-G,

caudal vertebra 21 in right lateral (B), left lateral (C), anterior (D), posterior (E), ventral (F) and dorsal (G) views. Abbreviations: Cd, caudal

vertebra; Ch, chevron. Scale bars = 50 mm (A) and 25 mm (B-G).

longitudinal groove. There are no significant differences in the

morphology of the neural arches between Cdl-4 and Cd5-7 that

cannot be accounted for by deformation or damage.

The next block in the caudal sequence contains Cd8-13

(though only the anterior part of Cdl3, which is poorly preserved,

with the remainder of this vertebra in the next block in the caudal

sequence) (fig. 3D). In terms of overall morphology, these

vertebrae are smaller versions of Cd5-7 and only differences

from the latter are noted here. Cd8-13 are mainly visible in left

lateral view and have suffered crushing so that the caudal ribs

have been rotated upwards to extend parallel to the neural spines.

The left caudal ribs of Cd8 and CdlO are complete, demonstrating

that they terminate in bluntly squared-off distal tips. The fossae

present on the proximoventral surfaces of the caudal ribs of

Cdl-8 become less prominent from Cd9 onwards and are absent

from Cdl2 and all subsequent caudal vertebrae. Ventral grooves

on the centra are present in Cd8-10, but their presence/absence

in Cdll-13 is not determinable due to the orientation in which

these vertebrae are preserved.

A clear break in morphology occurs between Cdl2/13 and

Cdl4/15 and the Cdl3/14 boundary is regarded here as the

transition point from the proximal to the middle caudal series.

Cdl4-20 are preserved in sequence in lateral view within a

single block of matrix (with the posterior part of Cdl3 at its

anterior end), although Cdl4 is not well preserved and Cdl8 is

slightly disarticulated from the other vertebrae, with its anterior

end extending into the block (fig. 4A). Cdl5-20 are much better

preserved than any of the preceding caudals. An isolated phalanx

from the right pes is also present in this block, overlying the

chevron associated with Cdl4.

New Valdosaurus

Figure 5. Middle caudal vertebrae of Valdosaurus canaliculatus (IWCMS 2013.175). A-C, caudal vertebrae 22-25 in right lateral (A), ventral

(B) and dorsal (C) views (caudal 22 is to the right in each case). D-F, caudal vertebrae 26-29 in right lateral (D), ventral (E) and dorsal (F) views

(caudal 26 is to the right in each case). G, detail of caudal vertebrae 27 and 28 in left lateral view showing bundles of ossified tendons. Abbreviation:

Cd, caudal vertebra. Scale bars = 50 mm.

P.M. Barrett

Figure 6. Distal caudal vertebrae of Valdosaurus canaliculatus (IWCMS 2013.175). A-C, caudal vertebrae 30-35 in right lateral (A), ventral (B)

and dorsal (C) views (caudal 30 is to the right in each case). D-F, caudal vertebrae 36-42 in right lateral (D), ventral (E) and dorsal (F) views

(caudal 36 is to the right in each case). G-I, caudal vertebrae 43-45 in right lateral (G), ventral (H) and dorsal (I) views (caudal 43 is to the right

in each case). Abbreviation: Cd, caudal vertebra. Scale bars = 25 mm.

New Valdosaurus

The centra of Cdl4-20 are more elongate and spool-like than

those of the preceding caudals (ratio of centrum length to height

>1.6 in middle caudals whereas in proximal caudals this ratio is

<1.6 and often closerto 1.0). Thelateral surfaces are longitudinally

concave and dorsoventrally convex, but the ventral, anterior and

posterior surfaces are generally obscured and many details are

not visible. Well-defined, but small, posterior chevron facets are

present, but the anterior chevron facets are harder to distinguish.

Cdl9-20 exhibit an incipient longitudinal ridge, which extends

along the full length of the centrum lateral surface, at centrum

midheight.

The neural arches of the middle caudals differ from those of

proximal caudals in several respects. For example, caudal ribs

appear to be completely absent from Cdl4 onwards. Also, the

prezygapophyses of Cdl4-20 are reduced in length and no longer

extend beyond the anterior margins of the centrum. The

postzygapophyses are no longer separate processes, but have

become small facets that are positioned on the base of the neural

spine. In contrast to the preceding vertebrae, the neural spines

become almost parallel-sided in lateral view (whereas there is

some anteroposterior expansion in the distal part of the spine in

proximal caudals), and curve dorsally along their length.

Together with the short prezygapophyses the neural spines form

a unified, curved and scimitar-shaped structure in lateral view.

Cd21 is fully prepared out and separated from the rest of the

caudals (fig. 4B-G); Cd22-25 and Cd26-29 are also fully

prepared, but as two articulated series (fig. 5). Cd21-29 are well

preserved and are either complete or missing only small parts of

the neural spine summit. All of these vertebrae share most of the

features seen in Cdl4-20, but exhibit new anatomical details and/

or reveal morphological trends along the tail. The centra of

Cd27-29 are more elongate than those preceding, with centrum

length/height ratios of >2.0. Where exposed, anterior and

posterior articular surfaces of Cd21-25 are shield-shaped,

shallowly concave and subequal in their mediolateral and

dorsoventral dimensions. However, in Cd26-29 the anterior and

posterior articular surfaces have a more hexagonal outline. This

change is due to stronger development of the longitudinally

extending ridges on the lateral surfaces of the centra in the latter

vertebrae. These lateral ridges, which were incipiently present in

Cdl9-20, exhibit a progressive increase in prominence from

Cd21-29. They divide the lateral surfaces of Cd21-29 into

separate dorsal and ventral excavations. In Cd21 the lateral ridge

is at centrum midheight, but the position of the ridge moves

dorsally in Cd22-29 to a position around one-third of the distance

from the dorsal margin of the centrum. A small, elliptical nutrient

foramen is variably present on the lateral surfaces of the centra in

Cd21-29. In addition, all of these vertebrae also bear faint

longitudinal striations that extend for short distances along the

lateral surfaces from their junctions with the anterior and posterior

articular surfaces. In ventral view, the centra are spool-like and

constricted at midlength. A distinct groove is present along the

ventral midline, bounded by low, sharp ridges. A remnant of the

anterior chevron facet is present in Cd21, represented by a small

bevelled surface, but is absent from Cd22 onward. A distinct

posterior facet is present in Cd21-29, but reduces in size through

the series. The posterior chevron facet is bifurcated ventrally by

the ventral midline groove giving it a ‘W’-shaped outline.

The neural arches of Cd21-29 are almost identical to those

of Cdl4-20 and the neural spines remain very elongate, although

from Cd23 onwards the orientation of the spine changes, to form

an angle of only ~15° with the horizontal (in contrast to ~30° in

the preceding caudals). From Cd22 onward, the postzygapophyses

are reduced to indistinct facets on the neural spine.

Cd30-45 represent the distal caudals (fig. 6). They are all

fully prepared and generally well preserved (although Cd43-45

lack neural spines) and are grouped into several articulated series

(Cd30-35, Cd36-42 and Cd43-45). Although similar to the

middle caudals in many respects, they can be distinguished from

them on the basis of several features. For example, the trend

toward vertebral elongation continues, with centrum length/

height ratios of ~2.7-3.0 in the posterior-most vertebrae (e.g.

Cd42-45). The ventral groove present on the centra of the middle

caudals is present in Cd30-35, but is reduced to a short notch

that bifurcates the ventral margin of the posterior chevron facet

from Cd36-40. Posterior chevron facets and ventral grooves are

both absent from Cd41 onward. Cd36-40 retain the hexagonal

transverse cross-section and particular surface outlines present

in the middle caudal vertebrae, but the lateral ridge extending

along the centrum reduces in prominence and disappears from

Cd41 onward. As a result, the centra of Cd41-45 possess a

simplified, spool-like morphology in which the lateral and

ventral surfaces are smoothly excavated and continuous with

each other.

Perhaps the most obvious difference between the middle and

distal caudal vertebrae is a progressive reduction in the length

and prominence of the neural spine: from Cd30 onward, the

neural spine extends no further posteriorly than the posterior

margin of the successive vertebra (rather than extending beyond

this point, as in the more anterior caudals: see above). In Cd36-42

this trend continues, with the neural spine terminating at a point

only halfway along the next caudal in the series. As the

prezygapophyses and neural spines are present as distinct

processes in Cd43-45, it is likely that several more distal

vertebrae would have been present in vivo, as in other dinosaurs

with complete tails the terminal vertebra lacks neural arch

structures and tapers to a blunt point lacking an articular surface

(Hone, 2012).

Chevrons. With the exceptions of those associated with Cd4 and

Cd8 (which are visible in anterior view), the chevrons articulated

with the proximal caudals (Cd3-13) are visible in lateral view

only (fig. 3D). The chevron for Cdl4 is too poorly preserved to

offer any useful information. Measurements of selected chevrons

are provided in table 3. In anterior view, the chevrons have a ‘Y’-

Table 3. Measurements (in mm) of selected chevrons (Ch) of

Valdosaurus canaliculatus (IWCMS 2013.175). indicates that the

measurement as been affected by breakage.

Chevron length

Ch8

116

Ch9

111

Chl5

Chl6

37*

P.M. Barrett

Figure 7. Partially articulated pelvic girdle of Valdosaurus canaliculatus (IWCMS 2013.175) in right lateral view. Note that majority of the right

femur and adhered left ischium have been removed for clarity. Photograph (A) and interpretative diagram (B). Abbreviations: at, anterior

trochanter; bf, brevis fossa; cs, caudosacral; ds, dorsosacral; l.isc, left ischium; l.pub, left pubis; pap, preacetabular process; pp, pubic peduncle;

r.f, right femur; r.il, right ilium; r.isc, right ischium; r.pub, right pubis; s, sacral vertebra; sr, sacral rib. Scale bar = 100 mm.

New Valdosaurus

Figure 8. Left pelvic girdle and articulated proximal caudal vertebrae of Valdosaurus canaliculatus (IWCMS 2013.175) in left lateral view. Note

that majority of the left femur has been removed for clarity. Photograph (A) and interpretative diagram (B). Abbreviations: at, anterior trochanter;

bs, brevis shelf; Cd, caudal vertebra; cs, caudosacral; ds, dorsosacral; l.f, left femur; l.il, left ilium; l.isc, left ischium; l.pub, left pubis; r.il, right

ilium; r.pub, right pubis; ten, ossified tendons. Scale bar = 100 mm.

P.M. Barrett

shaped morphology, with the proximal branches enclosing a

short, narrow and triangular haemal canal that is closed dorsally.

The haemal canal is recessed within a shallow fossa that extends

for approximately 30% of total chevron length. Ventral to the

haemal canal the chevron shaft extends as a thin vertical strut. In

dorsal view, the articular surface of the chevron has a dumbbell¬

shaped outline that is mediolaterally concave.

In lateral view, the chevrons associated with Cd3-ll are

straight, parallel-sided rods, whose shafts possess no constrictions

or expansions along their length and that terminate in bluntly

rounded ventral margins. They are very elongate, reaching up to

2.0 times the length of their associated centrum where complete

(e.g. those associated with Cd8-9). The chevrons articulated

with Cdl2-13 possess a subtly different morphology in lateral

view, in which the distal-most end of the shaft becomes slightly

expanded anteroposteriorly relative to the rest of the shaft.

Chevrons associated with the middle caudal series (Cdl5-18:

the chevron for Cdl7 is missing) differ in morphology from

those in the proximal part of the tail (fig. 4A). In lateral view,

they are reduced in length relative to the proximal caudals and

are approximately equal in length to their respective centra. The

distal ends of the chevron shafts are anteroposteriorly expanded

relative to their proximal portions: most of this expansion occurs

posteriorly to form an asymmetrical distal flange with a

subtriangular outline. Numerous fine striations are present on

the ventrolateral surfaces of these expanded flanges. No chevrons

are preserved posterior to Cdl8, but the presence of posterior

chevron facets up to Cd40 indicates they were present for most of

the length of the tail.

Ossified tendons. Fragments of numerous ossified tendons are

either adhered to the neural spines of the caudal vertebrae or

found in the matrix adjacent to the tail along its entire length.

These vary between 1.5 and 5 mm in diameter and some have a

circular cross-section, while that of others is flattened (figs. 3A,

5G). Those tendons that are preserved in their natural

orientations (e.g. those preserved alongside Cd26-29) extend

subparallel to each other, and the overlapping trellis-like

arrangement seen in ankylopollexian ornithopods (e.g. Norman,

1980, 1986) was absent.

Ilium. Both ilia are substantially complete and preserved in

articulation with the sacrum and femoral heads, though each is

only visible in lateral view (figs. 7, 8). The right ilium is lacking

the anterior-most part of the preacetabular process, has been

slightly crushed, and its brevis fossa has been artificially enlarged

by plastic deformation (fig. 7). By contrast, the left ilium is well

preserved and undeformed, but has been broken so that its

postacetabular process is associated with Cdl and 2 and is not

currently attached to the sacral block (fig. 8).

The ilium is long and low, with the dorsoventral distance

between the dorsal rim of the acetabulum relatively short in

comparison to overall iliac length (ratio of height above

acetabulum to total length of the left ilium is 0.17). The dorsal

margin of the ilium is slightly sinuous in outline, due to the

downturning of the preacetabular process in combination with a

dorsally convex margin over the pubic peduncle, a slightly

concave margin over the ischiadic peduncle, and a strongly

concave dorsal margin of the postacetabular process. The

preacetabular process extends almost anteriorly in the right

ilium, but anteroventrally in the left ilium: as the latter is less

deformed this is considered to be closer to the original

morphology. The preacetabular process is strap-like in lateral

view, and has parallel dorsal and ventral margins. It is incomplete

anteriorly in both ilia. The ventral margin of the preacetabular

process and the anterior margin of the pubic peduncle are

separated by a deep, concave sulcus.

The pubic peduncle has a thin, elongate subtriangular outline

in lateral view and is anteroposteriorly narrow. It has a

subtriangular transverse cross-section, with the apex of this

triangle pointing laterally and is considerably smaller than the

ischiadic peduncle. The lateral surface of the iliac body dorsal to

the acetabulum was shallowly concave both dorsoventrally and

anteroposteriorly, with this concavity accentuated by deformation

in the right ilium. Neither ilium possesses a distinct

supraacetabular flange. In lateral view, the dorsal margin of the

acetabulum describes a low, gentle curve and is not strongly

arched. The ischiadic peduncle has a stout subtriangular outline

and is pyramidal in shape. Both the pubic and ischiadic peduncles

appear to have extended for the same distance ventrally.

The postacetabular process is elongate and bears an extensive

brevis fossa, which faces ventrally in the left ilium and

ventrolaterally in the deformed right ilium. The fossa is separated

from the lateral surface of the postacetabular process by a robust

anteroventrally extending ridge that merges with the posterior

margin of the ischiadic peduncle. The fossa was not visible in

lateral view in the left ilium, but has been exposed by deformation

in the right ilium. In ventral view, the brevis fossa is very strongly

flared laterally and expands posteriorly to form an equilateral

triangle-shaped flange in ventral view. In lateral view, the dorsal

margin of the iliac body is slightly thickened mediolaterally, but

it is not folded to form an antitrochanter. This area bears

numerous short striations indicative of muscular attachment (for

the M. iliotibialis: e.g. Norman, 1986; Maidment and Barrett,

2011). A shallow groove extends parallel to the dorsal margin of

the ilium along the central part of the iliac body in lateral view.

It is not clear if this groove is a natural feature or due to

deformation, but it is present in both ilia.

A large and well-preserved bundle of ossified tendons is

present in the matrix between the ilia, which extend parallel to

each other and do not form a trellis-like arrangement (fig. 8).

Measurements of both ilia are presented in table 4.

Table 4. Selected measurements (in mm) of the pelvic elements of

Valdosaurus canaliculatus (IWCMS 2013.175). Abbreviations: L,

left; R, right. indicates damaged or deformed.

Ilium

Total ilium length

410 (L) 390* (R)

Height of iliac body above acetabulum

70 (L) 52* (R)

Length of postacetabular process

135 (L) 170* (R)

Maximum transverse width of

postacetabular process

90 (L) 120* (R)

Pubis

Length of prepubic process (as preserved)

195 (L) 220 (R)

New Valdosaurus

Figure 9. Selected hind limb elements of Valdosaurus canaliculatus (IWCMS 2013.175). A, left femur in lateral view (proximal end missing as

attached to pelvic block). B, articulated left tibia, fibula, astragalus and calcaneum in posterior view. C, right calcaneum in lateral view. D, right

metatarsus in anterior view. E, left metatarsus in anterior view. Scale bars = 100 mm (A, B, D, E) and 25 mm (C).

P.M. Barrett

Pubis. The proximal parts of each pubis are present, but it is

unclear how much of the postpubic processes might extend into

the matrix that encases most of the sacral block. The prepubic

processes are elongate, strap-like and extended beyond the

anterior margin of the preacetabular process of the ilium in

lateral view (see table 4 for measurements) (figs. 7, 8). They

have subparallel dorsal and ventral borders and do not expand

distally, but end in a bluntly rounded terminus. The prepubic

processes are mediolaterally-compressed plates, with rounded

dorsal and ventral margins in transverse cross-section.

Prominent depressions that extend along the lateral surface of

the right prepubis and medial surface of the left prepubis are

probably due to crushing. More posteriorly, the prepubes curve

ventrally then dorsally as they merge into the main body of the

pubis, but this may be a preservational artefact.

The pubic body is expanded relative to the plate-like

prepubis, is anteroposteriorly short and block-like and bears a

strongly convex surface for articulation with the ischium and

ilium. The pubic contribution to the acetabulum was relatively

small. An open obturator notch was present, but was partially

enclosed posteriorly by a ventral projection from the posterior

articular surface of the pubic body. The proximal part of the

postpubic rod is cylindrical in cross-section.

Ischium. Both ischia are present, but the proximal ends are

obscured by the overlying femora and by matrix and the shafts

are friable and poorly preserved (figs. 7, 8). The left ischial

shaft is attached to the right femur, whereas the right ischial

Table 5. Selected measurements (in mm) of the hind limb elements of

Valdosaurus canaliculatus (IWCMS 2013.175). Abbreviations: L,

left; R, right. ‘*’ indicates damaged or deformed.

Femoral length

460* (L)

499* (R)

Tibia length

562* (L)

555* (R)

Distal width of tibia

102 (L)

121* (R)

Fibula length

510* (L)

505* (R)

Length of fibula proximal end

88 (L)

- (R)

Fibula midshaft diameter

20 (L)

21 (R)

Mediolateral width of astragalus

83 (L)

90 (R)

Dorsoventral height of calcaneum

-(L)

51 (R)

Mediolateral width of calcaneum

-(L)

24 (R)

Metatarsal II length

199 (L)

200 (R)

Metatarsal II proximal width

27 (L)

26 (R)

Metatarsal II midshaft diameter

14 (L)

12 (R)

Metatarsal II distal width

29 (L)

29 (R)

Metatarsal III length

239 (L)

247 (R)

Metatarsal III proximal width

-(L)

46* (R)

Metatarsal III midshaft diameter

34 (L)

43* (R)

Metatarsal III distal width

52 (L)

52 (R)

Metatarsal IV length

195 (L)

221* (R)

Metatarsal IV proximal width

57* (L)

29 (R)

Metatarsal IV midshaft diameter

35 (L)

35 (R)

Metatarsal IV distal width

29 (L)

28 (R)

shaft is free. The right ischium has a fan-shaped proximal

plate, whose iliac articulation is visible in lateral and dorsal

views. The articular surface is oval in outline and mediolaterally

expanded relative to the rest of the exposed part of the proximal

plate. The shaft arises from the posteroventral corner of the

proximal plate. There is no indication of a groove along the

dorsal margin of the proximal part of the shaft, which is smooth

and rounded. The presence/absence of an obturator process

cannot be determined due to breakage. In lateral view, the

ischial shafts are straight, extend posteroventrally from the

proximal plate and have parallel dorsal and ventral margins.

There is no sign of any ventral curvature, although both ischia

are incomplete distally so its absence might be artefactual. The

shafts are mediolaterally compressed. It is not possible to

determine the extent of any symphysis between them due to

breakage and poor surface preservation.

Hind limb. Although they are almost complete, the hind limbs of

IWCMS 2013.175 provide no new anatomical information. Some

elements, such as the femora, tibiae and fibulae are crushed,

deformed and possess poor surface preservation, although the

proximal tarsals, metatarsi and pedal elements are well preserved

(fig. 9). As far as can be determined, all of these elements are

identical to those of other recently described specimens of

Valdosaurus (see Barrett et al., 2011). Measurements for all hind

limb elements are presented in table 5.

Both femora are complete, but each is preserved in two

parts, with the proximal part of the bone in articulation with

each acetabulum (figs. 7, 8) and the distal part prepared

separately (fig. 9A). The right femur is adhered to the left ischial

shaft. Valdosaurus possesses several diagnostic femoral

characteristics: 1) a deep cleft between the anterior and greater

trochanters that is visible in both medial and anterior views; 2)

a scar for the M. caudofemoralis that is placed close to the base

of the fourth trochanter; 3) a ‘U’-shaped anterior intercondylar

groove that is deeply incised into the femur with near parallel

sides; and 4) the proximal end of the anterior trochanter of the

femur is level with, or only slightly ventral to, the proximal end

of the greater trochanter (Barrett et al., 2011). Unfortunately, it

is not possible to assess the presence or absence of characters 1)

and 2) in IWCMS 2013.175 due to damage, but characters 3)

and 4) are present, supporting referral to Valdosaurus

canaliculatus (see above). It appears that the anterior

intercondylar groove was partially closed by a lip of bone

arising from the medial condyle, as occurs in other large

individuals (Galton, 2009; Barrett et al., 2011).

The tibiae are also complete, but preserved in several

sections. The left tibia is preserved in two parts (fig. 9B): a

proximal part comprising the proximal expansion and two-

thirds of the shaft and a distal part consisting of the rest of the

shaft, which articulated with the distal part of the left fibula,

the left astragalus and a fragment of the left calcaneum. The

right tibia is also preserved in two sections: the small proximal

section is in articulation with the proximal region of the right

fibula, while the larger distal section, comprising around three-

quarters of the length of the bone, is articulated with the distal

fibula shaft and the right astragalus. The proximal ends of both

tibiae are very poorly preserved: that of the right tibia is

New Valdosaurus

flattened, while that of the left tibia has experienced both

crushing and strong torsion, although all three of the major

proximal processes (cnemial crest, inner and outer condyles)

can be identified. Torsion has caused the cnemial crest of the

left tibia to extend laterally rather than anteriorly.

Both fibulae are almost complete and each is preserved in

two sections with small sections of the shaft missing. The

proximal end of the left fibula is generally well preserved, but

has been crushed mediolaterally, whereas the distal part of the

shaft is extensively cracked and warped. By contrast, the

proximal end of the right fibula is poorly preserved and

extensively crushed, but the shaft is more three-dimensional. A

small contact was present between the distal fibula and

the astragalus.

In general, the proximal tarsals are well preserved and

undeformed. Both astragali are present and in articulation with

their respective tibiae. The left calcaneum is also in articulation

with the left tibia, but is broken and incomplete, whereas the

complete right calcaneum has become separated from the rest

of the right ankle (fig. 9C). The only differences between these

elements and those previously described for Valdosaurus are

that the medial surfaces of the calcanea, which form the

articulation for the astragalus, are strongly rugose and that the

articular surfaces for the distal tarsals for both the astragali and

calcanea are slightly corrugated, in contrast to the smooth

surfaces present in other specimens referred to this taxon

(Barrett et al., 2011).

Both metatarsi, comprising metatarsals (Mt) 2-4, are

preserved in articulation and each metatarsal is complete (fig.

9D, E). The left metatarsus has suffered some minor

deformation, but has good surface preservation; the right is

slightly crushed, but also in good condition. A distal tarsal is

articulated with the proximal surface of the left metatarsus and

positioned primarily over Mt3, but also partially overlaps the

proximal surfaces of Mt2 and Mt4. It has an elliptical outline

and concave proximal surface. Mt2 and Mt3 are closely

appressed along their entire lengths, whereas Mt4 is kinked

laterally at a point about halfway along its length so its distal

end diverges slightly from that of Mt3 (both conditions are

present in both metatarsi). There is no evidence for the presence

of a first or fifth metatarsal in either foot.

Several non-ungual phalanges are preserved, though they

were not found in articulation with the metatarsals. Several are

very well preserved and are uncracked and undistorted: it is

possible that these are associated with the similarly well-

preserved left metatarsus. Conversely, the other more heavily

cracked and crushed phalanges might pertain to the slightly

crushed right metatarsus. Preserved phalanges (Ph) include

both Phll.l and one PhIII.1. Several other phalanges are also

present, but their positions within the pes cannot be determined.

Discussion

The anatomy of dryosaurids is poorly known: all known taxa

are represented by either isolated femora ( Callovosaurus,

Elrhazosaurus), partial skeletons ( Dryosaurus,

Eousdryosaurus, Kangnasaurus, Valdosaurus ), or bonebed

material that is disarticulated and lacks some key skeletal

elements ( Dysalotosaurus ) (see Janensch, 1955; Shepherd et

al., 1977; Galton, 1975, 1981, 1983, 1989, 2009; Galton and

Taquet, 1982; Cooper, 1985; Blows, 1998; Ruiz-Omenaca et al.,

2007; Hiibner and Rauhut, 2010; Barrett et al., 2011; Escaso et

al., 2014). Thus, the new specimen reported herein not only

increases the amount of anatomical information available for

Valdosaurus , but also provides some information that might be

more generally applicable for the clade as a whole.

IWCMS 2013.175 confirms the validity of a previously

proposed autapomorphy for Valdosaurus : the presence of an

open obturator notch on the pubis (Barrett et al., 2011). Other

dryosaurids for which the pubis is known (. Dryosaurus,

Dysalotosaurus ) have a closed notch (Janensch, 1955; Galton,

1981). One new feature of the tail is proposed herein as a potential

autapomorphy of Valdosaurus : the presence of elongate neural

spines in the middle of the caudal series that extend to reach over

more than one subsequent vertebra (fig. 5D). This condition

appears to be absent in Kangnasaurus (although on the basis of

fragmentary material: Cooper, 1985) and is absent in non-

iguanodontian ornithopods like Hypsilophodon (NHMUK

OR28707; NHMUK PV R196), Tenontosaurus (Forster, 1990),

rhabdodontids (Weishampel et al., 2003) and ankylopollexians

(Norman, 1986). It is difficult to assess this character in

Dryosaurus due to damage, but the orientation of the preserved

bases of the neural spines suggests that they might have been

more vertically inclined and thus might not have extended

beyond the posterior margin of the succeeding vertebra (Galton,

1981). Unfortunately this feature cannot be assessed in

Callovosaurus, Dysalotosaurus, Elrhazosaurus or

Eousdryosaurus due to incomplete preservation (Janensch,

1955; Galton and Taquet, 1982; Ruiz-Omenaca et al., 2007;

Escaso et al., 2014).

IWCMS 2013.175 enables comparisons with other taxa

that were not previously possible. With the exception of the

potentially autapomorphic middle caudal neural spines, the

preserved axial column of IWCMS 2013.175 is generally

similar to that of Dryosaurus, Dysalotosaurus,

Eousdryosaurus and Kangnasaurus (Janensch, 1955; Galton,

1981; Cooper, 1985; Escaso et al., 2014). The neural arches of

the proximal caudal vertebrae in Eousdryosaurus possess

small, anteriorly projecting, “thorn-like” processes that have

been proposed as diagnostic for this taxon (Escaso et al.,

2014: 1104): these processes are absent in Valdosaurus. The

distal ends of the middle caudal chevrons of Valdosaurus are

anteroposteriorly expanded in lateral view (fig. 4A): although

comparative material is lacking for other dryosaurids, similar

chevrons are known in Gasparinisaura (Coria and Salgado,

1996; MUCPv-212), Leaellynasaura (NMV P185992, NMV

P186047) and Tenontosaurus (Forster, 1990). Fortuitously,

IWCMS 2013.175 possesses the most complete tail of any

dryosaurid specimen described to date and demonstrates that

Valdosaurus is likely to have possessed no more than 46-50

caudal vertebrae (fig. IB). This number is similar to that seen

in many other ornithopods (Norman, 2004; Norman et al.,

2004), but much lower than those recorded for the

exceptionally long tails of Leaellynasaura (>70 caudal

vertebrae; Herne, 2009) and Tenontosaurus (60-65 caudal

vertebrae: Forster, 1990).

P.M. Barrett

The overall shape of the ilium in Valdosaurus (IWCMS

2013.175) falls within the range of variation seen in Dryosaurus

and Dysalotosaurus (Janensch, 1955; Galton, 1981). In

Eousdryosaurus the brevis fossa is exposed in lateral view

(Escaso et al., 2014), but that of Valdosaurus is largely obscured

and faces posteroventrally (fig. 8: the right ilium of IWCMS

2013.175 has a laterally open brevis fossa, but this has resulted

from plastic deformation: see above). The ilia of Eousdryosaurus,

some Dryosaurus individuals and of an indeterminate dryosaurid

from the Tunbridge Wells Sands Formation of Cuckfield, UK

(NHMUK OR2132) possess a straight dorsal margin in lateral

view (Escaso et al., 2014). By contrast, those of Dysalotosaurus,

Valdosaurus and two other indeterminate dryosaurid specimens

from the Tunbridge Wells Sands Formation (BMB 004274;

NHMUK OR2150) are gently sinuous (Janensch, 1955; Galton,

2009; see above). In Valdosaurus, BMB 00472 and NHMUK

OR2150 the postacetabular processes are posterodorsally

inclined (BMB 004274; NHMUK OR2150; Galton, 2009),

whereas in NHMUK OR2131, this process extends posteriorly.

Escaso et al. (2014) proposed that the calcaneum of

Eousdryosaurus could be distinguished from those of other

dryosaurids on the basis of the rounded (rather than more sharply

triangular) ‘proximal projection’, which comprises the triangular

process that divides the facets for the tibia and fibula in lateral

view. However, this process is similarly rounded in IWCMS

2013.175 (fig. 9C) and hence, this feature cannot be regarded as a

reliable autapomorphy for Eousdryosaurus. Moreover, Escaso et

al. (2014) also proposed that the calcaneum of Valdosaurus

differed from that of other dryosaurids in possessing a distinct

offset between the facets for the tibia and fibula in lateral view

(with the tibia facet positioned more ventrally), whereas in other

dryosaurids the facets were at the same level. However, this is

not the case: if the calcanea are figured in their ‘natural’

orientations (i.e. as if in articulation with the distal tibia) the

fibula facets in all of these taxa (including Eousdryosaurus )

would be positioned somewhat dorsally relative to the level of the

tibia facets, reflecting the shorter length of the fibula relative to

the tibia in dryosaurids (Galton, 1981; Barrett et al., 2011; Escaso

et al., 2014).

Finally, IWCMS 2013.175 enables the scoring of several

phylogenetic characters for Valdosaurus that were unknown

from other specimens and scored as missing data in the analyses

of McDonald et al. (2010), Barrett et al. (2011) and Escaso et al.

(2014). For example, it reveals the longitudinal (rather than

basket-like) arrangement of ossified tendons along the vertebral

column (McDonald et al., 2010: character 95[0]). Several iliac

characters can also be scored: the preacetabular process of the

ilium was not twisted along its length (McDonald et al., 2010:

character 110[0]); the dorsal margin of the main iliac body is

only slightly sinuous, rather than displaying the extreme sinuosity

seen in more derived iguanodontians (McDonald et al., 2010:

character 111[0]); and the dorsal margin of the iliac body dorsal

to the ischial peduncle is smooth and lacks the development of

any prominent processes (McDonald et al., 2010: character

112[0]). All of these character scores are identical to those

present in Dryosaurus and Dysalotosaurus (McDonald et al.,

2010; Barrett et al., 2011).

Conclusions

IWCMS 2013.175 is the most complete individual of

Valdosaurus canaliculatus yet found and offers new

information on the axial skeleton and pelvis of this poorly

known iguanodontian ornithopod. Its discovery highlights the

fact that Valdosaurus was a more common component of the

Wessex Formation dinosaur fauna than usually thought.

Frustratingly, all known individuals are represented by their

hindquarters only, limiting comparisons with other dryosaurids.

Additional discoveries of specimens that include both hind

limb material and elements from the presacral region are now

critical to enable further integration of Valdosaurus into both

phylogenetic and palaeoecological scenarios.

Acknowledgements

This paper is dedicated to Thomas Rich in recognition of his

work on the dinosaurs of the Australian Early Cretaceous,

which were roaming Victoria around the time that Valdosaurus

lived in the Northern Hemisphere. Tom has been generous with

his time and material and has done much to illuminate the

previously poorly known Cretaceous ecosystems in this region.

Numerous individuals and organisations contributed to the

collection and preparation of the specimen, and I would like to

express my sincere thanks to the following: the National Trust

(owners of the locality) for donating the specimen to Dinosaur

Isle; Nick Chase for his discovery of the specimen, leading role

in its excavation and initial preparation; Steve Hutt, Penny

Newbury and Jeremy Lockwood for their involvement in the

excavation; and Gary Blackwell and Martin New for preparation

of the specimen. I am extremely grateful to Alex Peaker for his

continued assistance and hospitality at Dinosaur Isle, as well

the provision of key information on the specimen. Martin Munt

is thanked for bringing the specimen to my attention and for

logistical support while working on the Isle of Wight. In

addition to those already listed, a number of curators provided

access to material in their care, including S. Hutt, J. Porfiri, M.

Simms and T. Rich. Photographs were taken by Kevin Webb

(NHM Image Resources) and the NHM Earth Sciences

Departmental Investment Fund provided funding for travel. I

thank referees Susannah Maidment and Andrew McDonald

and editor Erich Fitzgerald for their helpful comments and the

editor for the invitation to contribute to this volume.

References

Allen, P., and Wimbledon, W.A. 1991. Correlation of NW European

Purbeck-Wealden (nonmarine Lower Cretaceous) as seen from the

English type-areas. Cretaceous Research 12: 511-526.

Barrett, P.M., Butler, R.J., Twitchett, R.J., and Hutt, S. 2011. New

material of Valdosaurus canaliculatus (Ornithischia: Ornithopoda)

from the Lower Cretaceous of southern England. Special Papers in

Palaeontology 86: 131-163.

Beneden, P.-J. van 1881. Sur Pare pelvien chez les dinosauriens de

Bernissart. Bulletins de l’Academie Royale des Sciences, des

Lettres et des Beaux-Arts de Belgique, Classe des Sciences (Series

3) 1: 600-608.

New Valdosaurus

Blows, W.T. 1998. A review of Lower and middle Cretaceous dinosaurs

of England. New Mexico Museum of Natural History and Science,

Bulletin 14: 29-38.

Carpenter, K., and Ishida, Y. 2010. Early and “middle” Cretaceous

iguanodonts in time and space. Journal of Iberian Geology 36:

145-164.

Cooper, M.R. 1985. A revision of the ornithischian dinosaur

Kangnasaurus coetzeei Haughton, with a classification of

Ornithischia. Annals of the South African Museum 95: 281- 317.

Coria, R.A., and Salgado, L. 1996. A basal iguanodontian (Ornithischia:

Ornithopoda) from the Late Cretaceous of South America. Journal

of Vertebrate Paleontology 16: 445-457.

Escaso, F., Ortega, F., Dantas, P., Malafaia, E., Silva, B., Gasulla, J.M.,

Mocho, P., Narvaez, I., and Sanz, J.L. 2014. A new dryosaurid

ornithopod (Dinosauria, Ornithischia) from the Late Jurassic of

Portugal. Journal of Vertebrate Paleontology 34: 1102-1112.

Forster, C.A. 1990. The postcranial skeleton of the ornithopod dinosaur

Tenontosaurus tilletti. Journal of Vertebrate Paleontology 10:

273-294.

Galton, P.M. 1974. The ornithischian dinosaur Hypsilophodon from the

Wealden of the Isle of Wight. Bulletin of the British Museum

(Natural History), Geology 25: 1-152.

Galton, P.M. 1975. English hypsilophodontid dinosaurs (Reptilia:

Ornithischia). Palaeontology 18: 741-752.

Galton, P.M. 1981. Dryosaurus, a hypsilophodontid dinosaur from the

Upper Jurassic of North America and Africa. Postcranial skeleton.

Palaontologische Zeitschrift 55: 271-312.

Galton, P.M. 1983. The cranial anatomy of Dryosaurus, a

hypsilophodontid dinosaur from the Upper Jurassic of North

America and East Africa, with a review of hypsilophodontids from

the Upper Jurassic of North America. Geologica et Palaeontologica

17: 207-243.

Galton, P.M. 1989. Crania and endocranial casts from ornithopod

dinosaurs of the families Dryosauridae and Hypsilophodontidae

(Reptilia: Ornithischia). Geologica et Palaeontologica 23: 217-239.

Galton, P.M. 2009. Notes on Neocomian (Lower Cretaceous) ornithopod

dinosaurs from England - Hypsilophodon, Valdosaurus,

“ Camptosaurus ”, “Iguanodon ” - and referred specimens from

Romania and elsewhere. Revue de Paleobiologie 28: 211-273.

Galton, P.M. and Taquet, P. 1982. Valdosaurus, a hypsilophodontid

dinosaur from the Lower Cretaceous of Europe and Africa. Geobios

15: 147-159.

Herne, M. 2009. Postcranial osteology of Leaellynasaura

amicagraphica (Dinosauria: Ornithischia) from the Early

Cretaceous of Southeastern Australia. Journal of Vertebrate

Paleontology 29 (3, Supplement): 113A.

Hone, D.W.E. 2012. Variation in the tail length of non-avian dinosaurs.

Journal of Vertebrate Paleontology 32: 1082-1089.

Hooley, R.W. 1925. On the skeleton of Iguanodon atherfieldensis sp.

nov., from the Wealden Shales of Atherfield (Isle of Wight).

Quarterly Journal of the Geological Society of London 81: 1-61 +

pis 1-2.

Hiibner, T.R., and Rauhut, O.W.M. 2010. A juvenile skull of

Dysalotosaurus lettowvorbecki (Ornithischia: Iguanodontia), and

implications for cranial ontogeny, phylogeny, and taxonomy in

ornithopod dinosaurs. Zoological Journal of the Linnean Society

160: 366-396.

Hulke, J.W. 1879. Vectisaurus valdensis, a new Wealden dinosaur.

Quarterly Journal of the Geological Society of London 35: 421-

424 + pi. 21.

Hulke, J.W. 1882. Description of some Iguanodon remains indicating a

new species, I. seeleyi. Quarterly Journal of the Geological Society

of London 38: 135-144 + pi. 4.

Huxley, T.H. 1869. On Hypsilophodon, a new genus of Dinosauria.

Abstracts of the Proceedings of the Geological Society of London

204:3-4.

Janensch, W. 1955. Der Ornithopode Dysalotosaurus der

Tendaguruschichten. Palaeontographica, Supplement 7: 105-176 +

pis 9-14.

Lydekker, R. 1888. Catalogue of the fossil Reptilia and Amphibia in the

British Museum. Part I. Containing the orders Ornithosauria,

Crocodilia, Dinosauria, Squamata, Rhynchocephalia and

Proterosauria. British Museum of Natural History: London, xii +

309 pp.

Maidment, S.C.R., and Barrett, P.M. 2011. The locomotor musculature

of basal ornithischian dinosaurs. Journal of Vertebrate

Paleontology 31: 1265-1291.

Martill, D.M., and Naish, D. 2001. The geology of the Isle of Wight. Pp.

25-43 in: Martill, D.M. and Naish, D. (eds). Dinosaurs of the Isle of

Wight (Palaeontological Association Field Guides to Fossils, 10).

Palaeontological Association: London. 433 pp.

McDonald, A.T., Barrett, P.M., and Chapman, S.D. 2010. A new basal

iguanodont (Dinosauria: Ornithischia) from the Wealden of

England. Zootaxa 2569: 1-43.

McDonald, A.T. 2012. The status of Dollodon and other basal

iguanodonts (Dinosauria: Ornithischia) from the Lower Cretaceous

of Europe. Cretaceous Research 33: 1-6.

Naish, D., and Martill, D.M. 2001. Ornithopod dinosaurs. Pp. 60-132

in: Martill, D.M. and Naish, D. (eds). Dinosaurs of the Isle of Wight

(Palaeontological Association Field Guides to Fossils, 10).

Palaeontological Association: London. 433 pp.

Norman, D.B. 1980. On the ornithischian dinosaur Iguanodon

bernissartensis of Bernissart (Belgium). Memoire de Vlnstitut

Royal des Sciences Naturelles de Belgique 178: 1-104.

Norman, D.B. 1986. On the anatomy of Iguanodon atherfieldensis

(Ornithischia: Ornithopoda). Bulletin de Vlnstitut Royal des

Sciences Naturelles de Belgique 56: 281-372.

Norman, D.B. 2004. Basal Iguanodontia. Pp. 413-437 in: Weishampel,

D.B., Dodson, P., and Osmolska, H. (eds). The Dinosauria (Second

Edition). University of California Press: Berkeley. 861 pp.

Norman, D.B. 2012. Iguanodontian taxa (Dinosauria: Ornithischia)

from the Lower Cretaceous of England and Belgium. Pp. 174-212

in: Godefroit, P. (ed.), Bernissart dinosaurs and Early Cretaceous

terrestrial ecosystems. Indiana University Press: Bloomington and

Indianapolis. 629 pp.

Norman, D.B., Sues, H.-D., Witmer, L.M., and Coria, R.A. 2004. Basal

Ornithopoda. Pp. 393-412 in: Weishampel, D.B., Dodson, P., and

Osmolska, H. (eds), The Dinosauria (Second Edition). University of

California Press: Berkeley. 861 pp.

Paul, G.S. 2008. A revised taxonomy of the iguanodont dinosaur genera

and species. Cretaceous Research 29: 192-216.

Rawson, PF. 2006. Cretaceous: sea levels peak as the North Atlantic

opens. Pp. 365-393 in: Brenchly, P.J., and Rawson, PF. (eds). The

geology of England and Wales (Second Edition). The Geological

Society: London. 559 pp.

Ruiz-Omenaca, J.L, Pereda Suberbiola, X., and Galton, P.M. 2007.

Callovosaurus leedsi, the earliest dryosaurid dinosaur

(Ornithischia: Euornithopoda) from the Middle Jurassic of England.

Pp. 3-16 in: Carpenter, K. (ed.), Homs and beaks: ceratopsian and

ornithopod dinosaurs. Indiana University Press: Bloomington and

Indianapolis, xi + 369 pp.

Shepherd, J.D., Galton, P.M., and Jensen, J.A. 1977. Additional

specimens of the hypsilophodontid dinosaur Dryosaurus altus

from the Upper Jurassic of western North America. Brigham Young

University Geology Studies 24: 11-15.

P.M. Barrett

Sweetman, S.C., and Insole, A.N. 2010. The plant debris beds of the

Early Cretaceous (Barremian) Wessex Formation of the Isle of

Wight, southern England: their genesis and palaeontological

significance. Palaeogeography, Palaeoclimatology, Palaeoecology

292:409-424.

Weishampel, D.B., Jianu, C.-M., Csiki, Z., and Norman, D.B. 2003.

Osteology and phylogeny of Zalmoxes (n. g.), an unusual

euomithopod dinosaur from the latest Cretaceous of Romania.

Journal of Systematic Palaeontology 1: 65-123.

Wilson, J.A. 1999. A nomenclature for vertebral laminae in sauropods

and other saurischian dinosaurs. Journal ofVertebrate Paleontology

19:639-653.

Wilson, J.A., D'Emic, M.D., Ikejiri, T., Moacdieh, E.M., and Whitlock,

J.A. 2011. A nomenclature for vertebral fossae in sauropods and

other saurischian dinosaurs. PLoS ONE 6(2): el7114 (doi:10.1371/

j ournal .p one. 0017114).

Memoirs of Museum Victoria 74:49-61 (2016) Published 2016

ISSN 1447-2546 (Print) 1447-2554 (On-line)

http://museumvictoria.com.au/about/books-and-journals/journals/memoirs-of-museum-victoria/

Phylogenetic relationships of the Cretaceous Gondwanan theropods Megaraptor

and Australovenator: the evidence afforded by their manual anatomy

Fernando E. Novas 1 ’ 2 *, Alexis M. Aranciaga Rolando 1 and Federico L. Agnolin 13

1 Museo Argentino de Ciencias Naturales, Avenida Angel Gallardo 470,1405DJR Buenos Aires, Argentina

2 CONICET, Consejo Nacional de Investigaciones Cientfficas y Tecnicas, Argentina

3 Fundacion de Historia Natural “Felix de Azara”, Universidad Maimonides, Hidalgo 775,1405BDB, Buenos Aires,

Argentina

* To whom correspondence should be addressed. E-mail: fernovas@yahoo.com.ar

Abstract Novas, F.E., Aranciaga Rolando, A.M. and Agnolm, F.L. 2016. Phylogenetic relationships of the Cretaceous Gondwanan

theropods Megaraptor and Australovenator: the evidence afforded by their manual anatomy. Memoirs of Museum

Victoria 74: 49-61.

General comparisons of the manual elements of megaraptorid theropods are conducted with the aim to enlarge the

morphological dataset of phylogenetically useful features within Tetanurae. Distinctive features of Megaraptor are

concentrated along the medial side of the manus, with metacarpal I and its corresponding digit being considerably

elongated. Manual ungual of digit I is characteristically enlarged in megaraptorids, but it is also transversely compressed

resulting in a sharp ventral edge. We recognize two derived characters shared by megaraptorans and coelurosaurs (i.e.,

proximal end of metacarpal I without a deep and wide groove continuous with the semilunar carpal, and metacarpals I and

II long and slender), and one derived trait similar to derived tyrannosauroids (i.e., metacarpal III length <0.75 length of

metacarpal II). However, after comparing carpal, metacarpal and phalangeal morphologies, it becomes evident that

megaraptorids retained most of the manual features present in Allosaurus. Moreover, Megaraptor and Australovenator

are devoid of several manual features that the basal tyrannosauroid Guanlong shares with more derived coelurosaurs (e.g.,

Deinonychus), thus countering our own previous hypothesis that Megaraptora is well nested within Tyrannosauroidea.

Keywords Dinosauria, Theropoda, Megaraptoridae, Cretaceous, Argentina, Australia, morphology.

Introduction

Megaraptoridae is a Cretaceous theropod family including

several taxa recorded from different regions of Gondwana

(Novas et al., 2013). The best known megaraptorids are

Megaraptor namunhuaiquii (Novas, 1998; Calvo et al., 2004;

Porfiri et al., 2014), Orkoraptor bukei (Novas et al., 2008),

and Aerosteon riocoloradensis (Sereno et al., 2008), coming

from different formations of Turonian through Santonian age

of Argentina; and Australovenator wintonensis (Hocknull et

al., 2009; White et al., 2012, 2013), from Cenomanian rocks

of Australia.

The megaraptorids and their sister taxon Fukuiraptor

kitadaniensis (Azuma and Currie, 2000), from Barremian beds

of Japan, constitutes the clade Megaraptora, originally coined

by Benson et al. (2010a). After a comprehensive phylogenetic

analysis, these authors considered megaraptorans as

allosauroids closely related with carcharodontosaurid

theropods, an interpretation subsequently followed by later

authors (Carrano et al., 2012; Zanno and Makovicky, 2013).

However, recent studies conducted by some of us (e.g., Novas et

al., 2013; Porfiri et al., 2014) have suggested that megaraptorans

are not representative of archaic allosauroid tetanurans, but

instead argued that megaraptorans are coelurosaurs, and

representatives of a basal tyrannosauroid radiation in particular

(Novas et al., 2013). Recent discovery of cranial remains of a

juvenile specimen of Megaraptor namunhuaiquii (Porfiri et

al., 2014) offered novel anatomical information that supported

this phylogenetic interpretation.

The fossil record of megaraptorids in Gondwana has

increased over the last few years. Additional evidence of the

presence of megaraptorids in regions of South America other

than Argentina comes from Brazil, from which isolated caudal

vertebrae have been described (Mendez et al., 2013).

Cretaceous formations of Australia have yielded several

isolated elements referred to Megaraptoridae, including

Rapator ornitholestoides (Huene, 1932; Agnolm et al., 2010;

White et al., 2012), an isolated ulna closely similar to that of

Megaraptor and Australovenator (Smith et al., 2008), more

than one hundred isolated teeth (Benson et al., 2012), and

probably an isolated astragalus (Molnar et al., 1981; Fitzgerald

et al., 2012), and paired pubes originally described as

tyrannosauroid (Benson et al., 2010b; Novas et al., 2013).

F.E. Novas, A.M. Aranciaga Rolando & F.L. Agnolin

Available information demonstrates that megaraptorans

were a diverse and relatively abundant clade of large predatory

dinosaurs in the southern landmasses (Novas, 1998, 2008;

Calvo et al., 2004; Benson et al., 2010a; Novas et ah, 2013),

sharing with abelisauroids and carcharodontosaurids the role

of top predators.

We offer here a comparative survey of the manual bones of

Megaraptor and Australovenator with the aim to recognize

anatomical features characterizing these theropods. Also, we

briefly discuss the distribution of some manual features among

theropods that may inform the phylogenetic relationships of

megaraptorid among Tetanurae.

Institutional abbreviations

AODF, Australian Age of Dinosaurs Fossil, Winton, Australia;

BMNH, British Museum of Natural History, London, England;

IVPP, Institute of Vertebrate Paleontology and

Paleoanthropology, Beijing, China; MUCPv, Museo de la

Universidad Nacional del Comahue, Neuquen, Argentina;

UUVP, University of Utah Vertebrate Paleontology, Utah,

USA; YPM, Yale Peabody Museum, New Haven, USA.

Materials and Methods

Material examined. A comparative study of the holotype and

referred specimens of Megaraptor namunhuaiquii (MUCPv

595, MUCPv 1353, and MUCPv 341), Australovenator

wintonensis (AODF 604), and cast of Rapator ornitholestoides

(cast of BMNH R3718) was conducted. The following

specimens were also studied: Guanlong wucaii (IVPP V14531),

Allosaurus fragilis (cast of UUVP 6000), Deinonychus

antirrhopus (cast of YPM 5205), Xuanhanosaurus qilixiaensis

(cast of IVPP V6729), Coelurus fragilis (cast of YPM 2010),

and Ornitholestes hermanni (cast of AMNH 619).

Comparative Anatomy

Megaraptor and Australovenator are currently the only

megaraptorans in which the forelimb bones are fairly well

documented (Calvo et al., 2004; Hocknull et al., 2009; White

et al., 2012). Specimen MUCPv 341 of Megaraptor

namunhuaiquii preserves articulated forearm bones (i.e., ulna

and radius) and manus, but no humerus (fig. 3). However, the

recent discovery of a juvenile specimen of M. namunhuaiquii

(Porfiri et al., 2014) documents for the first time the humeral

morphology in this genus. Although the humerus does not

preserve complete proximal and distal ends, it offers reliable

information to calculate humeral proportions in this

Patagonian taxon. The type specimen of Australovenator

preserves most of the forelimb except metacarpal III and some

manual phalanges.

Humerus. The humerus of Megaraptor (Porfiri et al., 2014) and

Australovenator (White et al., 2012) resembles basal tetanurans

(e.g., Allosaurus, Acrocanthosaurus, Piatnitzkysaurus; Madsen,

1976; Currie and Carpenter, 2000; Bonaparte, 1986) and basal

coelurosaurs (e.g., Coelurus, Ornitholestes, Guanlong; Osborn,

1903; Carpenter, 2005; Xu et al., 2006; fig. 1) in being sigmoid¬

shaped in anterior and lateral views, with a prominent

deltopectoral crest. These characters are absent in non-

coelurosaurian theropods like Xuanhanosaurus (Dong, 1984),

Ceratosaurus (Madsen and Welles, 2000), Torvosaurus (Galton

and Jensen, 1979), Baryonyx (Charig and Milner, 1997), and

some coelurosaurs including ornithomimids (Kobayashi and Lti,

2003; Nichols and Russel, 1985), and tyrannosaurids (Brochu,

2002) (fig. 1). The internal tuberosity also resembles basal

tetanurans in being conical-shaped (e.g., Bonaparte et al., 1990).

However, the humerus of both Megaraptor and Australovenator

exhibits a deep longitudinal furrow that runs on the medial

surface of the shaft, distally to the internal tuberosity, a feature

also present in Fukuiraptor and some coelurosaurs ( Deinonychus,

tyrannosaurids; Ostrom, 1969; Brochu, 2002). This character is

absent in other coelurosaurs like Chilantaisaurus, Ornitholestes,

Coelurus, oviraptorosaurs (Benson and Xu, 2008; Osborn, 1903;

Carpenter, 2005; Lu, 2002), and non-coelurosaurian tetanurans

(e.g. Allosaurus, Acrocanthosaurus, Piatnizkysaurus; Madsen,

1976; Currie and Carpenter, 2000; Bonaparte 1986) (fig. 1).

Furthermore the entire distal end bends anteriorly, showing a

sigmoid shape in lateral view. Notably, the distal humeral

condyles of Australovenator (White et al., 2012) are well-defined

and much more rounded anteriorly than those of Allosaurus,

Acrocanthosaurus or Xuanhanosaurus (Madsen 1976; Currie

and Carpenter, 2000; Dong 1984) (fig. 2), and are separated by

deep extensor and flexor grooves not present in non-celurosaurian

tetanurans. In this regards, the distal end of the humerus of

Australovenator (White et al., 2012) resembles coelurosaurs like

Coelurus, Ornitholestes (Carpenter, 2005), Guanlong (IVPP

IVPP V14531), Deinonychus (Novas, 1996) (fig. 2) and Aves, and

may suggest a more complex folding system than in basal

theropods, a hypothesis that needs to be tested properly. Apart

from the similarity with some coelurosaurs described for the

distal end, the robust construction of the humerus in Megaraptor

and Australovenator is closer to Allosaurus (width:length ratio

approximately 40; Madsen, 1976; Hocknull et al., 2009; Porfiri

et al., 2014) than the elongate and more gracile humeral

proportions of Guanlong and Deinonychus (widthdength ratio

approximately 30; pers. obs.).

Ulna. As already noted by previous authors (e.g.. Novas, 1998;

Calvo et al., 2004; Smith et al., 2008; Agnolin et al., 2010;

Benson et al., 2010a; Hocknull et al., 2009; White et al., 2012;

Novas et al., 2013), the megaraptorid ulna exhibits a

transversally compressed blade-like olecranon process, and a

robust and dorsoventrally extended lateral tuberosity. These

two features are absent in the remaining theropods, including

the basal megaraptoran Fukuiraptor, thus they have been

interpreted as unambiguous synapomorphies of Megaraptoridae

(Novas et al., 2013). The megaraptorid ulna narrows distally, a

condition similar to that of Allosaurus (e.g., Madsen, 1976) or

basal coelurosaurs (e.g. Guanlong, Ornitholestes, Coelurus;

Ornithomimids; Nichols and Russel, 1985; Xu et al., 2006;

Osborn, 1903; Carpenter, 2005). But absent in megalosauroids

(Dong, 1984; Charig and Milner, 1997) and derived

coelurosaurs (e.g. Deinonychus-, Ostrom, 1969).

Remarkable features characterizing megaraptorids

correspond to the manus, in particular the formidable

development of the manual unguals of digits I and II, and the

Megaraptor manual osteology

Figure 1. Humerus in lateral (C-I) and medial (A-B,J) views of: A, Megaraptor (MUCPv 341), B, Australovenator, C, Allosaurus, D,

Acrocanthosaurus, E, Coelurus, F, Ornitholestes, G, Xuanhanosaurus, H, Torvosaurus, and I, Baryonyx. J, Fukuiraptor. B, modified from

White et al. (2012). D, modified from Currie and Carpenter (2000). H, modified from Galton and Jensen (1979). I, modified from Charing and

Milner (1997). Scale bar: 5cm. Abbreviations: it, internal tuberosity; If, longitudinal furrow.

transverse compression and ventral sharpness of the ungual of

digit I (Calvo et al., 2004; Novas et al., 2013).

Carpus. In Megaraptor (Calvo et al., 2004) and Australovenator

(White et al., 2012) two carpal elements are documented: a disk¬

shaped radiale, and an enlarged distal carpal described as distal

carpal 1 by White et al. (2012). Because the homology of this

bone among theropods is difficult to interpret (e.g., Xu et al.,

2006,2009,2014), we will informally describe it as a “semilunate

carpal”, based on its proximally arched profile in dorsal view.

Semilunate carpals of Megaraptor and Australovenator

resemble Allosaurus (Madsen, 1976) in being gently convex

proximally (figs. 3, 4, 5). As in the latter taxon, the semilunate

carpal is in contact with most of the proximal end of metacarpal

I, and also the medial half of the proximal end of metacarpal

II. The semilunate carpal of megaraptorids bears a pair of

distal projections for articulation with metacarpal bones, also

present in Allosaurus, Acrocanthosaurus and the basal

coelurosaur Guanlong (Madsen, 1976; Currie and Carpenter,

2000; Xu et al., 2006). One of these projections is visible in

ventral view, and wedges between metacarpals I and II. The

other projection is seen in dorsal view, and lodges into a socket

on the proximal end of metacarpal I. Such interlocking among

the semilunate carpal and metacarpals I and II probably

constitutes a tetanuran feature, apomorphically lost among

derived coelurosaurs (e.g., oviraptorosaurs, paravians) in

which the distal surface is flat or slightly concave, without

projecting between metacarpals I and II (Rauhut, 2003).

Aside from the general similarities noted with Allosaurus,

the semilunate carpal of megaraptorids exhibits a

proximodistally deep profile, mainly due to the bulged

condition of the distal projection that lodges into the proximal

end of metacarpal I. In this regard, the semilunate carpal of

Megaraptor and Australovenator differs from the

F.E. Novas, A.M. Aranciaga Rolando & F.L. Agnolin

Figure 2. Distal end of humerus in anterior (A,C,E,G,I,K,) and distal

(B,D,F,H,J,L) views of Australovenator (A,B), Allosaurus (C,D),

Xuanhanosaurus (E,F), Chilantaisaurus (G,H), Guanlong (I,J), and

Coelurus (K,L). Not to scale. A,B, modified from White et al. (2012).

G,H, modified from Benson and Xu (2008).

proximodistally shallower semilunate carpal of basal

tetanurans (e.g., Allosaurus, Acrocanthosaurus) and basal

coelurosaurs, such as Tanycolagreus (Carpenter et al., 2005),

Sinosauropteryx (Currie and Chen, 2001), Scipionyx (Dal

Sasso and Maganuco, 2011), Coelurus, and ornithomimosaurs

(Kobayashi and Lii, 2003).

In sum, megaraptorids retained a carpal morphology

diagnostic at the level of Tetanurae. No derived features shared

with coelurosaurs are identified. The distally convex condition

of the semilunate carpal probably represents a synapomorphic

feature for Megaraptoridae.

Metacarpus. Comparing the forearms of Megaraptor with

those of Allosaurus and Acrocanthosaurus (equaling the

length of the ulna), permits recognition that the manus of the

first taxon is much more elongate and slender than in those

basal tetanurans. In particular, metacarpal I of Megaraptor is

less massive than the block-like Metacarpal I of Allosaurus,

Acrocanthosaurus, and Torvosaurus (figs. 3, 6; Madsen, 1976;

Gabon and Jensen, 1979; Currie and Carpenter, 2000). In

Megaraptor the ratio between transverse diameter and total

length of the metacarpal I results in, approximately 40, whereas

in Allosaurus the same ratio is of 50 (Novas, 1998). Digits II

and III of Megaraptor are considerably elongate, in particular

their respective ungual phalanges. The exception is digit III,

which is not proportionally longer with respect to Allosaurus.

In this regard, the shortness of digit III was considered as a

derived feature shared by megaraptorids and tyrannosaurids

(Novas et al., 2013). Moreover, the ungual phalanx of digit III

of Megaraptor is less curved and trenchant than its homologue

in Allosaurus. Australovenator also exhibits slender

metacarpals as in Megaraptor, as well as an enlarged ungual on

digit I. However, proportions of the remaining phalanges are

intermediate between those of Allosaurus and Megaraptor.

Metacarpal I. As pointed out by Rauhut (2003), metacarpal I in

most coelurosaurs is much longer than broad. Rauhut (2003)

proposed that a length:width ratio greater than 2.2 is diagnostic

for derived coelurosaurs (e.g., Ornitholestes, troodontids,

oviraptorids, dromaeosaurids), and that was <2 in other

theropods. Metacarpal I of megaraptorids exhibits slender

proportions resembling those of coelurosaurs, contrasting with

most non-coelurosaurian theropods in which the metacarpal is

approximately as broad as long (e.g., Allosaurus, Torvosaurus,

Acrocanthosaurus-, fig. 7). In megaraptorids the metacarpal I

has a length: width ratio of 1.85 for Megaraptor, and 2 for

Australovenator. This contrasts with non-coelurosaurian

theropods, such as Allosaurus, Acrocanthosaurus, and

Xuanhanosaurus, in which the relationship between

length: width is 1.52, 1,24 and 1,67 respectively (Madsen, 1976;

Dong, 1984; Currie and Carpenter, 2000). In addition, the

elongation of metacarpal I is also shared by the Australian

“Rapator ” (see White et al., 2013). On the other hand, in

coelurosaurians like Deinonychus and Guanlong, the ratio is

1,89 and 1,86 respectively (Ostrom, 1076; obs. pers.), resembling

in this aspect the megaraptoran condition.

As already said, the proximal end of metacarpal I bears a

deep embayment to lodge the semilunate carpal. This proximal

concavity of metacarpal I is also present in basal tetanurans (e.g.,

Allosaurus ) as well as basal tyrannosauroids (e.g., Guanlong ),

but in megaraptorids it is emphasized by the presence of a

prominent proximal projection on the medial corner of the bone.

Huene (1932), in the original description of Rapator

ornitholestoides, pointed out the peculiar proximomedial

process of metacarpal I (figs. 7, 8). This feature was usually

considered as a probable autapomorphic trait diagnostic for this

taxon (e.g., Molnar, 1980, 1990). However, Agnolin et al. (2010)

recognized that a similar process is also present in

Australovenator and Megaraptor, thus suggesting that it may

constitute a synapomorphy of Megaraptoridae (see also White et

al., 2012). The proximal concavity on metacarpal I and its

associated proximomedial process are less well developed in

basal coelurosaurs (e.g., Scipionyx-, Dal Sasso and Maganuco,

2011), basal tyrannosauroids (e.g., Tanycolagreus-, Carpenter,

Miles and Cloward, 2005), and paravians (e.g., Deinonychus-,

Ostrom, 1976), in which the proximal margin of metacarpal I is

almost straight and a proximomedial process is lacking. The

only possible exception among basal coelurosaurs is the

compsognathid Sinosauropteryx, which appears to posseses a

metacarpal I that is proximally notched and bears an associated

proximomedial process (Figure 6; Currie and Chen, 2001).

In the Australian megaraptorids Australovenator and

“Rapator ” the lateral margin of metacarpal I is straight (in

dorsal and ventral views), and the lateral surface for articulation

with metacarpal II is slightly faced dorsally (fig. 8). This

morphology resembles metacarpal I of basal tyrannosauroids

(e.g., Guanlong-, Xu et al., 2006) and derived coelurosaurs

(e.g., Deinonychus-, Ostrom, 1969), and differs from basal

tetanurans (e.g., Torvosaurus, Allosaurus, Acrocanthosaurus-,

Madsen, 1976; Currie and Carpenter, 2000; Galton and Jensen,

Megaraptor manual osteology

Figure 3. Left manus of Megaraptor namunhuaiquii (MUCPv 341) in dorsal view (A) and schematicrepresentation (B). Scale bar: 1 cm.

1979) in which metacarpal I possesses a well-developed

posterolateral surface (also partially faced proximally) for

articulation with metacarpal II. The latter bone has a

transversely expanded its proximal head, embracing

metacarpal I ventrally. The morphology of the proximolateral

portion of metacarpal I and the way it articulates with

metacarpal II is not uniform among megaraptorids, as shown

by Megaraptor in which the proximolateral corner of

metacarpal I is truncated in a similar condition to that

described for Allosaurus (Madsen, 1976). In other words,

Megaraptor exhibits the ancestral tetanuran condition, but its

close relative Australovenator developed an articulation of

metacarpal I that is morphologically closer to that of

coelurosaurian theropods. This suggests that character

transformation within Megaraptoridae has been more complex

than we expected.

In megaraptorids (i.e., Australovenator, Megaraptor,

“Rapator ”) the medial edge of metacarpal I is transversely

rounded and dorsoventrally deep (as seen in proximal view;

fig. 8). This prominent medial margin resembles Allosaurus,

being different from the dorsoventrally depressed and sharp

medial margin present in some coelurosaurs, such as Guanlong

and Deinonychus (Ostrom, 1976; Xu et al., 2006).

In megaraptorids (e.g., Megaraptor, Australovenator,

“Rapator”) the medial distal condyle of metacarpal I is more

distally placed than in other theropods (Calvo et al, 2004;

F.E. Novas, A.M. Aranciaga Rolando & F.L. Agnolin

Figure 4. Left manus of (A,C), Allosaurus fragilis, and (B,D), Australovenator wintonensis in (A,B) dorsal, and (C,D) ventral views. Not to scale.

B,D, mofied from White et al. (2012).

Figure 5. Left “semilunate” carpal in proximal (upper row) and dorsal (lower row) of A, Allosaurus, B, Acrocanthosaurus (modified from Currie

and Carpenter, 2000); C, Megaraptor, D, Guanlong (modified from Xu et al., 2014); E, Ornitholestes (mofiied from Carpenter et al., 2005); F,

Tanycolagreus (modified from Carpenter et al., 2005); G, Alxasaurus (modified from Xu et al., 2014); H, Deinonychus (modified from Ostrom,

1969); and I, Australovenator (modified from White et al., 2012). Not to scale. Abbreviations: ag, anterior groove; dp, distal projections.

White et al., 2012; fig. 8). In addition, the distal end of

metacarpal I in megaraptorids is distally oriented, lacking a

medial tilting (Calvo et al., 2004; Agnolin et al., 2010; White

et al., 2012, 2013). This morphology results in a metacarpal I

that is distally less asymmetrical than in other theropods, with

the exception of derived paravians, including Archaeopteryx,

dromaeosaurids and troodontids, in which the distal end lacks

the medial twisting present in other theropods (Rauhut, 2003).

In most saurischians, including theropods, the distal end of

the first metacarpal I shows asymmetrically developed

articular condyles, in which the lateral condyle is larger than

the medial condyle (Galton, 1971). This pattern is also present

in all known megaraptorids (Calvo et al., 2004; White et al.,

2012, 2013). However, the distal end of metacarpal I shows

some minor distinctions among megaraptorids: in Megaraptor

metacarpal I differs from Allosaurus and Australovenator in

the presence of a greatly developed lateral distal condyle,

which is ventrally wider than in the above mentioned taxa. In

Australovenator, the medial distal condyle is prominently

projected ventrally (as seen in distal view; see White et al.,

2012, fig.l3C), constituting a condition hitherto unreported

among theropods, with the exception of of Guanlong in which

the medial condyle projects incipiently ventrally. Differences

between Megaraptor and Australovenator may reveal subtle

variations in the way digit I functioned. Contrasting with

Acrocanthosaurus and Allosaurus (Madsen, 1976; Currie and

Carpenter, 2000), Megaraptor has a metacarpal I that bears

distal articular condyles that are little-developed dorsally and

lack the globe-shaped morphology characteristic of the

aforementioned allosauroids. In the same way, Xuahanosaurus

has a poorly developed distal articular surface in both views

(Dong, 1984). The extensor ligament pit of metacarpal I in

Megaraptor is roughly triangular in outline, unlike the

transversely elongate and elliptical form of this feature

Acrocanthosaurus and Allosaurus (Madsen, 1976; Currie and

Carpenter, 2000; fig. 7). Guanlong has a similar condition to

Megaraptor (Xu et al., 2014).

The dorsal surface of metacarpal I in non-coelurosaurian

theropods (e.g., Allosaurus, Acrocanthosaurus, Torvosaurus;

Madsen, 1976; Galton and Jensen, 1979; Currie and Carpenter,

2000) is longitudinally grooved. This groove is contiguous

with a similar trough on the dorsal surface of the semilunate

carpal (fig. 5). By contrast, in coelurosaurs (e.g., Scipionyx,

Tyrannosaurus, Falcarius, Gallimimus, Deinonychus-,

Ostrom, 1979; Brochu, 2003; Zanno, 2010; Dal Sasso and

Maganuco, 2011) the dorsal surface of metacarpal I and its

Megaraptor manual osteology

Figure 6. Left manus in dorsal view of A, Dilophosaurus (modified from Welles, 1980); B, Allosaurus-, C, Megaraptor, D, Sinocalliopteryx\ E,

Tanycolagreus (modified from Carpenter et al., 2005); F, Deinonychus (modified from Ostrom, 1969); G, Scipionyx (modified from Dal Sasso

and Maganuco, 2011); H, Guanlong (modified from Xu et al., 2009); and I, Sinosauropteryx (modified from Currie and Chen, 2001). Not to scale.

Figure 7. Right metacarpals II and I in dorsal view of A, Acrocanthosaurus (modified from Currie and Carpenter, 2000); B, Torvosaurus

(modified from Galton and Jensen, 1979); C, Megaraptor, D, Deinonychus ( modified from Ostrom, 1969); E, Guanlong (modified from Xu et

al., 2009). Not to scale. Abbreviations: ep, extensor pit; pdp, proximomedial process; ps, proximolateral surface.

corresponding carpal is almost flat or slightly concave. In

Australovenator the dorsal surfaces of both metacarpal I and

the semilunate carpal are almost flat, resembling the condition

described for coelurosaurs. In Megaraptor the metacarpal I is

slightly concave, and although the semilunar carpal is

damaged, its dorsal surface is flattened. A similar condition to

Megaraptor is retained in other basals coelurosaurs like

Guanlong, Ornitholestes and Tanycolagreus which possesses

a deep groove in dorsal view (Carpenter et al., 2005; Xu et al.,

2006). In sum, the absence of a continuous proximodistal

groove on metacarpal I and semilunate carpal may constitute

a sinapomorphic trait uniting megaraptorids with coelurosaurs

retained in some basals coelurosaurs.

Metacarpal II. In Megaraptor and Australovenator the

metacarpal II is long and slender, with a distal ginglymoid

transversely narrower than the proximal end of the bone. This

condition differs from that of Syntarsus, Dilophosaurus,

Allosaurus, and Acrocanthosaurus, in which the distal end of

metacarpal II bears a prominent ginglymus that flares on both

sides, with a transverse diameter equals to that of the proximal

end. The just condition described for megaraptorids resembles

that of Compsognathus (Ostrom, 1969) and Sinocalliopteryx

(Ji et al., 2007). An intermediate step between the allosauroid

and the megaraptorid condition is seen in Guanlong (Xu et al.,

2006). Scaled at the same size, the distal ginglymoid of

metacarpal II of Megaraptor is considerably narrower than that

of Allosaurus, representing half the transverse diameter of the

latter taxon's metacarpal II. Another condition is seen in

derived coelurosaurids (Deinonychus; Ostrom, 1969) which

has a slender metacarpal I with equally developed extremities.

In congruence with the narrow condition of distal ginglymus,

the extensor ligament pit of metacarpal II in Megaraptor has a

proximodistally extended sub-triangular contour, similar to

Sciurumimus (Rauhut et al., 2012), but different from the

proximodistally short and transversely wide ligament pit of

Allosaurus (Madsen, 1976).

As mentioned above, in Megaraptor the proximal head of

metacarpal II is medially expanded, ventrally embracing

metacarpal I. This condition differs from that of most

coelurosaurs, including Compsognathus, tyrannosauroids

(e.g., Guanlong, Tanycolagreus, Tyrannosaurus; Xu et al.,

2009; Carpenter et al., 2005; Brochu, 2003), and more

crownward forms (e.g., Ornitholestes, Deinonychus,

Velociraptor, Carpenter et al., 2005; Ostrom, 1976), in which

F.E. Novas, A.M. Aranciaga Rolando & F.L. Agnolin

Figure 8. A-C, left first metacarpal in dorsal view of A, Megaraptor,

B, Australovenator, and C, Rapator ; D-F, proximal view of left

metacarpus of D, Guanlong (modified from Xu et al.,2009), E,

Tanycolagreus (modified from Carpenter et al., 2005), and F,

Deinonychus (modified from Ostrom, 1969); G-FI, proximal view of

right first metacarpal of G, Rapator, and H, Australovenator. Not to

scale. Abbreviations: pdp, proximomedial process; vpl, ventral

process of metacarpal I; vpll, ventral process of metacarpal II.

the lateroventral margin of metacarpal I is laterally projected,

thus embracing the ventral surface of metacarpal II.

Metacarpal III. Among megaraptorans, this bone has been

solely recorded in Megaraptor. Calvo et al. (2004) described

the metacarpal III of Megaraptor as transversally compressed,

its distal end being narrower than its proximal end. This

condition is also present in most tyrannosaurids (e.g.,

Daspletosaurus, Tyrannosaurus, Albertosaurus; Russell,

1970; Lipkin and Carpenter, 2008), in which metacarpal III is

extremely slender. This condition has been interpreted as

diagnostic of advanced tyrannosauroids (Holtz, 2004).

The reduction of metacarpal III is correlated with the

reduction of the entire digit III. In Megaraptor the phalanges

of digit III are proximodistally shortened and transversely

compressed, thus resulting in a digit III shorter and more

slender than in basal tetanurans (e.g., Allosaurus,

Acrocanthosaurus-, Madsen, 1976; Currie and Carpenter,

2000). This peculiar morphology may be regarded as

autapomorphic for Megaraptor.

In Megaraptor, the length of metacarpal III represents 71%

of metacarpal II, a ratio that matches that of specialised

tyrannosauroids (Russell, 1970; Barsbold, 1982; Rauhut, 2003;

Holtz, 2004). This proportion, as well as the short length of the

entire digit III may be a condition shared between both groups.

Megaraptor retained a small and rod-like metacarpal IV,

and no evidence of phalanges of digit IV have been found in the

preserved manus (Calvo et al., 2004), thus it is probable that

digit IV was completely lost. The only available specimen of

Australovenator does not preserve metacarpal IV (Hocknull et

al., 2009; White et al., 2012). Presence of metacarpal IV in

Megaraptor is here interpreted as an apomorphic reversal from

the neotetanuran ancestral state, in which metacarpal IV is

absent (e.g., Sciurumimus, Allosaurus, Acrocanthosaurus;

Rauhut, 2003). This conclusion agrees with Rauhut et al. (2012)

who recognized a high level of homoplasy in this characteristic,

given that the basal allosauroid Sinraptor (Currie and Zhao,

1993) and the basal tyrannosauroid Guanlong (Xu et al., 2006)

retained a rudimentary fourth metacarpal.

Manual phalanges. In Megaraptor and Australovenator,

manual phalanges exhibit shallow and triangular-shaped

extensor ligament pits, which lack well-defined margins and

are not proximally delimited by a transverse ridge (fig. 7).

Rauhut (2003) pointed out that coelurosaurs lack well-defined

extensor pits on manual phalanges. In contrast, in non-

coelurosaurian theropods, extensor ligament pits are deep and

transversely extended, as shown for example in Eoraptor,

Dilophosaurus, Syntarsus, Xuanhanosaurus, Torvosaurus,

Allosaurus, Acrocanthosaurus, Sinraptor, and Baryonyx

(Raath, 1969; Madsen, 1976; Galton and Jensen, 1979; Welles,

1984; Dong, 1984; Currie and Zhao, 1993; Sereno et al., 1993;

Charig and Milner, 1997; Currie and Carpenter, 2000; Rauhut,

2003). In contrast most coelurosaurian theropods have shallow

or absent extensor pits (e.g. Deinonychus, Nothronychus,

Tyrannosaurus, Troodon; Ostrom, 1969; Currie and Russel,

1987; Bochu, 2003; Zanno et al., 2009; Zanno, 2010).

Phalanges of digit I. Megaraptor is distinguished from the

remaining theropods, including Australovenator, in the

remarkable elongation of the internal bones of the manus (i.e.,

metacarpal I, phalanx 1.1, and especially the ungual phalanx).

The tip of digit I ungual ends at the level of the mid-length of

the second ungual digit (fig. 3).

Phalanx 1 of digit I of Megaraptor exhibits a proximodorsal

lip. In most basal theropods (e.g., coelophysoids, Torvosaurus,

Spinosaurus, Allosaurus, Acrocanthosaurus; Rauhut, 2003;

Ibrahim et al., 2014) the phalanx 1.1 bears a transversely wide

proximodorsal lip on phalanx 1 of digit I. Such a wide lip appears

to be related with a transversely extended, deep, and well-defined

extensor ligament pit on distal metacarpal I, a condition regarded

as plesiomorphic among theropods (Sereno et al., 1993; Rauhut,

2003). However, among coelurosaurs (e.g., Tanycolagreus,

Guanlong, Tyrannosaurus, Gallimimus, Deinonychus-, Ostrom,

1976; Brochu, 2003; Carpenter et al., 2005; Xu et al., 2006) the

proximodorsal lip of phalanx 1 is narrower. In Megaraptor and

Australovenator the proximal surface of the proximal phalanx

presents a pointed proximodorsal lip, which is different from the

condition described for the remaining theropods. This pointed

process appears to be related with a reduction in the distal

extensor pits of the metacarpals, as diagnostic of coelurosaurs

(Rauhut, 2003).

In Megaraptor and Australovenator the proximal end of

phalanx 1.1 is sub-quadrangular in outline (fig. 9). It shows

robust and thickened lateral, medial, and dorsal margins,

conforming to an expanded articular surface for metacarpal I.

The lateral margin is even more thickened than the medial one

and is strongly proximally expanded. This set of features

Figure 9. Proximal end of right phalanx 1.1 of A, Megaraptor ; B, Australovenator, C, Allosaurus-, D, Tyrannosaurus (modified from Brochu,

2003); and E, Deinonychus (modified from Ostrom, 1969). Not to scale.

appears to be unique to megaraptorids: in other theropods, the

proximal end is transversely narrow and dorsoventrally deep,

being sub-rectangular in shape (e.g., Allosaurus,

Acrocanthosaurus, Torvosaurus, Tyrannosaurus ; Galton and

Jensen, 1979; Madsen, 1976; Currie and Carpenter, 2000;

Brochu, 2003) or subtriangular in outline (as in Guanlong and

Deinonychus ; Ostrom, 1976; Xu et al., 2006). Furthermore, in

Megaraptor the proximal articular surface is transversely

wider dorsally than ventrally (Novas, 1998). This condition is

unknown in other theropods, including Australovenator

(White et al., 2012), in which the proximal end is transversely

wider on its ventral margin than on its dorsal edge.

Phalanx 1 of digit I in Megaraptor shows a deep and wide

furrow along its ventral surface (Novas, 1998). As a result, both

lateral and medial margins of this surface acquired the form of

sharp longitudinal ridges (fig. 10). These features are also

documented in Australovenator (White et al., 2012). In other

theropods, phalanx 1.1 is ventrally excavated, but the furrow is

restricted on the proximal half of the bone, and it is not as deep

as in megaraptorids. No longitudinal ridges are present. It is

interesting to note that in megaraptorids, the ventral margin of

the proximal articular surface of phalanx 1.1 is concave,

reflecting the deep furrow present along the ventral surface of

the bone. This is in contrast with other theropods, in which this

margin is straight (e.g., Allosaurus-, Madsen, 1976) or convex

(e.g., Guanlong, Deinonychus-, Ostrom, 1976; Xu et al., 2006).

The distal ginglymus of phalanx 1.1 of Megaraptor is

dorsoventrally deeper and transversely narrower than in other

theropods (including Australovenator ), and the dorsoventral

sulcus is much more incised.

Megaraptor is well-known by its extremely large and

elongate manual ungual on digit I (Calvo et al., 2004), which is

subequal in length to the ulna. This condition is unusual

among theropods, being absent among basal tetanurans (e.g.,

Allosaurus-, Madsen, 1976), basal coelurosaurs (e.g., Scipionyx,

Tanycolagreus, Chilantaisaurus-, Dal Sasso and Maganuco,

2011; Carpenter et al., 2005; Benson and Xu, 2008),

ornithomimosaurs (e.g., Gallimimus), oviraptorosaurs, basal

therizinosaurs (e.g., Falcarius, Nothronychus\ Zanno, 2010;

Zanno et al., 2009), and paravians (e.g., Deinonychus-, Ostrom,

1969). Furthermore, in the megaraptorans Australovenator

and Fukuiraptor, the ungual of digit I is much shorter than the

ulna, representing approximately half of its length. Basal

tetanurans that evolved an enlarged ungual in manual digit I

are the compsognathid Sinosauropteryx (Currie and Chen,

2001), and the megalosauroids Baryonyx and Torvosaurus

(Galton and Jensen, 1979; Charig and Milner, 1997).

In the original description of Megaraptor (Novas, 1998), it

was remarked that the ungual phalanx bore a sharp longitudinal

ventral keel. This trait was later considered as a synapomorphy

of Megaraptoridae (Novas et al., 2013). In Megaraptor, towards

the proximal end of the claw, the ventral keel gradually displaces

laterally, joining the lateral margin of the claw on its most

proximal portion, a condition also reported in Australovenator

(White et al., 2012; fig. 11). Other theropods, including

Fukuiraptor (Azuma and Currie, 2000), basal tyrannosauroids

(e.g., Guanlong-, Xu et al., 2006), megalosauroids (e.g.,

Baryonyx, Torvosaurus-, Galton and Jensen, 1979; Charig and

Milner, 1997) and the problematic Chilantaisaurus (Benson

and Xu, 2008) have unguals with a transversely rounded

expanded ventral surface, without traces of a ventral keel. In

sum, such a transverse compression of the enlarged ungual

constitutes a distinctive feature of Megaraptoridae.

In addition, the manual ungual I of Megaraptor and

Australovenator share very deep and well-defined flexor

facets on the lateral and medial surfaces of the flexor tubercle.

These facets are deep, wide, and more well-defined than in

other theropods, including Allosaurus, Baryonyx and

Torvosaurus (Madsen, 1976; Galton and Jensen, 1979; Charig

and Milner, 1997). Furthermore, in Megaraptor such facets

are delimited by acute ridges of bone (Figure 12). It is worth

nothing that similar facets were described for Fukuiraptor

(Azuma and Currie, 2000).

Digit II. In Megaraptor, phalanx l.II is shorter than phalanx

2.II, a condition similar to that of some allosauroids, such as

Allosaurus (Gilmore, 1920; Madsen, 1976) and

Acrocanthosaurus (Currie and Carpenter, 2000), and selected

coelurosaurs, as for example Sinocalliopteryx (Ji et al., 2007),

Sinosauropteryx (Currie and Chen, 2001), Scipionyx (Dal Sasso

and Maganuco, 2011), Guanlong and Deinonychus. Distribution

of this feature (i.e., length ratio of pre-ungual phalanges of digit

II) is not uniform among tetanurans. For example, in the

megaraptorid Australovenator and the basal tyrannosauroid

Tanycolagreus (Carpenter et al., 2005), phalanges 1 and 2 of

F.E. Novas, A.M. Aranciaga Rolando & F.L. Agnolin

Figure 10. Right manual phalanx 1 of digit I in ventral view and schematic representations of Megaraptor (A, C), Australovenator (B,D). Scale

bar: 2 cm. Note the well-developed longitudinal ventral furrow.

Figure 11. Right manual ungual phalanx of digit I in ventral view and schematic representation of Megaraptor (A,C); and Australovenator (B,D).

Not to scale.

Megaraptor manual osteology

Figure 12. Right manual ungual phalanx of digit I in A,C, ventral, and B,D, lateral views. A-B, Megaraptor, C-D, Australovenator and schematic

representation in E,G, ventral, and F,H, lateral views. E-F, Megaraptor; G-H, Australovenator. Scale bar: 2 cm. Abbreviations: ff, flexor facets.

digit II are subequal in length, and the megalosauroid

Sciurumimus (Rauhut et al., 2012) shows coelurosaur-like

proportions, with phalanx l.II shorter than phalanx 2.II.

In Megaraptor, the proximal articular surface of phalanx

l.II describes a dorsoventrally deep ovoid contour. Its ventral

margin bears a rounded process that projects proximomedially,

a feature shared with Australovenator (White et al., 2012) and

Fukuiraptor (Azuma and Currie, 2000). This results in a

relatively narrow ventral margin of the proximal end of phalanx

1.1. This shape is in contrast with other theropods, such as

Allosaurus and Tyrannosaurus (Madsen, 1976; Brochu, 2003),

in which the ventral margin is straight. Furthermore, in

Megaraptor, and probably also in Australovenator and

Fukuiraptor, the proximal articular surface phalanx l.II is

obliquely oriented with respect to the distal articular trochlea,

a condition unknown in other theropods, in which the main

axes of both proximal and distal ends are sub-parallel.

In Megaraptor, metacarpal II and its corresponding non-

ungual phalanges have respective distal articular trochleae

with a medial condyle more ventrally projected than the lateral

one. In probable correlation with this shape, it is seen that non-

unguals of digit II exhibit a longitudinal keel that runs along

their ventromedial margins. Such strong asymmetry of distal

condyles and longitudinal ridges appear to be absent in other

theropods, including Australovenator, although in the

available phalanx l.II of Fukuiraptor (Azuma and Currie,

2000) a similar ventromedial ridge seems to be present.

Digit III. In Megaraptor phalanges of this digit look similar in

proportions to those of Allosaurus (Gilmore, 1920; Madsen,

1976), except for the ungual, which is proportionally shorter

and smaller. The pre-ungual phalanx of digit III of Megaraptor

is longer than phalanges 1 and 2 of the same digit, as generally

occurs among tetanurans, although it does not reach the

elongation that characteristically occurs in coelurosaurs (e.g.,

Sinocalliopteryx, Dilong, Guanlong, Deinonychus; Ostrom,

1976; Xu et al., 2006; Ji et al., 2007).

Conclusions

Shared presence of a longitudinal groove along the medial side

of humeral shaft in megaraptorans and tyrannosaurids

conforms a novel feature supporting close relationships

between these theropod families. Comparison of the manus in

Megaraptor and Australovenator allowed the recognition of

several features that may shed light on the phylogenetic

relationships of megaraptorids. The manus of Megaraptor

exhibits the following unique traits that are not present in

other theropods, and are here interpreted as autapomorphies

of this genus: 1) metacarpal I with an acute medial condyle on

distal gynglimus; 2) phalanges of digit II with ventromedial

ridges; and 3) an extremely elongate manual ungual on digit I,

approximating the length of the ulna.

Manual characters here interpreted as diagnostic of

Megaraptoridae include: symmetrical-shaped metacarpal I,

proximal end of phalanx 1.1 transversally expanded, phalanx 1.1

with a longitudinal ventral furrow, and ungual phalanx of digit

I with a laterally displaced sharp ventral margin. Manual

characters diagnostic of Megaraptora are more difficult to

recognize because the manus of the basal megaraptoran

Fukuiraptor is poorly known. Nevertheless, two possible

derived features have been identified: asymmetrical phalanx 1.

II; and first digit ungual with deep facets on the flexor tubercle.

After comparing carpal, metacarpal and phalangeal

morphology, it becomes evident that megaraptorids retained

several of the manual features present in basal tetanurans, such

as Allosaurus. In this regard, Megaraptor and Australovenator

are devoid of several manual features that the basal

tyrannosauroid Guanlong shares with more derived coelurosaurs

(e.g., Deinonychus ). However, there are some manual characters

that support Megaraptora as members of Coelurosauria,

including the elongate and slender shaft of metacarpals I and II,

and the presence of separated flexor and extensor distal end of

the humerus, and the absence of a longitudinal furrow on the

dorsal surface of metacarpal I, and a semilunar carpal.

Furthermore, megaraptorans are similar to specialised members

of Tyrannosauroidea in having a transversely narrow metacarpal

III that represents 0.75 the length of metacarpal II, a set of

features previously interpreted as synapomorphies uniting both

clades (Novas et al., 2013; Porfiri et al., 2014).

Acknowledgements

The senior author wishes to thank deeply Tom and Pat Rich for

their genereous invitation to visit Australia, and for allowing

the study of the valuable Cretaceous theropod materials they

collected in Victoria. Scott Hocknull and his crew kindly

facilitated access to the specimen of Australovenator. Xu

F.E. Novas, A.M. Aranciaga Rolando & F.L. Agnolin

Xing allowed access to several theropod specimens under his

care. Special thanks to Steve Brusatte and an anonymous

reviewer that made clever observations that improved the

quality of the manuscript.

References

Agnoh'n, F.L., Ezcurra, M.D., Pais, D.F. and Salisbury, S.W. 2010. A

reappraisal of the Cretaceous non-avian dinosaur faunas from

Australia and New Zealand: evidence for their Gondwanan

affinities. Journal of Systematic Palaeontology 8; 257-300.

Azuma, Y. and Currie, P.J. 2000. A new carnosaur (Dinosauria:

Theropoda) from the Lower Cretaceous of Japan. Canadian

Journal of Earth Sciences 37; 1735-1753.

Benson, R.B.J., Carrano, M.T. and Brusatte, S.L. 2010a. A new clade

of archaic large-bodied predatory dinosaurs (Theropoda:

Allosauroidea) that survived to the latest Mesozoic.

Naturwissenschaften 97; 71-78.

Benson, R.B.J., Barrett, P.M., Rich, T.H., Vickers-Rich, P., 2010b. A

southern tyrant reptile. Science 327, 1613.

Benson, R.B.J., Rich, T.H., Vickers-Rich, P. and Hall, M. 2012.

Theropod fauna from Southern Australia indicates high Polar

Diversity and Climate-Driven dinosaur provinciality. PlosOne

7(5), e37122.

Benson, R.B.J. and Xu X. 2008. The anatomy and systematic position

of the theropod dinosaur Chilantaisaurus tashuikouensis Hu,

1964 from the Early Cretaceous of Alashan, People's Republic of

China. Geological Magazine 145; 778-789.

Brochu, C.A., 2003. Osteology of Tyrannosaurus rex: insights from a

nearly complete skeleton and high-resolution computed

tomographic analysis of the skull. Society of Vertebrate

Paleontology Memoir 7; 1-138.

Bonaparte, J. 1986. Les Dinosaures (Carnosaures, Allosaurides,

Sauropodes, Cetiosaurides) du Jurassique moyen de Cerro

Condor (Chubut, Argentine). Annales de Pateontologie, 72; 247-

289, 326-386.

Calvo, J.O., Porfiri, J.D., Veralli, C., Novas, F.E. and Poblete, F. 2004.

Phylogenetic status of Megaraptor namunhuaiquii Novas based

on a new specimen from Neuquen, Patagonia, Argentina.

Ameghiniana 41; 565-575.

Carpenter, K., Miles, C. and Cloward, K. 2005. New small theropod

from the Upper Jurassic Morrison Formation of Wyoming. In:

Carpenter, K. (ed). The Carnivorous Dinosaurs. Indiana

University Press, Bloomington, pp. 23-48.

Charig, A.J. and Milner, A.C. 1997. Baryonyx walkeri, a fish-eating

dinosaur from the Wealden of Surrey. Bulletin of the Natural

History Museum of London 53; 11-70.

Currie, P.J. and Zhao, X. 1993. A new carnosaur (Dinosauria,

Theropoda) from the Jurassic of Xinjiang, People's Republic of

China. Canadian Journal of Earth Sciences 30; 2037-2081.

Currie, P.J. and Carpenter, K. 2000. A new specimen of

Acrocanthosaurus atokensis (Theropoda, Dinosauria) from the

Lower Cretaceous Antlers Formation (Lower Cretaceous,

Aptian) of Oklahoma, USA. Geodiversitas 22; 207-246.

Currie, P.J. and Chen, P.J. 2001. Anatomy of Sinosauropteryx prima

from Liaoning, northeastern China. Canadian Journal of Earth

Sciences 38; 705-727.

Currie, P.J. and Russell, D. A. 1988. Osteology and relationships of

Cirosrenotes pergracilis (Saurischia, Theropoda) from the Judith

River (Oldman) Formation of Alberta, Canada. Canadiall

Journal of Earth Sciences, 25\ 972-986.

Dal Sasso, C. and Maganuco, S. 2011. Scipionyx samniticus

(Theropoda: Compsognathidae) from the Lower Cretaceous of

Italy. Osteology, ontogenetic assessment, phylogeny, soft tissue

anatomy, taphonomy and palaeobiology. Memorie Della Societa

Italiana de Scienze Naturali e del Museo Civico di Storia

Naturale di Milano 37; 1-281.

Dong, Z., 1984. A new theropod dinosaur from the Middle Jurassic of

Sichuan Basin. Vertebrata PalAsiatica 22; 213-218. [In Chinese].

Fitzgerald, E.M.G., Carrano, M.T., Holland, T., Wagstaff, B.E.,

Pickering, D., Rich, T.H. and Vickers-Rich, P. 2012. First

ceratosaurian dinosaur from Australia. Naturwissenschaften 99:

397-405.

Galton, PM. 1971. Manus movements of the coelurosaurian dinosaur

Syntarsus and opposability of the theropod hallux. Arnoldia 15:

1 - 8 .

Galton, PM. and Jensen, J. A. 1979. A new large theropod dinosaur

from the Upper Jurassic of Colorado. Brigham Young University

Geology Studies 26; 1-12.

Gilmore, C.W. 1920. Osteology of the carnivorous Dinosauria in the

United States National Museum, with special reference to the

genera Antrodemus (Allosaurus ) and Ceratosaurus. Bulletin of

the United States National Museum 110, 1-154.

Hocknull, S.A., White, M.A., Tischler, T.R., Cook, A.G., Calleja,

N.D., Sloan, T. and Elliott, D.A. 2009. New Mid-Cretaceous

(Latest Albian) Dinosaurs from Winton, Queensland, Australia.

PLoS ONE 4, e6190.

Holtz, T.R. Jr. 2004. Tyrannosauroidea. In: Weishampel, D.B.,

Dodson, P, Osmolska, H. (eds). The Dinosauria, Second Edition.

University of California Press, pp. 111-136.

Huene, F. von., 1932. Die fossile Reptil-Ordnung Saurischia, ihre

Entwicklung und Geschichte. Monographie fur Geologie und

Palaontologie 4; 1-361.

Ibrahim, N., Sereno, P. C., Dal Sasso, C., Maganuco, S., Fabbri, M.,

Martill, D. M., Zouhri, S., Myhrvold, N. and Iurino, D. A. 2014.

Semiaquatic adaptations in a giant predatory dinosaur. Science,

doi:10.1126/science.l258750

Ji, S., Ji, Q., Lu J. and Yuan, C. 2007. A new giant compsognathid

dinosaur with long filamentous integuments from Lower

Cretaceous of Northeastern China. Acta Geologica Sinica 81;

8-15.

Kobayashi, Y. and Lii, J.-C. 2003. A new ornithomimid dinosaur with

gregarious habits from the Late Cretaceous of China. Acta

Palaeontologica Polonica 48; 235-259.

Lipkin, C. and Carpenter, K. 2008. Looking again at the forelimb of

Tyrannosaurus rex. In: Larson P.L., Carpenter K. (eds)

Tyrannosaurus rex , the tyrant king. Indiana University Press, pp.

167-192.

Lu, J., 2002. A new oviraptorosaurid (Theropoda: oviraptorosauria)

from the late Cretaceous of southern of China. Journal of

Vertebrate Paleontology 22(4):871-875.

Madsen, J.H., Jr. 1976. Allosaurus fragilis: a revised osteology. Utah

Geological and Mineralogical Survey Bulletin 109; 3-163.

Madsen, J.H. and Welles, S.P. 2000. Ceratosaurus (Dinosauria,

Theropoda). A revised osteology. Utah Geological Survey.

Miscellaneous Publication, 00-2, 80 pp.

Molnar, R.E., 1980. Australian Late Mesozoic terrestrial tetrapods:

some implications. Memoires de la Societe Geologique de France

139;131-143.

Molnar, R.E., 1990. Problematic Theropoda: “Carnosaurs”. In:

Weishampel, D.B., Dodson, P., and Osmolska, H. (eds). The

Dinosauria. University of California Press, Berkeley, pp. 306-

317.

Megaraptor manual osteology

Molnar, R.E., Flannery, T.F. and Rich, T.H.V. 1981. An allosaurid

theropod dinosaur from the Early Cretaceous of Victoria,

Australia. Alcheringa 5: 141-146.

Nichols, E.L. and Russel, A.R 1985. Structure and function of the

pectoral girdle and forelimb of Struthiomimus altus (Theropoda:

ornithomimidae). Palaeontology 28; 643-677.

Novas, F.E. 1998. Megaraptor namunhuaiquii gen. et. sp. nov., a

large-clawed. Late Cretaceous Theropod from Argentina.

Journal of Vertebrate Paleontology 18: 4-9.

Novas, F.E., Ezcurra, M.D. and Lecuona, A. 2008. Orkoraptor burkei

nov.gen. et sp., a large theropod from the Maastrichtian Pari Aike

Formation, Southern Patagonia, Argentina. Cretaceous Research

29; 468-480.

Novas, F.E., Agnolin, F.L., Ezcurra, M.D., Porfiri, J. and Canale, J.I.

2013. Evolution of the carnivorous dinosaurs during the

Cretaceous: the evidence from Patagonia. Cretaceous Research

45; 174-215.

Osborn, H.F. 1903 .Ornitliolestes hermanni, a new compsognathoid

dinosaur from the upper Jurassic. Bulletin American Museum of

Natural History 19; 459-464.

Ostrom, J.H. 1969. Osteology of Deinonychus antirrhopus, an

unusual theropod from the Lower Cretaceous of Montana.

Bulletin of the Peabody Museum of Natural History 30; 1-165.

Porfiri, J.D., Novas, F.E., Calvo, J.O., Agnolin, F.L., Ezcurra, M.D.

and Cerda, I.A. 2014. Juvenile specimen of Megaraptor

(Dinosauria, Theropoda) sheds light about tyrannosauroid

radiation. Cretaceous Research 51; 35-55.

Raath. M.A. 1969. A new coelurosaurian dinosaur from the Forest

Sandstone of Rhodesia. Amoldia, 4, 1-25. 1985. The theropod

Syntarsus and ilS bearing on the origin of birds. 219-227. In

Hecht. M. K., Ostrom.J. H., Viohl. G., and Wellnhofer. P. (ed).

The beginning of birds. Freunde des Jura Museums, Eichsliitt,

382 pp.

Rauhut, O.W.M. 2003. Interrelationships and evolution of basal

theropod dinosaurs. Special Papers in Palaeontology 69; 1-215.

Rauhut, O.W.M. Foth, C., Tischlinger, H. and Norell, M.A. 2012.

Exceptionally preserved juvenile megalosauroid theropod

dinosaur with filamentous integument from the Late Jurassic of

Germany. Proceedings of the National Academy of Sciences of

the United States of America 109; 11746-11751.

Russell, D.A. 1970. Tyrannosaurs from the Late Cretaceous of

western Canada. National Museum of Natural Sciences

Publications in Paleontology 1; 1-34.

Sereno, P.C. and Novas, F.E. 1993. The skull and neck of the basal

theropod Herrerasaurus ischigualastensis. Journal of Vertebrate

Paleontology 13; 451-476.

Sereno, P.C., Martinez, R.N., Wilson, J.A., Varricchio, D.J. and

Alcober, O.A. 2008. Evidence for avian intrathoracic air sacs in

a new predatory dinosaur from Argentina. Plos One 3, e3303.

Smith, N.D., Makovicky, P.J., Agnolin, F.L., Ezcurra, M.D., Pais,

D.F. and Salisbury, S.W. 2008. A Megaraptor- like theropod

(Dinosauria: Tetanurae) in Australia: support for faunal exchange

across eastern and western Gondwana in the Mid-Cretaceous.

Proceedings of the Royal Society of London 275; 2085-2090.

Welles, S.P. 1984. Dilophosaurus wetherilli (Dinosauria, Theropoda).

Osteology and comparisons. Palaeontographica A. 185; 85-180.

White, M.A., Cook, A.G., Hocknull, S.A., Sloan, T., Sinapius, G.H.K.

and Elliott, D.A. 2012. New forearm elements discovered of

holotype specimen Australovenator wintonensis from Winton,

Queensland, Australia. Plos One 7 (6), e39364.

White, M.A., Falkingham, P.L., Cook, A.G., Hocknull, S.A. and

Elliott, D.A. 2013. Morphological comparisons of metacarpal I

for Australovenator wintonensis and Rapator ornitholestoides:

implications for their taxonomic relationships. Alcheringa 37;

1-7.

Xu, X., Clark, J.M., Forster, C.A., Norell, M.A., Erickson, G.M.,

Eberth, D.A., Ji, A.C. and Zhao, Q. 2006. A basal tyrannosauroid

dinosaur from the Late Jurassic of China. Nature 439; 715-718.

Xu, X., Clark, J.M., Mo, J., Choiniere, J., Forster, C.A., Erickson,

G.M., Hone, D.W.E., Sullivan, C., Eberth, D.A., Nesbitt, S., Zhao,

Q., Hernandez, R., Jia, C.-K., Han, F.-L. and Guo, Y. 2009. A

Jurassic ceratosaur from China helps clarify avian digital

homologies. Nature 459; 940-944.

Xu, X., Han, F. and Zhao, Q. 2014. Homologies and homeotic

transformation of the theropod “semilunate” carpal. Scientific

Reports 4, 6042.

Zanno, L.E., Gillette, D. D., Albright L. B. and Titus, L. A. 2009. A

new North American therizinosaurid and the role of herbivory in

‘predatory’ dinosaur evolution. Proceedings of the Royal Society

276; 3505-3511.

Zanno, L.E. 2010. Osteology of Falcarius utahensis (Dinosauria:

Theropoda): characterizing the anatomy of basal therizinosaurs.

Zoological Journal of the Linnean Society 158; 196-230.

Zanno, L.E. and Makovicky, P.J. 2013. Neovenatorid theropods are

apex predators in the Late Cretaceous of North America. Nature

Communications 3827; 1-9.

Memoirs of Museum Victoria 74:63-71 (2016) Published 2016

ISSN 1447-2546 (Print) 1447-2554 (On-line)

http://museumvictoria.com.au/about/books-and-journals/journals/memoirs-of-museum-victoria/

A close look at Victoria’s first known dinosaur tracks

Anthony J. Martin

Abstract

Keywords

Department of Environmental Sciences, Emory University, Atlanta, Georgia 30322 USA (geoam@emory.edu)

Martin, A.J. 2016. A close look at Victoria's first known dinosaur tracks. Memoirs of Museum Victoria 74: 63-71.

Lower Cretaceous (Aptian-Albian) rocks of Victoria, Australia are well known for their dinosaur body fossils, but

not so much for their trace fossils. For example, the first known dinosaur track from the Eumeralla Formation (Albian) of

Knowledge Creek, Victoria, was not discovered until 1980. This specimen, along with two more Eumeralla tracks found

at Skenes Creek in 1989, constituted all of the dinosaur tracks recognised in Lower Cretaceous strata of southern Australia

until the late 2000s. Unfortunately, none of these first-known dinosaur tracks of Victoria were properly described and

diagnosed. Hence, the main purpose of this study is to document these trace fossils more thoroughly. Remarkably, the

Knowledge Creek and one of the Skenes Creek tracks are nearly identical in size and form; both tracks are attributed to

small ornithopods. Although poorly expressed, the second probable track from Skenes Creek provides a search image for

less obvious dinosaur tracks in Lower Cretaceous strata of Victoria. The Skenes Creek tracks were also likely from the

same trackway, and thus may represent the first discovered dinosaur trackway from Victoria. These tracks are the first

confirmed ornithopod tracks for Victoria, augmenting abundant body fossil evidence of small ornithopods

(‘hypsilophodontids’) in formerly polar environments during the Early Cretaceous.

Cretaceous, dinosaur, footprint, ichnology, ornithopod, trace fossil, track.

Introduction

Victoria is world famous for its dinosaur body fossils, which

reflect the best-documented polar-dinosaur assemblage in the

Southern Hemisphere (Rich et al., 2002; Rich and Vickers-

Rich, 2003; Kear and Bruce, 2011; Benson et al., 2012). The

first known dinosaur body fossil in Victoria, a theropod ungual

from the Wonthaggi Formation (Aptian) found by William

Ferguson in 1903, was also the first Australian dinosaur fossil

known to science (Rich and Vickers-Rich, 2000; Rich and

Vickers-Rich, 2003). However, dinosaur trace fossils, such as

tracks, nests, burrows, and other direct evidence of dinosaur

behavior in the Cretaceous rocks of Victoria remained

unnoticed by palaeontologists until 1980, 77 years after the

first recognised body fossil. This ichnological drought ended

when Thomas H. Rich and Patricia Vickers-Rich discovered

and collected a dinosaur track from the Eumeralla Formation

(Albian) at Knowledge Creek, Victoria (Rich and Vickers-

Rich, 2000) (fig. 1).

Another eight years passed before two more dinosaur

tracks were noticed in a Eumeralla Formation stratum at

Skenes Creek in early 1989. In 2006, I recognized two large

theropod tracks in the Wonthaggi Formation at the Flat Rocks

(“Dinosaur Dreaming”) dinosaur dig site, near Inverloch,

Victoria; Tyler Lamb then found another at the same site in

2007 (Martin et al., 2007). Three years later, the largest

assemblage of polar-dinosaur tracks in the Southern

Hemisphere - made by small- to moderate-sized theropods -

was discovered in the Eumeralla Formation at Milanesia

Beach (Martin et al., 2012). Recently, closely associated

tridactyl and tetradactyl tracks were described from Dinosaur

Cove, and were interpreted as non-avian theropod and avian in

origin, respectively (Martin et al., 2014). Otherwise, the only

other trace fossils ascribed to non-avian dinosaurs in Lower

Cretaceous strata of Victoria include possible burrows

(Martin, 2009). Nests, toothmarks, gastroliths, coprolites and

other such trace fossils apparently have not yet been discovered

(Martin, 2014).

The Knowledge Creek track has been figured in numerous

publications, and was much reproduced for educational

purposes (Rich and Vickers-Rich, 2000, 2003). However, it

and the Skenes Creek tracks have not been described nor

interpreted in detail. Thus the main purposes of this study are

to: (1) thoroughly document these tracks; (2) interpret their

dinosaur makers and preservational modes; (3) assess the

palaeontological importance of the tracks; and (4) suggest how

this information might be used to prospect for more such

tracks in Lower Cretaceous strata of Victoria.

Methods

The three specimens are in the Museum Victoria Palaeontology

Collection (NMV P); thus they are available for further study

by qualified researchers. I measured the tracks with Mitutoyo

digital calipers, using minimum-outlines for track widths,

lengths, and other parameters (fig. 2). Digit-impression lengths

A.J. Martin

Figure 1. Locality and outcrop map for Early Cretaceous dinosaur

tracks found thus far in Victoria, with the three tracks described in

this study coming from Knowledge Creek and Skenes Creek. Key and

latitude-longitude coordinates for each of the tracksites, from east to

west: Flat Rocks (FR), S38° 45.3’, E145° 40.9’; Skenes Creek (SC),

S38° 42.9’, E143° 44.4’; Dinosaur Cove (DC), S38° 46.9’, E143° 24.3’;

Knowledge Creek (KC), S38° 45.3’, E143° 20.9’; Milanesia Beach

(MB), S38° 45.3’, E143° 19.3’.

I_2

l_3

Figure 2. Track parameters measured in this study. Key: TW = total

width; TL = total length; L1-L3 = digit lengths; Wl-3 = digit widths;

IA1-IA2 = interdigital angles. Anterior triangle defined by total track

width (base of the triangle) and middle-digit length measured from

that base.

were measured from the midline of each digit, and digit-

impression widths were taken perpendicular to this midline

and medially along the length of each impression. Interdigital

angles were measured with a circular protractor, using a digit-

impression axis radiating from a single point on the rear

margin of the track, as depicted by Thulborn (1990, fig. 4.5).

The anterior triangle length:width ratio (sensu fig. 2 in

Lockley, 2009) was derived from measuring the base of a

triangle, defined by the width between the lateral digit

impressions and the length of the middle digit impression

from that base. Semiquantitative and qualitative information,

such as the host lithology and other descriptive traits of the

tracks, were also noted. All data are provided here and

summarized (table 1) so that future investigators may examine,

test, or otherwise attempt to correct the results reported here.

Descriptions

Knowledge Creek Track. On December 18, 1980, Thomas

Rich and Patricia Vickers-Rich discovered the Knowledge

Creek track, cataloged as NMV P159790 (fig. 3, Appendix I).

The track was located on a marine platform just above sea

level and about 100 m east of Knowledge Creek. Rich and

Vickers-Rich used hand chisels and rock hammers to extract

and collect the track, which they brought to the museum.

The track is in a very fine-fine lithic arenite, although the

track itself is filled with fine-coarse, moderately sorted quartz

and lithic sand held together with hematitic cement. The bed is

36-47 mm thick and horizontally laminated in cross section,

with no apparent disruptions of bedding by bioturbation. The

area surrounding the track is flat, and lacks other physical or

biogenic sedimentary structures on this surface. The track is

preserved as a nearly flat but positive-relief (raised) epichnion,

rather than a depression. It was weathered such that its form is

expressed in nearly full relief.

The track is tridactyl and mesaxonic. It is almost equant in

length and width: 106 mm long and 118 mm wide, with a

lengthiwidth ratio of 0.90. The anterior triangle length:width

ratio is 0.40, with a base (width) of the triangle of 118 mm and

length from that base of 47 mm. Outermost digit impressions

were 90 mm and 84 mm long (left and right, respectively), and

the central impression is also the length of the track, 106 mm.

Medial thicknesses of the three digit impressions, measured

perpendicular to the long axis of each digit, are from left to

right: 25 mm, 31 mm, and 30 mm. Using an average of 29 mm,

digit-impression widths are about 27% of footprint length.

Divarication between the outermost digit impressions is 85°,

which combines an angle of 47° between the left and middle,

and 38° between the right and middle. All three impressions

are outlined completely. Digit impressions narrow distally, but

are subrounded at their ends. The track bears three small, oval

protuberances, two on the middle digit impression and one on

the left. These structures on the middle impression are 4 x 8

mm wide (toward the distal end) and 4x5 mm wide (at the

right intersection with the right impression), whereas the one

on the left impression is 5 x 5 mm (outer edge). Each structure

is labeled as “B” (for “burrow”) on fig. 3b.

The sand fill varies from 4 mm thick at the posterior “heel”

(proximal) end of the track to 8-10 mm thick at the anterior

(distal) ends of each digit impression (fig.3c,d). Thus a

longitudinal profile of the track would show a gradual

thickening of the sand fill from posterior to anterior. A 1-mm

thick, slightly curved thread-like structure, filled with the

same reddish sand as the track, cross-cuts the grey lithic

arenite below the right posterior portion of the track (fig. 3c).

A close look at Victoria's first known dinosaur tracks

Table 1. Measurements of dinosaur tracks from: Knowledge Creek (KC1), specimen P.159790; and Skenes Creek (SCI and SC2), specimen

P.208232. Key: L = length, W = width, L:W = length:width, IA1 = left-middle interdigital angle, IA2 = middle-right interdigital angle,

D = divarication (interdigital angle between left and right), LI = left digit length, L2 = middle digit length (same as track length), L3 = right digit

length, W1 = left digit width, W2 = middle digit width, W3 = right digit width, at-L = anterior triangle length, at-W = anterior triangle width, at-L:

W = anterior triangle length:width, n/a = not applicable. All measurements are in millimeters except for IAl, IA2, and D, which are in degrees.

Track

L:W

1 A 1

1A2

at-L

at-W

at-

L:W

KC1

106

118

0.90

47°

38°

85°

106

118

0.40

SCI

106

115

0.92

37°

46°

83°

106

n/a

115

0.37

SC2

127

138

0.92

44 °

27°

71°

102

127

103

n/a

138

0.31

Figure 3. Knowledge Creek track, NMV P159790. a. Overall top view of track in collected slab, with bedding and chisel marks evident along

edge; scale = 5 cm. b. Outline of track and parameters measured, with anterior triangle indicated (see Figure 2 for key), B = invertebrate burrows;

scale = 5 cm. c. Posterior-edge view of track, showing full relief of track, thin fill of coarse sand, and possible small burrow below (arrow); scale

= 1 cm. D. Anterior-oblique view of track, showing gradually thicker sand fill toward distal ends of digits; scale = 1 cm.

Skenes Creek Track 1. Helmut Tracksdorf, a local citizen and

geologist from the Skenes Creek area, discovered two dinosaur

footprints there in January 1989, with both cataloged as NMV

P208232 (figs. 4-5, Appendix I). Tracksdorf alerted Museum

Victoria about them, and personnel from the Museum collected

the tracks on March 18, 1989. Although no additional

information is available about who collected or cataloged these

tracks, the exact field location of the tracks was just recently

verified, having come from the supratidal marine platform of

rock exposed between Skenes Creek and Browns Creek

(Appendix II).

The most clearly defined of the two tracks (herein

designated Skenes Creek Track 1) is in a 16 x 20 cm cut slab of

very fine-fine, well-sorted lithic arenite. The bed is 25-41 mm

thick, with seven parallel and symmetrical ripples sharing the

top surface with the track. Assuming an arbitrary “north”

A.J. Martin

Figure 4. Skenes Creek Track 1, NMV P208232. a. Overall top view of track in collected slab, with rock saw cuts evident on three sides, as well as

symmetrical and parallel ripple marks below track; scale = 5 cm. b. Close-up of track, showing raised relief and thin, laminated “platform” above

ripple marks; scale = 5 cm. c. Outline of track and parameters measured, with anterior triangle indicated (see Figure 2 for key); scale = 5 cm.

Figure 5. Skenes Creek Track 2, NMV P208232 (shared with Track 1). a. Overall top view of track in collected slab, with rock saw cuts evident

on five sides and fracture cutting across anterior part of track; scale (left) in centimetres and millimetres, b. Close-up of track, showing raised

relief and thin, laminated “platform” above ripple marks; scale = 5 cm. c. Outline of track and parameters measured, with anterior triangle

indicated (see Figure 2 for key); scale = 5 cm.

defined by the axis of the middle digit impression, ripple-crests

trend “northeast-southwest.” Ripples have amplitudes of 4-7

mm and wavelengths of 30-40 mm; ripple bedding was also

evident in cross-section. The track is preserved as a nearly flat,

1-2 mm thick positive-relief (raised) epichnion, but is filled

with sand texturally and compositionally identical to the main

host lithology. It is circumscribed and somewhat mimicked in

outline by a 4-8 mm thick quadrilateral platform between it

and the rippled surface. Bedding in the track and its platform is

mostly planar and laminated, although slight variations in

surface topography synch with underlying ripples.

The track is tridactyl, mesaxonic, and nearly equant in

length and width, at 106 mm long and 115 mm wide; this

results in a length:width ratio of 0.92. Outermost digit

impressions were 91 mm and 87 mm long (left and right,

respectively), and the central impression is the length of the

track, 106 mm. The right impression has a seemingly complete

outline, whereas the left is apparently expressed partially. The

anterior triangle length:width ratio is 0.37, with a base (width)

at 115 mm and length measured from that base of 43 mm.

Medial thicknesses of the two complete digit impressions are,

from middle to right, 34 and 36 mm. With an average of 35

mm, digit-impression widths are about 33% of footprint

length. Divarication between the outermost impressions is 83°,

which combines an angle of 37° between the left and middle,

and 46° between the right and middle. Both the middle and

right digit impressions narrow distally, but are subrounded at

their ends. An unidentified modern barnacle is attached to the

lower right-side edge of the track.

Skenes Creek Track 2. This probable dinosaur track (herein

called Skenes Creek Track 2) was also cataloged as part of

NMV P208232 (fig. 5, Appendix I). Based on its same specimen

A close look at Victoria's first known dinosaur tracks

number and having many identical sedimentary traits as the

Skenes Creek Track 1,1 conclude that it was also discovered by

Helmut Tracksdorf in 1989, then later collected from the same

marine-platform bedding plane by Museum of Victoria

personnel at the same time as Skenes Creek Track 1. This

supposition is supported by two adjacent rock-saw cuts at the

probable discovery site, which were relocated by Helmut

Tracksdorf in 2013 and Mike Cleeland in 2014 (Appendix II).

The track is in a pentagonal slab (cut by a rock saw) with

long dimensions of 24 x 21 cm; a fracture runs transversely

across the track (fig. 5a). The lithology is identical to that

hosting the Skenes Creek Track 1, consisting of a very fine-

fine, well-sorted lithic arenite, with a bed thickness of 30-40

mm. The top surface of the bed has six parallel, symmetrical,

and low-amplitude ripples underneath the track. Using an

arbitrary “north-south” defined by the medial axis of the track,

ripple-crests trend “northeast-southwest.” Ripples have

amplitudes of 5-7 mm and wavelengths of 25-40 mm; ripple

bedding was visible in cross-section. Also like Skenes Creek

Track 1, it is preserved as an almost-flat, positive-relief

epichnion, filled with sand texturally and compositionally

identical to that of its host rock. The track is circumscribed by

two levels, one about 5 mm above the rippled surface and

another inner and topmost level that is about 2 mm thick.

Again, like Skenes Creek Track 1, bedding in both upper

levels is mostly planar and laminated, with variations in

surface topography corresponding with underlying ripples.

Skenes Creek Track 2 is apparently tridactyl, with three

rounded points opposite another rounded end, which are

assumed as the anterior and posterior parts of the track,

respectively. Using this configuration, track length was 127

mm, whereas the width was 138 mm, resulting in a length: width

ratio of 0.92. Owing to vague outlines of presumed digit

impressions, width measurements were not attempted, but

lengths could be measured, yielding 102 mm for the left digit

impression and 103 mm for the right. Assuming this

configuration, the interdigital angles were 44° (left-middle)

and 27° (middle-right), for a divarication of 71°. The anterior

triangle length:width ratio was also calculable, yielding a

value of 0.31, with the base (width) the same as track width

(138 mm) and length measured from that base of 43 mm. The

posterior part of the track is slightly indented along its margin,

and the lower-level outline just below the measured part of the

track is bilobed. Three unidentified modern barnacles are

attached on the lowermost bedding surface, all to the lower

right of the track, but with one proximal and two more distal.

Interpretations and Discussion

The two most completely preserved tracks from Knowledge

Creek and Skenes Creek (Track 1) were very likely pes

impressions made by small ‘hypsilophodontid’ ornithopods

akin to Atlascopcosaurus,Fulgurotherium, or Leaellynasaura,

all of which are in the Eumeralla Formation (Rich and

Vickers-Rich, 1999; Rich et al., 2010). These trace fossils thus

constitute the first ornithopod footprints discovered in Lower

Cretaceous rocks of Victoria; all others found since the 1980s

have been attributed to theropods (Martin et al., 2007, 2012,

2014). The interpretation of tracks as ornithopod tracks is

based on track forms, sizes, and their preservation in strata

chronologically close to those containing the skeletal remains

of these ornithopods in the Eumeralla Formation (Rich and

Vickers-Rich, 2003; Kear and Bruce-Hamilton, 2011). The

second probable dinosaur track from Skenes Creek (Track 2),

although less definite in outline, is also likely from a small

ornithopod and comes from the same bed as the other track.

Furthermore, it may have been part of the same trackway as

the more completely defined track.

The two comparable tracks from Knowledge Creek and

Skenes Creek Track 1 are remarkably similar in size and form

(Figure 6). Both tracks are tridactyl, presumably reflecting

digits II-IV, with digit II-IV divarications of 85° (Knowledge

Creek) and 83° (Skenes Creek Track 1). Their lengths and

widths are nearly identical, their digit-impression widths vary

by only a few millimeters, and middle digit-impression widths

(digit III) are 27% and 33% of footprint length (Knowledge

Creek and Skenes Creek Track 1, respectively). Moreover,

their anterior triangle length: width ratios are nearly convergent,

at 0.40 for the Knowledge Creek track and 0.37 for Skenes

Creek Track 1. Their expression as positive-relief epichnia,

along with the other less completely preserved Skenes Creek

ichnite, is also noteworthy, implying their preservational

conditions may have been similar as well. However, because

the tracks are nearly symmetrical and do not show any

pressure-release structures related to movement ( sensu Martin

et al., 2012), I cannot identify digit impressions II or IV in

either tracks, nor state with certainty whether either represents

a right or left footprint.

Both quantitative and qualitative traits indicate the

Knowledge Creek and Skenes Creek Track 1 trace fossils are

ornithopod tracks. Length:width ratios of about 0.9, digit II-IV

divarications of 83-85°, anterior-triangle length-width ratios of

about 0.4, relatively thick digits, and rounded (blunt) ends on

digit impressions are all consistent with ornithopod tracks

(Moratallo et al., 1988; Lockley, 2009; Mateus and Milan, 2010;

Martin et al., 2012; Farlow et al., 2012). Relatively small sizes of

the tracks also agree with assessments of Eumeralla Formation

dinosaur assemblages, which are dominated by small

‘hysilophodontids’ (Rich and Vickers-Rich, 1999; Rich et al.,

2002). A possibly comparable ichnogenus in size and form to

these tracks is Dinehichnus isp. ( sensu Gierlinski et al, 2009,

fig. 7), reported from the Late Jurassic of Poland and North

America, and interpreted as that of a small ornithopod (Lockley

and Foster, 2006; Foster, 2007; Gierlinski et al., 2009; Lockley

et al., 2009). Furthermore, the Victoria tracks resemble

Iguanodontipus, which has been attributed to Early Cretaceous

iguanodontids (Sarjeant et al., 1998; Cobos and Gasco, 2012),

although these are often much larger (Castanera et al., 2013). In

Australia, Wintonopus of the Lower Cretaceous Winton

Formation (Queensland) is another small ornithopod track

comparable to the Victoria specimens (Thulborn and Wade,

1984; Romilio et al., 2013). In short, similarities between the

Victoria tracks and these ichnogenera affirm that their likely

tracemakers were small ornithopods. The tracks can be further

used to estimate tracemaker size via a footprint formula of 4.0 x

footprint length = hip height (Alexander, 1976; Henderson,

A.J. Martin

Figure 6. Skenes Creek Track 1 (left) and Knowledge Creek track (right) together, allowing for a side-by-side comparison. The slab holding the

Knowledge Creek track is slightly thicker (~10 mm) than the Skenes Creek slab, hence they are not being viewed on exactly the same horizon;

scale =10 cm.

2003). Using this formula, the Knowledge Creek and Skenes

Creek (Track 1) ornithopod tracemakers had hip heights of

about 42 cm (based on 10.6 x 4.0 = 42.4 cm). Owing to its poor

definition, the hip height of the trackmaker for Skenes Creek

Track 2 was not calculated, but if it does indeed belong to the

same trackway, it can be assumed as similar to that of Track 1.

Other dinosaur tracemakers that could have made tridactyl

tracks include small theropods, such as those interpreted from

the Eumeralla Formation dinosaur tracksite at Milanesia

Beach (Martin et al., 2012) or a single non-avian theropod

track from Dinosaur Cove (Martin et al., 2014). However, the

Milanesia and Dinosaur Cove theropod tracks were “thin¬

toed,” with middle digit widths of only 5-16% of footprint

lengths. In contrast, digits of the Knowledge Creek and Skenes

Creek tracks were more than twice as thick, at 27 and 33% of

footprint lengths (respectively). Furthermore, most of the

Milanesia Beach tracks had thin (sharp) clawmarks, which are

also characteristic of theropods (Martin et al ., 2012, and

references therein). Avian theropods (birds) were also

discounted as possible tracemakers for the Knowledge Creek

and Skenes Creek tracks for the same reasons, as well as for

having digital divarications of 83-85°; bird tracks more

typically have divarications of 95-120° (Lockley et al., 1992;

Falk et al., 2011; Martin et al., 2014). Both tracks also lack

evidence of a digit I impression, a trait noted in two similarly

sized avian tracks from the Eumeralla Formation at Dinosaur

Cove (Martin et al., 2014).

The positive-relief (raised) expression of all three tracks,

yet on bed tops as epichnia, is unusual for most fossil tracks.

Fossil tracks are normally preserved either as depressions

(negative-relief epichnia) or as natural casts on bed bottoms

(positive-relief hypichnia) (Lockley, 1991; Farlow et al., 2012).

Because all three specimens were recovered from marine

platforms eroded by tides and waves, their positive-relief

preservation implies that sediment filling the tracks was better

cemented - and hence more resistant to weathering - than

their host rocks. This differential cementation and weathering

that resulted in convex dinosaur tracks on bed tops was also

noted for large theropod tracks in the Wonthaggi Formation

(Aptian) at the Flat Rocks (“Dinosaur Dreaming”) dig site

(Martin et al., 2007). An uneven fill and cementation probably

contributed to the vague definition of Skenes Creek Track 2, in

which the “true track” is buried below the outwardly expressed

positive epichnion.

Track surfaces are mainly uniform, but the Knowledge

Creek track contained three oval-outlined protuberances,

which I interpret as cross-sections of invertebrate burrows

(figs. 3a,b). The burrows would have been made in an originally

thicker bed composed of the same medium-coarse sand that

filled the track. A thin vertical structure with the same sand

fill underneath and toward the rear of the track is also likely a

burrow, but one in the underlying host lithology and passively

filled by sand from above. Owing to insufficient details,

ichnotaxa were not assigned to these burrows.

Skenes Creek Track 2, despite its larger dimensions and less

definite outline, is similar enough to its companion track from

the same site that it is also interpreted as a dinosaur track, and

probably that of a small ornithopod. This supposition is based on

its tridactyl form, rounded ends to the probable digit impressions,

a length:width ratio of 0.92, and an anterior-triangle length:width

ratio of 0.31, which again are consistent with ornithopod

tracemakers (Lockley, 2009). One of the “interdigital” angles on

Skenes Creek Track 2 roughly corresponded with that of Track 1

(44° versus 46°, respectively); however, divarication was notably

different (71° versus 83°, respectively). As mentioned before, the

original depressions (“true track”) for both Tracks 1 and 2 are

likely underneath the currently expressed positive-relief outline,

filled with sand that later cemented differently from the

A close look at Victoria's first known dinosaur tracks

surrounding substrate. These depressions may even cut across

the ripples underneath the track or deformed surrounding

sediments so that the subsequent fill and cementation of that fill

affected the outward appearance and weathering of the tracks on

the same marine platform.

Although no information was recorded about the spatial

relationship of the Skenes Creek tracks on the marine platform

when and where they were recovered in 1989, the rock-saw

cuts corresponding to their locations were found by Helmut

Tracksdorf in June 2014, which was confirmed by Michael

Cleeland in February 2015 (Appendix II). Interestingly, the

rectangular outlines of the rock-saw cuts are directly aligned

and in a sandstone bed with low-amplitude ripples, with ripple

crests oriented obliquely relative to the outlines. Furthermore,

both tracks have similar forms and are atop ripples oriented

the same with respect to footprint directions, i.e., “northeast-

southwest” with tracks pointing toward an arbitrary “north.”

Consequently, these tracks likely belong to the same trackway

and were made by the same individual ornithopod. If so, these

would constitute the first discovered dinosaur trackway in

Victoria, usurping a small-theropod trackway discovered at

Milanesia Beach in 2010 (Martin et ah, 2012). To test this

preliminary interpretation, the exact distance between

incisions will need to be measured, and rock-saw outlines

should be compared to the shapes and orientations of the two

recovered slabs. Conversely, if the tracks are not aligned (e.g.,

point in opposite directions) and the collected slabs do not

correspond to the outlines, they can be reasonably attributed

to separate trackways made by similar tracemakers.

Unfortunately, all three tracks are isolated specimens taken

out of context from their original field exposures in 1980

(Knowledge Creek) and 1989 (Skenes Creek). Thus very little

additional information can be said about the palaeoenvironments

trodden by their ornithopod tracemakers. The Eumeralla

Formation is interpreted as a series of fluvial channel-fill,

overbank, and floodplain facies deposited in circumpolar rift

valleys, with less common alluvial or lacustrine facies (Bryan et

al., 1997; Tosolini et al., 1999; Vickers-Rich et ah, 1999). Given

this broad framework, the most probable palaeoenvironments

for dinosaurs making preservable tracks would have been point

bars or floodplains, which are common sites for dinosaur track

preservation (Martin, 2014). If the Skenes Creek trackmaker

was walking on a floodplain, ripples underneath these tracks

might have been current ripples, exposed after water flowed

over that surface. Similar modern occurrences of current ripples

later cross-cut by vertebrate tracks, providing a possible

analogue for the Skenes Creek tracks, were described from a

Arctic point bar in Alaska (Martin, 2009b). The largest

assemblage of dinosaur tracks from the Eumeralla Formation -

found at Milanesia Beach and described from two separate slabs

of the same sandstone bed - was also in what were likely

floodplain sandstones (Martin et al., 2012). Indeed, the

circumpolar setting of these river valleys during the Early

Cretaceous meant that dinosaur tracks might have been made

and preserved only seasonally, from spring through fall (Martin

et al., 2012). If so, this limiting factor may account for the

relative rarity of dinosaur tracks and other trace fossils in Lower

Cretaceous strata of Victoria (Martin et al., 2012).

Conclusions

The first known dinosaur tracks from Victoria may be small

and few, but nonetheless carry useful information about Early

Cretaceous dinosaurs in Victoria. For one, the Knowledge

Creek and Skenes Creek tracks are from localities where no

dinosaur bones are yet known, therefore confirming a

dinosaurian presence at each of these places. Secondly, the

tracks demonstrate that dinosaurs - specifically small

ornithopods - actually lived in the palaeoenvironments of

these places. In contrast, most dinosaur body fossils in Victoria,

such as those from the Flat Rocks (“Dinosaur Dreaming”) site

at Inverloch and Dinosaur Cove, were likely transported and

deposited in fluvial channels (Rich and Vickers-Rich, 2000;

Rich et al., 2003). Lastly, these tracks are the first discovered

Early Cretaceous ornithopod tracks from Victoria, and the

Skenes Creek tracks may represent the first discovered dinosaur

trackway in Victoria. These finds thus supplement comparatively

abundant body fossils of ‘hypsilophodontids’ in the Wonthaggi

and Eumeralla Formations.

The preservation of these small ornithopod tracks as

positive-relief epichnia, as well as those of large theropod

tracks in the Wonthaggi Formation (Martin et al., 2007), also

may be typical modes of preservation for dinosaur tracks in

Lower Cretaceous strata of Victoria. Hence future researchers

scanning bedding planes of the Wonthaggi and Eumeralla

Formations might adjust their search images for raised tracks

on bed tops, rather than just depressions. Furthermore, less

definite forms of dinosaur, such as that of Skenes Creek Track

2 , should not be so easily ignored or dismissed once found.

Given all of these insights, I have every confidence that more

dinosaur tracks will be discovered, whether from ornithopods,

theropods, or other tetrapod taxa whose trace fossil records

are not yet known from this otherwise palaeontologically

well-studied area of Australia.

Acknowledgements

I am extremely grateful to Thomas (Tom) Rich and Patricia

(Pat) Vickers-Rich for their guidance, mentorship, and

friendship over the years, starting with a sabbatical I enjoyed

at Monash University in 2006 and visits to Australia since

then. I thank Erich Fitzgerald (Museum Victoria) for asking

me to write this manuscript, as well as Lisa Buckley and

Anthony Romilio for their insightful and helpful reviews.

Appreciation is extended to David Pickering, Lesley and

Gerry Kool, Michael and Naomi Hall, and Michael Cleeland

for their time in the field with me, as well as educating this

inexperienced Yank whenever he dropped in for a too-brief

glance at their lovely Lower Cretaceous rocks. Helmut

Tracksdorf and Michael Cleeland also were invaluable in

helping to document the original discovery site of the Skenes

Creek tracks. The Emory University Faculty Travel Fund and

the Center for International Programs Abroad assisted with

some of my travel expenses during various visits. Last but not

least, I am indebted to my wife, Ruth, nicknamed “Lefty” by

her fellow southpaw Tom, for her sinistral support during

much of the research.

A.J. Martin

References

Alexander, R.M. 1976 Estimates of speeds of dinosaurs. Nature 262:

129-130.

Benson, R.B., Rich, T.H., Vickers-Rich, P. and Hall, M. 2012.

Theropod fauna from Southern Australia indicates high polar

diversity and climate-driven dinosaur provinciality. PLoS ONE

7(5): e37122.

Bryan, S.E., Constantine, A.E., Stephens, C.J., Ewart, A., Schon,

R.W., and Parianos, J. 1997. Early Cretaceous volcano¬

sedimentary successions along the eastern Australian continental

margin: implications for the break-up of eastern Gondwana. Earth

and Planetary Science Letters 153: 85-102.

Castanera, D., Pascual, C., Razzolini, N.L., Vila, B., Barco, J.L., and

Canudo, J.I. 2013. Discriminating between medium-sized

tridactyl trackmakers: Tracking ornithopod tracks in the base of

the Cretaceous (Berriasian, Spain). PLoS ONE 8(11): e81830.

Cobos, A. and Gasco, F. 2012 Presencia del icnogenero Iguanodontipus

en el Cretacico Inferior de la provincia de Teruel (Espana).

Geogaceta 52: 185-188.

Farlow, J.O., Chapman, R.E., Breithaupt, B., and Matthews, N. 2012.

The scientific study of dinosaur footprints. Pp. 713-759 in: Brett-

Surman, M.K., Holtz, T.R., Jr., and Farlow, J.O. (eds). The

Complete Dinosaur (2nd Edition). Indiana

University Press: Bloomington, Indiana. 1112 pp.

Foster, J.R. 2007. Jurassic West: The Dinosaurs of the Morrison

Formation and Their World. Indiana University Press.

Bloomington, Indiana. 389 pp.

Gierlinski, G.D., Niedzwiedzki, G. and Nowacki, P. 2009. Small

theropod and ornithopod tracks in the Jurassic of Poland. Acta

Geologica Polonica 59: 221-234.

Henderson, D.M. 2003. Footprints, trackways, and hip heights of

bipedal dinosaurs: testing hip height predictions with computer

models. Ichnos 10: 99-114.

Kear, B.P, and Hamilton-Bruce, R.J. 2011. Dinosaurs in Australia:

Mesozoic Life from the Southern Continent. CSIRO Publishing:

Collingwood, Victoria. 190 pp.

Lockley, M.G. 1991. Tracking Dinosaurs: A New Look at an Ancient

World. Cambridge University Press: Cambridge, U.K. 264 pp.

Lockley, M.G., and Foster, J.R. 2006. Dinosaur and turtle tracks from

the Morrison Formation (Upper Jurassic) of Colorado National

Monument, with observations on the taxonomy of vertebrate

swim tracks. Pp. 193-198 in: Foster, J.R., and Lucas, S.G. (eds).

Paleontology and Geology of the Upper Jurassic Morrison

Formation, New Mexico Museum of Natural History and Science

Bulletin , 36: 249 pp.

Lockley, M.G., Yang, S.Y., Matsukawa, M., Fleming, F. and Lim, S K.

1992. The track record of Mesozoic birds: evidence and

implications. Philosophical Transactions of the Royal Society of

London 336: 113-134.

Lockley, M.G. 2009. New perspectives on morphological variation in

tridactyl footprints: clues to widespread convergence in

developmental dynamics. Geological Quarterly 53: 415-432.

Martin, A.J., 2009a. Dinosaur burrows in the Otway Group (Albian)

of Victoria, Australia, and their relation to Cretaceous polar

environments. Cretaceous Research 30: 1223-1237.

Martin, A.J., 2009b. Neoichnology of an Arctic fluvial point bar.

North Slope, Alaska. Geological Quarterly 53: 383-396.

Martin, A.J. 2014. Dinosaurs Without Bones: Dinosaur Lives Revealed

by Their Trace Fossils. Pegasus Press: New York. 460 pp.

Martin, A.J., Vickers-Rich, P., Rich, T.H. and Kool, L. 2007. Polar

dinosaur tracks in the Cretaceous of Australia: though many were

cold, few were frozen. Journal of Vertebrate Paleontology 27

(Supplement3): 112A.

Martin, A.J., Rich, T.H., Hall, M., Vickers-Rich, P. and Vazquez-

Prokopec, G. 2012. A polar dinosaur-track assemblage from the

Eumeralla Formation (Albian), Victoria, Australia. Alcheringa 36:

171-188.

Martin, A.J., Vickers-Rich, P, Rich, T.H., and Hall, M. 2014. Oldest

known avian footprints from Australia: Eumeralla Formation

(Albian), Dinosaur Cove, Victoria. Palaeontology 57: 7-19.

Mateus, O., and Milan, J. 2010. A diverse Upper Jurassic dinosaur

ichnofauna from central-west Portugal. Lethaia 43: 245-257.

Moratalla, J.J., Sanz, S.L., and Jiminez, S., 1988. Multivariate analysis

on Lower Cretaceous dinosaur footprints: discrimination between

ornithopods and theropods. Geobios 21: 395-408.

Rich, T.H., and Rich, P.V. 1989. Polar dinosaurs and biotas of the Early

Cretaceous of southeastern Australia. National Geographic

Research 5: 15-53.

Rich, T.H., and Vickers-Rich, P. 1999. The Hypsilophodontidae from

southeastern Australia. Pp. 167-180 in: Tomada, Y., Rich, T.H., and

Vickers-Rich (eds.). Proceedings of the Second Gondwana Dinosaur

Symposium. National Science Museum Monographs 15. 296 pp.

Rich, T.H., and Vickers-Rich, P. 2000. Dinosaurs of Darkness. Indiana

University Press: Bloomington, Indiana. 222 pp.

Rich, T.H., and Vickers-Rich, P. 2003. A Century of Australian

Dinosaurs. Queen Victoria Museum and Art Gallery: Launceston,

Tasmania. 124 pp.

Rich, T.H., Vickers-Rich, P., and Gangloff, R.A. 2002. Polar dinosaurs.

Science 295: 979-980.

Rich, T.H., Galton, P.M., and Vickers-Rich, P. 2010. The holotype

individual of the ornithopod dinosaur Leaellynasaura

amicagraphica Rich & Rich, 1989 (late Early Cretaceous, Victoria,

Australia). Alcheringa 34: 385-396.

Rich, P.V., Rich, T.H., Wagstaff, B.E., McEwen-Mason, J., Douthitt,

C.B., Gregory, R.T., and Felton, E.A. 1988. Evidence for low

temperatures and biologic diversity in Cretaceous high latitudes of

Australia. Science 242: 1403-1406.

Romilio, A., Tucker, R.T., and Salisbury, S.W. 2013. Reevaluation of the

Lark Quarry dinosaur Tracksite (late Albian-Cenomanian Winton

Formation, central-western Queensland, Australia): no longer a

stampede? Journal of Vertebrate Paleontology 33: 102-120.

Sarjeant, W.A.S., Delair, J.B., and Lockley, M.G. 1998. The footprints

of Iguanodon: A history and taxonomic study. Ichnos 6: 183-202.

Thulborn, R.A. 1990. Dinosaur Tracks. Chapman & Hall: London. 410

pp.

Thulborn, R.A., and Wade. M. 1984. Dinosaur trackways in the Winton

Formation (Mid-Cretaceous) of Queensland. Memoirs of the

Queensland Museum 21: 413-517.

Tosolini, A.-M.P., McLoughlin, A., and Drinnan, A.N. 1999.

Stratigraphy and fluvial sedimentary facies of the Neocomian

lower Strzelecki Group, Gippsland Basin, Victoria. Australian

Journal of Earth Sciences 46: 951-970.

Vickers-Rich, P, Rich, T.H., and Constantine, A. 1999. Environmental

setting of the polar faunas of southeastern Australia and adaptive

strategies of the dinosaurs. Pp. 181-195 in: Tomada, Y., Rich, T.H.,

and Vickers-Rich (eds). Proceedings of the Second Gondwana

Dinosaur Symposium, National Science Museum Monographs 15.

296 pp.

Appendices

Appendix I. Specimen Label Information

Specimen label information for Knowledge Creek track:

P.159790; Dinosaur track, Otway Group, Knowledge Creek,

Victoria on wave platform about 100 m east of mouth of

A close look at Victoria's first known dinosaur tracks

Knowledge Creek. T. Rich Exp., 18-12-1980 [December

18, 1980]

Specimen Label Information for Skenes Creek tracks:

P.208232; T.H. Rich Exp., 18-3-1989 [March 18, 1989] Skenes

Creek (Catalog 149). Locality: Shore platform, Skenes Creek.

Appendix II. Discovery of the Skenes Creek Tracks

The specimen label for the two Skenes Creek tracks does not

credit their original discoverer, and repeated inquiries posed to

personnel at Museum Victoria and long-time volunteers did not

result in anyone taking credit for finding them. So I was

gratified to learn in 2013 that credit for their discovery should

go to Helmut Tracksdorf, a geologist who lived in Victoria near

Skenes Creek at the time of their discovery.

In October 2013, Mr. Tracksdorf read a blog post written by

me referring to the track. Mr. Tracksdorf sent me an e-mail

message, received on October 20, 2013, revealing that he was

the person who found the tracks. According to his message, he

then reported these tracks and their location to Museum

Victoria personnel. Several months after reporting them, he did

not receive a confirmation from Museum Victoria on whether

or not the tracks were recovered. However, he later visited the

site and saw where they had been cut out of the marine platform.

The e-mail message was sent by Helmut Tracksdorf and

received by me (Anthony J. Martin) at 5:43 p.m. on October 20,

2013. The full, verbatim text of the e-mail is available for

reading with permission of both Mr. Tracksdorf and myself.

Later, on July 9, 2014, Trackdorf contacted me again via

e-mail with more information about the probable original

location of the tracks; he copied Thomas Rich, David

Pickering, Lesley Kool, and Michael Cleeland onto this

message. In this message, he described the rock-saw cuts on

the marine platform as 50-100 m west of Browns Creek (east

of both Skenes Creek and Petticoat Creek) and about 50 m

south of the Great Ocean Road. Along with this description,

Tracksdorf provided a Google Earth image of the locality, as

well as photographs of the site and rock-saw cuts in the marine

platform, with the photos taken in June, 2014 by Trackdorf’s

brother (name not given). These descriptions aided me in

figuring the approximate latitude-longitude coordinates of the

tracks. On February 17, 2015, Cleeland and his wife (Pip)

stopped by the location to look for the rock-saw cuts on the

marine platform, relocated them, and photographed the rock-

saw cuts; these were aligned with one another and in a rippled

sandstone very similar to those of the Skenes Creek tracks

(NMV P208232). On March 2,2015, he sent these photographs

to Rich, Pickering, Kool, Trackdorf, and me. Given this

confirmation, we were satisfied that this is indeed the discovery

site of the tracks.

In deference to their long-established nickname as the

“Skenes Creek tracks,” I recommend retaining this location

designation, rather than adopting the more geographically

appropriate “Browns Creek tracks.” I also suggest that future

researchers searching for more such dinosaur tracks in this area

might concentrate their efforts on rippled sandstones in the

marine platform between Petticoat Creek and Browns Creek.

Memoirs of Museum Victoria 74:73-79 (2016) Published 2016

ISSN 1447-2546 (Print) 1447-2554 (On-line)

http://museumvictoria.com.au/about/books-and-journals/journals/memoirs-of-museum-victoria/

Organic geochemistry of a high-latitude Lower Cretaceous lacustrine sediment

sample from the Koonwarra Fossil Beds, South Gippsland, Victoria, Australia

Michael L. Tuite* David T. Flannery and Kenneth H. Williford

NASA Jet Propulsion Laboratory, California Institute of Technology, Pasadena, California, USA

* To whom correspondence should be addressed. E-mail: mtuite@jpl.nasa.gov

Abstract Tuite, M.L., Flannery , D.T., and Williford, K.H. 2016. Organic geochemistry of a high-latitude Lower Cretaceous

lacustrine sediment sample from the Koonwarra Fossil Beds, South Gippsland, Victoria, Australia. Memoirs of Museum

Victoria 74: 73-79.

The Koonwarra Fossil Beds are widely recognized for their high-fidelity preservation of freshwater/terrestrial

vertebrate and invertebrate fossils. A preliminary investigation suggests that organic biomarkers are also exceptionally well

preserved and could contribute significantly to understanding the ecology of this ancient lake system. Solvent-extractable

organic matter was collected from a single feldspathic siltstone/mudstone sample and analyzed using gas chromatography-

mass spectrometry (GC-MS). The distribution of n-alkanes suggests a significant input of terrestrial plant material into the

lake. The very low ratio of eukaryotic steranes to bacterial hopanes may reflect the decomposition of abundant plant material

in the lake. Polycyclic aromatic hydrocarbons may record wildfire activity in the surrounding watershed.

Keywords Koonwarra fossil beds, Cretaceous, paleolimnology, biomarkers.

Introduction

The Lower Cretaceous Koonwarra Fossil Beds in South

Gippsland, Victoria, Australia (fig. 1), were discovered during

road works in 1961 (Jell and Duncan, 1986). They are thought

to be a freshwater lacustrine deposit of Barremian-Aptian age

based on plant and animal fossils, palynology, and fission

track dating (Douglas, 1969, 1974; Dettmann, 1986; Drinnan

and Chambers, 1986; Jell and Duncan, 1986). Paleogeographic

reconstructions place southern Australia well within the

Antarctic Circle at this time, and the S 18 0 values of early

diagenetic carbonate concretions imply mean annual

temperatures of ~5°C, despite a generally warm Cretaceous

climate (Embleton and McElhinny, 1982; Rich et al., 1988).

The Koonwarra Fossil Beds are known for high-fidelity

preservation of freshwater/terrestrial fossils, including several

fish groups, insects, crustaceans, bird feathers and a freshwater

xiphosuran (e.g. Riek, 1970; Riek and Gill, 1971; Jell and

Duncan, 1986; Vickers-Rich, 1991; Krzeminski et al., 2015).

Waldman (1971) interpreted varves and multiple horizons of

fish fossils to be the result of winter ice covering a shallow

lake and causing anoxia, mass fish kills, and the settling of

clay from suspension (which would have further prevented the

decomposition of covered carcasses). Warmer conditions

during the spring might also be expected to cause

decomposition and re-floatation of fish carcasses in a shallow-

water environment (Wilson, 1977). Alternatively, Elder and

Smith (1988) proposed a stratified lake model, wherein fish

carcasses sunk to deep, cold waters, where scavenging and

decomposition were inhibited by oxygen depletion. The

formation of the fish beds could also be related to toxic

summer algal blooms, as suggested by McGrew (1975) for the

Eocene Green River Formation.

In this brief report, we present the results of a preliminary

study of the organic geochemistry of a sediment sample

collected from the Koonwarra Fossil Beds in early 2013. The

sample analysed here was collected from the fossil fish beds;

approximately 5 m from the bottom of the unit. It is a

feldspathic siltstone/mudstone laminated on a mm-cm scale.

Fish fossils are abundant in this zone (fig. 2). Our results reveal

a well-preserved biomarker record in the Koonwarra deposit

and suggest that further geochemical investigation is likely to

yield significant insights into Cretaceous paleolimnology and

the exceptional taphonomy at Koonwarra. These data also

suggest that terrigenous organic matter, deriving both from

higher plants and perhaps soil bacteria, was an important

supplement to aquatic primary production as a trophic resource

for lake consumers.

Materials and Methods

Geochemical analyses were performed in the

Astrobiogeochemistry Laboratory (abcLab) at the Jet

Propulsion Laboratory. A sample without visible macrofossils

was powdered using an alumina grinding dish in a model 8530

Shatterbox (SPEX Sample Prep). A ~0.5 g aliquot was

M.L. Tuite, D.T. Flannery and K.H. Williford

decarbonated with excess IN HC1, washed to neutrality with

ultrapure (<18 M£2) deionized water, and dried at 50°C for 48

hours. For determination of total organic carbon (TOC),

approximately 30 mg of dried sample was weighed in a tin

capsule and combusted at 980°C in a Costech 4010 elemental

analyzer. The resulting C0 2 was chemically dried and

transferred via He carrier flow to a Delta V Plus (Thermo)

isotope ratio mass spectrometer. The mass of organic C was

determined by comparison of the area of the mass 44

chromatogram of the sample with the regression of a series of

acetanilide standards of known C content (r 2 = 0.999).

Thirty grams of the powdered sample were extracted for

48 hours using a soxhlet apparatus in a DCM:methanol (9:1

v:v) mixture. The extraction yielded 0.57 mg of total lipid

extract that was subsequently separated into saturate, aromatic,

and polar fractions using small column chromatography

(Bastow et al., 2007). The three fractions were eluted using

«-hexane, w-hexane:DCM (7:3 v:v), and DCM:methanol (1:1

v:v), respectively. Saturate and aromatic fractions were

analyzed via gas chromatography-mass spectrometry

(GC-MS) using a Trace GC Ultra (Thermo) connected to an

ISQ Series quadrupole MS (Thermo).

Results

The TOC value of the sample is 0.4% and the yield of

extractable lipids is 0.02 mg per gram of sample or 4.75 mg per

gram TOC. The saturate fraction total ion chromatogram

(fig. 3) has a small, unresolved complex mixture (C 15 -C 23 )

Organic geochemistry of sediment from the Cretaceous Koonwarra Beds, Victoria

Figure 2: An uncommon example of disarticulation of a fish carcass, collected during an excavation of the Koonwarra Fossil Beds led by Tom

Rich in 2013. This specimen was collected approximately 5 m from the bottom of the unit (defined here as the first > 20 cm thick unit of green

siltstone/mudstone; the underlying rocks are predominantly cross-bedded, fluviatile arkosic sandstone).

above the baseline suggesting that biodegradation of the fossil

organic matter has been minimal. The m/z 85 mass

chromatogram reveals a stepwise increase in the abundance of

short-chained w-alkanes (C 15 -C 22 ) with increasing molecular

weight likely associated with aquatic sources such as algae.

The longer chained w-alkanes (C 23 -C 31 ) that are indicative of

terrestrial higher-plant input (Bourbonniere and Meyers,

1996) show a clear odd-over-even carbon number

predominance (OEP = 1.48 where 1.0 means no predominance)

(Scalan and Smith, 1970). The OEP preserves a distribution

characteristic of terrestrial plant epicuticular waxes and

indicates that thermal alteration of the sample that would have

diminished the uneven distribution has been minimal (Peters

et al., 2005).

A measure of the relative contributions of terrestrial and

aquatic organic matter sources is the terrestrial/aquatic ratio

(TAR; Bourbonniere and Meyers, 1996) calculated using

w-alkane peak areas:

TAR = (C 15 + C 17 + C 19 ) / (C 27 + C 29 + C 31 ).

Although the ratio is most useful when comparing changes

in organic matter sources along a stratigraphic series of

samples, the value we determined (TAR = 4.7) indicates a

significant higher-plant contribution to the total carbon flux to

the sediment. This is supported by the ratio of the regular

isoprenoids pristane (Pr) and phytane (Ph) that are usually

understood to have derived from the phytol tail of the

chlorophyll molecule (Brooks et al., 1969). The ratio is

influenced by the source organic matter as well as by the redox

state of the environment of deposition. The value for Pr/Ph for

this sample of 3.1 is diagnostic of a predominantly terrestrial

organic matter source deposited under oxic conditions (Peters

et al., 2005).

Hopanes are pentacyclic triterpenoids that are derived

predominantly from cell wall lipids of prokaryotes (Ourisson

et al., 1979). Steranes derive from lipids found only in

eukaryotes including microbial photoautotrophs and

metazoans (Chapman and Schopf, 1983). A sterane/hopane

ratio of 0.03 was calculated using the summed peak areas of

17a-hopane isomers and C 27 , C 28 , and C 29 sterane isomers (fig.

4). Low sterane/hopane ratios are typically indicative of either

a dominantly terrigenous organic matter source or

biodegradation of organic matter (Tissot and Welte, 1984;

Peters et al., 2005). The small contribution of unresolved

complex mixture to the saturate fraction indicates that post-

M.L. Tuite, D.T. Flannery and K.H. Williford

Figure 3: Saturate fraction total ion chromatogram and m/z 85 mass chromatogram showing distribution and relative abundances of n-alkanes

and isoprenoids pristane and phytane.

Organic geochemistry of sediment from the Cretaceous Koonwarra Beds, Victoria

17D-hopanes

c c c

27 > ^ 28 ’ 29

regular steranes

m/z 217

Figure 4: Partial m/z 191 and 217 mass chromatograms used in calculation of sterane/hopane ratio. A ratio of 0.03 indicates that a very significant

proportion of overall biomass in the lake was derived from bacteria.

m/z 178

Figure 5: Partial m/z 178, 202 and 228 mass chromatograms showing the distribution of common polycyclic aromatic hydrocarbons (PAH) in the

aromatic fraction.

M.L. Tuite, D.T. Flannery and K.H. Williford

depositional biodegradation was probably not significant.

Possible sources of the abundant hopanes include soil bacteria

from the surrounding watershed and the aquatic bacterial

heterotrophs responsible for the decomposition of terrestrial

plant matter.

The aromatic organic fraction contains a range of

polycyclic aromatic hydrocarbons (PAH). PAHs in sediments

may derive from incomplete combustion of wood and plant

matter (Blumer and Youngblood, 1975) or from diagenesis of

organic precursors (Jiang et al., 1998). Figure 5 shows three

mass/charge-specific chromatograms that illustrate the

presence of potentially combustion-related PAHs

phenanthrene, fluoranthene, pyrene, and benzo[a]anthracene,

chrysene, and triphenylene. Venkatesan and Dahl (1989)

observed high concentrations of these and other pyrosynthetic

PAHs at a variety of Cretaceous/Tertiary boundary sites that

they interpreted as evidence of extensive wildfires. Similar

distributions of PAHs at the Permian/Triassic boundary also

suggest that wildfires were unusually common at the time of

the end-Permian extinction (Nabbefeld et al., 2010).

Conclusions

The Koonwarra beds provide fossil evidence of a complex,

high latitude lacustrine ecosystem. Organic geochemical

evidence suggests that supplementation of aquatic trophic

resources by terrigenous organic matter in the form of soil

bacteria and plant tissues washed into the lake may have

played an important role in sustaining that ecosystem. This

supplementation may have been particularly significant given

the impact of the high annual variability of solar insolation on

aquatic primary production at high latitudes. The presence of

combustion-related PAHs indicates an atmospheric p0 2

capable of sustaining combustion and confirms the presence of

forested land cover and soils in the vicinity of the lake.

In addition to illuminating the environmental context of

the extensive fossil record at Koonwarra, the sample also

clearly indicates that the state of organic matter preservation

in the unit is highly conducive to further integration of

paleontological and geochemical evidence in assembling a

comprehensive understanding of an exceptionally well-

preserved ecosystem. Further work is likely to yield insights

into redox conditions and other environmental factors that led

to the high-fidelity preservation of fossils.

Acknowledgements

We are foremost indebted to Tom Rich, who introduced us

to the Koonwarra Fossil Beds and led the excavations in

2013. Simon George provided a thoughtful and thorough

review. David Flannery thanks Abigail Allwood for

supporting his interest in this area. The research was carried

out at the Jet Propulsion Laboratory, California Institute of

Technology, under a contract with the National Aeronautics

and Space Administration.

Government sponsorship acknowledged.

References

Bastow, T.P., van Aarssen, B.G.K., and Lang, D. 2007, Rapid small-

scale separation of saturate, aromatic and polar components in

petroleum. Organic Geochemistry 38(8): 1235-50.

Blumer, M., and Youngblood, W.W. 1975. Polycyclic Aromatic-

Hydrocarbons In Soils And Recent Sediments. Science 188(4183):

53-5.

Bourbonniere, R.A., and Meyers, P.A. 1996. Sedimentary geolipid

records of historical changes in the watersheds and productivities

of Lakes Ontario and Erie. Limnology and Oceanography 41(2):

352-9.

Brooks, J.D., Gould, K., and Smith, J.W. 1969. Isoprenoid

Hydrocarbons in Coal and Petroleum Nature 222(5190): 257-259.

Chapman, D.J., and Schopf, J.W. 1983. Biological and biochemical

effects of an aerobic environment. Pp. 302-320 in: Schopf, J.W.

(ed). Earth’s Earliest Biosphere. Princeton University Press,

Princeton, NJ.

Dettmann, M.E. 1986. Early Cretaceous palynoflora of subsurface

strata correlative with the Koonwarra Fossil Bed, Victoria.

Association of Australasian Palaeontologists, Memoir 3: 79-110.

Douglas, J.G. 1969. The Mesozoic floras of Victoria, Parts 1 and 2.

Geological Survey of Victoria Memoire 28: 1-310.

Douglas, J.G. 1974. The Mesozoic floras of Victoria, Part 3. Geological

Survey of Victoria Memoire 28:1-185.

Drinnan, A.N., and Chambers, T.C. 1986. Flora of the Lower

Cretaceous Koonwarra Fossil Bed (Korumburra Group), South

Gippsland, Victoria. AssociationofAustralasianPalaeontologists,

Memoir 3: 1-77.

Elder, R.L., and Smith, G.R. 1988. Fish taphonomy and environmental

inference in paleolimnology. Palaeogeography,

Palaeoclimatology, Palaeoecology 62.1: 577-592.

Embleton, B.J., and McElhinny, M.W. 1982. Marine magnetic

anomalies, palaeomagnetism and the drift history of Gondwanaland.

Earth and Planetary Science Letters 58(2): 141-150.

Jell, P.A., and Duncan, P.M. 1986. Invertebrates, mainly insects, from

the freshwater Lower Cretaceous, Koonwarra Fossil Bed

(Korumburra Group), South Gippsland, Victoria. Association of

Australasian Palaeontologists, Memoir 3: 111-205.

Jiang, C.Q., Alexander, R., Kagi, R.I., and Murray, A.P 1998.

Polycyclic aromatic hydrocarbons in ancient sediments and their

relationships to palaeoclimate. Organic Geochemistry 29(5-7):

1721-35.

Krzeminski, W., et al. 2015. Revision of the unique Early Cretaceous

Mecoptera from Koonwarra (Australia) with description of a new

genus and family. Cretaceous Research 52:501-506.

McGrew, P. O. 1975. Taphonomy of Eocene fish from Fossil Basin,

Wyoming. Fieldiana 33(14): 257-270.

Nabbefeld, B., Grice, K., Summons, R.E., Hays, L.E., and Cao, C.

2010. Significance of polycyclic aromatic hydrocarbons (PAHs) in

Permian/Triassic boundary sections. Applied Geochemistry 25:

1374-1382.

Ourisson, G., Albrecht, P., and Rohmer, M. 1979. Hopanoids -

Palaeochemistry and Biochemistry of a Group of Natural-

Products. Pure and Applied Chemistry 51(4): 709-29.

Peters, K.E., Walters, C.C., and Moldowan, J.M. 2005. The Biomarker

Guide: Biomarker and Isotopes in Petroleum Exploration and

Earth History. Cambridge University Press, New York.

Rich P.V., Rich, T.H., Wagstaff, B.E., Mason, J.M., Douthitt, C.B.,

Gregory, R.T., and Felton, E.A. 1988. Evidence for low

temperatures and biologic diversity in Cretaceous high latitudes

of Australia. Science 242: 1403-6.

Riek, E.F. 1970. Lower Cretaceous fleas. Nature 227: 746-747

Organic geochemistry of sediment from the Cretaceous Koonwarra Beds, Victoria

Riek, E. F., and Gill, E.D. 1971. A new xiphosuran genus from Lower

Cretaceous freshwater sediments at Koonwarra, Victoria,

Australia. Palaeontology 14.2: 206-210.

Scalan, R.S., and Smith, J.E. 1970. An Improved Measure of Odd-

Even Predominance in Normal Alkanes of Sediment Extracts and

Petroleum. Geocliimica et Cosmochimica Acta 34(5): 611-620.

Tissot, B.P., and Welte, D.H. 1984. Petroleum Formation and

Occurrence. Springer-Verlag, New York.

Venkatesan, M.I., and Dahl, J. 1989. Organic Geochemical Evidence

for Global Fires at the Cretaceous Tertiary Boundary, Nature

338(6210): 57-60.

Vickers-Rich, P. 1991. The Mesozoic and Tertiary history of birds on

the Australian Plate. Pp. 721-808 in: Rich, P.V., Monaghan, J.M.,

Baird, R.F., and Rich, T.H. (eds.). Vertebrate Palaeontology

of Australasia. Pioneer, Victoria.

Waldman, M. 1971. Fish from the freshwater Lower Cretaceous of

Victoria, Australia, with comments on the paleo-environment.

Special Papers Palaeontology 9(1): 1-124.

Wilson, M.V.H. 1977. Paleoecology of Eocene lacustrine varves at

Horsefly, British Columbia. Canadian Journal of Earth Sciences

14: 953-962.

Memoirs of Museum Victoria 74:81-91 (2016) Published 2016

ISSN 1447-2546 (Print) 1447-2554 (On-line)

http://museumvictoria.com.au/about/books-and-journals/journals/memoirs-of-museum-victoria/

Morphological variation of stratigraphically important species in the genus

Pilosisporites Delcourt & Sprumont, 1955 in the Gippsland Basin,

southeastern Australia

Doris E. Seegets-Villiers 1 * and Barbara E. Wagstaff 2

1 School of Earth, Atmosphere and Environment, Monash University, Clayton Campus, Victoria 3800, Australia

(doris.seegets-villiers@monash.edu)

2 School of Earth Sciences, The University of Melbourne, Victoria 3010, Australia (wagstaff@unimelb.edu.au)

* To whom correspondence should be addressed. E-mail: doris.seegets-villiers@monash.edu

Abstract Seegets-Villiers, D.E. and Wagstaff, B.E. 2016. Morphological variation of stratigraphically important species in the

genus Pilosisporites Delcourt & Sprumont, 1955 in the Gippsland Basin, southeastern Australia. Memoirs of Museum

Victoria 74: 81-91.

Three hundred and ninety eight mudstone samples of Early Cretaceous age from the onshore part of the Gippsland

Basin in southeastern Australia were used to ascertain the morphological variation in three species of spores in the genus

Pilosisporites. In Australia all species of Pilosisporites are biostratigraphically useful and this study confirms that in the

Gippsland Basin the ranges of Pilosisporites notensis, Pilosisporites parvispinosus and Pilosisporites grandis are as

defined by some previous authors. Morphological variations of these three taxa from the published descriptions are

discussed. In the case of P. grandis and P. parvispinosus the main variation was in the size of specimens, however P.

notensis showed sculpture variations in regard to element size, type and distribution. Two distinct types of this species

were defined with only one occurring in the youngest part of the section. Modern fern species can exhibit similar spore

sculpture and size variations as a result of polyploidy. This could possibly be the cause of the variations in all three species

of Pilosisporites and also their short-lived, in geological terms, species ranges.

Keywords palynology. Early Cretaceous, Gippsland Basin, Pilosisporites, palynostratigraphy.

Introduction

In Australia all four species in the Early Cretaceous spore

genus Pilosisporites Delcourt and Sprumont, 1955 are

biostratigraphically important. The endemic Western

Australian species Pilosisporites ingramii Backhouse, 1988 is

the oldest representative of the genus in Australia and ranges

through the Berriasian Biretisporites enneabbaensis Zone in

that state (Backhouse, 1988). The remaining species

Pilososporites notensis Cookson and Dettmann, 1958

Pilosisporites parvispinosus Dettmann, 1963 and Pilosisporites

grandis Dettmann, 1963 have first and last appearance datums

that play critical roles in defining spore-pollen zones in eastern

Australia. Originally, the endemic Western Australian species,

later defined as P. ingramii (Backhouse, 1988), was identified

as P. notensis (Backhouse, 1978). The confusion over what

constituted the species P. notensis led to Morgan’s (1980)

conclusion that the species had a climate-controlled migration

across the continent. As such P. notensis was regarded as

exhibiting different ranges in the east and west of the continent

(Morgan, 1980; Helby et al., 1987). However, the recognition of

P. ingramii as a separate species, endemic to Western Australia

(Backhouse, 1988) meant that such discussion became

irrelevant and Partridge (2006) set in place the biostratigraphic

usefulness of all four Australian species.

Two species of Pilosisporites, P. notensis and P.

parvispinosus, first described from Australia, have

subsequently been recorded in Early Cretaceous strata in other

continents. Pestchevitskaya (2007) used the first appearance

of P. parvispinosus as a biostratigraphically important datum

in the late Berriasian in Siberia, and P. notensis just above the

early/late Hauterivian boundary and at the base of the

Barremian in different regions in Siberia (Pestchevitskaya,

2007, 2008). In the Athgarh Basin in eastern India both P.

notensis and P. cf. P. notensis were recorded (Goswami et al.,

2008), whereas in the Rajmahal Basin further north Tiwari

and Tipathi (1995) noted the absence of P. notensis and P.

parvispinosus from the Lower Cretaceous sections as being of

interest. Zhang et al. (2008) listed P. notensis in the Lower

Cretaceous of northeast China. The type species of the genus,

Pilosisporites trichopapillosus has never been recorded in

Australia but has been recorded globally in North and South

America, Europe, Africa and Asia (Alroy, 2014).

D.E. Seegets-Villiers and B.E. Wagstaff

Figure 1 . Gippsland Basin map showing the distribution of exposed and subsurface Cretaceous sediments and the locations of petroleum

exploration wells mentioned in text (redrawn from Douglas 1988 and Tosolini etal., 1999).

Morphological variations exhibited by the illustrated

holotypes of the four Australian species of Pilosisporites

(Backhouse, 1988; Dettmann, 1963; Cookson and Dettmann,

1958) are distinctive and should not present any difficulty in

identifying the spores. However, extensive studies on lower

Cretaceous surface and subsurface sedimentary successions

in the Gippsland Basin in Victoria have shown that the eastern

Australian species of Pilosisporites , i.e. P. notensis, P.

parvispinosus, and P. grandis, all exhibit morphological

variations that are not included in the original descriptions.

Study location, lithology and age

All samples used in this study were from the Lower Cretaceous

Strzelecki Group in the Gippsland Basin, Victoria, Australia

(fig. 1). This basin was formed during the Early Cretaceous as

a result of rifting between Australia and Antarctica (Duddy,

2003), with an approximately NW to SE basin axis (Willcox et

al., 1992). Continuous separation, accompanied by progressive

subsidence, resulted in the accumulation of vast sedimentary

successions, the sediment source of which was

contemporaneous volcanism from the active volcanic arc of

the Tasman Rise (Bryan et al., 1997). The rift valley floor was

a vast floodplain crossed by massive westward flowing

(O’Sullivan et al., 2000) primarily braided river systems

(Vickers-Rich et al., 1997) with subordinate lacustrine

conditions (Dettmann, 1986; Waldman, 1971). The resultant

lithologies are dominated by sandstone, mudstone and to a

lesser extent conglomerate and coals (Wagstaff and McEwen

Mason, 1989). Coastal outcrop sections of these rocks were

used for the oldest part of the Early Cretaceous ( Foraminisporis

wonthaggiensis- lowest Cyclosporites hughesii zones /

Hauterivian-early Aptian age) and samples from exploration

wells provided the lowest part (upper Crybelosporites striatus

to Coptospora paradoxa zones/Albian age). The wells used in

this study form part of Lakes Oil N.L. tight gas exploration

program in the onshore Gippsland Basin, Victoria, Australia.

Wombat-1 (Lat 38°21T5”S Long 147°09’32”E) and Wombat-3

(Lat 38°21’33”S Long 147°08’57”E) wells are located in the

Wombat Gas field and the shallow (total depth 366 m)

Boundary Creek-IA sidetrack bore (Lat 38°11’25”S Long

147°07’54”E). Due to lack of outcrop and core there is a gap in

the section examined between the upper part of the C. hughesii

and the lower part of the C. striatus zones therefore

encompassing most of the Aptian.

Materials and methods

Core, cuttings and outcrop samples used had been collected

over a period of 30 years by the authors of this study to

undertake a range of different research objectives including

palynostratigraphic dating, assessing vegetation abundance

changes through time and environmental interpretation. This

resulted in a total data set of 398 samples.

Samples used. At most of the outcrop localities only low-

resolution sampling had been carried out for the exposed

succession (a maximum of two or four samples) focusing on

mudstone beds. Two outcrop sites included in this study

however, involved high-resolution sampling. At the Flat Rocks

(Lat 38°39’39”S Long 145°40’52”E) vertebrate fossil locality

(fig. 1) near Inverloch, rock platform and cliffs had been

sampled to provide 70 samples through the stratigraphic

section. At Kilcunda (Lat 38°33’04”S Long 145°28T3”E) 38

samples had been taken at regular intervals through the

mudstones in a cliff section (fig. 1). One hundred and nineteen

samples were used from the Boundary Creek-IA sidetrack core

between 203.03-366.5 m. Forty-four samples were from

Wombat-1 between 1485-1971 m. They included eight

unwashed cuttings samples and 36 samples from washed

cuttings. Fifty-four washed cuttings samples were examined in

the section in Wombat-3 between 1386-2169 m. This study

also involved the re-examination of the type specimens of P.

parvispinosus , P. notensis and P. grandis.

Morphological variation in Pilosisporites

Table 1 Summary of the processing and mounting mediums used for all samples in this study.

Location/Well

Number of

samples

Hydrochloric

acid

Hydrofluoric

acid

Oxidation

Heavy liquid

separation

Sieving

Mounting

medium

Western coastal

outcrop (San Remo

to Black Head)

Nitric acid

Zinc bromide

10pm

acrylic resin

Schulze solution

Zinc bromide

none

acrylic resin

Kilcunda cliff

Schulze solution

Zinc bromide

none

silicone oil

Eastern coastal

outcrop (Harmers

Haven to Inverloch)

Nitric acid

Zinc bromide

10pm

acrylic resin

Schulze solution

Zinc bromide

none

acrylic resin

Flat Rocks

Schulze solution

Sodium

polytungstate

10pm

glycerine jelly

Nitric acid

Zinc bromide

10pm

acrylic resin

Wombat-1

Nitric acid

Zinc bromide

10pm

acrylic resin

Wombat-3

Nitric acid

Zinc bromide

10pm

acrylic resin

Boundary Creek-IA

Nitric acid

Zinc bromide

10pm

acrylic resin

Schulze solution

Zinc bromide

none

acrylic resin

Sample processing. Thirty-one samples from the Flat Rocks

site and all the samples (38) from Kilcunda were processed by

the authors in the laboratory of the School of Geography and

Environmental Science at Monash University. Morgan Palaeo

Associates processed the remaining 39 samples from the Flat

Rocks site. Laola Pty. Ltd. (subsequently Core Laboratories

Australia Pty. Ltd.) in Western Australia processed 37 outcrop

samples, the remainder being processed by Global Geolab in

Calgary, Canada. Core Laboratories Australia Pty. Ltd. in

Western Australia processed all samples from Wombat-1 and

Wombat-3 plus 24 samples from Boundary Creek-IA sidetrack

bore. Global Geolab in Calgary Canada processing all other

samples from this bore. Table 1 summarises the chemical and

physical processing of all samples. Although the slides were

initially examined and counted on a range of microscopes the

work for this paper was undertaken on a Zeiss Axioscope A1

microscope in the School of Earth Sciences at the University of

Melbourne and all photographs were taken on a Jentopik

ProgRes CT3 camera using ProgRes Mac CapturePro 2.7

software. All illustrated specimens and the type slides are

housed in the palaeontological collection of Museum Victoria

and have registered numbers in the collection (prefixed “P”).

The authors acknowledge that the different mounting

mediums may have an effect on the size of the fossils. Faegri

and Iverson (1975) considered the effects on pollen grains

mounted in glycerine jelly, indicating that the grains had a

tendency to swell in the medium. They (Faegri and Iverson,

1975) noted that this is more serious for large grains rather

than small ones. However, Faegri and Iverson (1975) also

noted that in silicone oil the size of the pollen grain does not

change with storage. Sluyter (1997) investigated the effects of

using silicone oil, glycerine jelly and acrylic resin on pollen

grain size and determined conversion factors to account for

the increase of grain sizes in the latter two mounting mediums.

Sluyter (1997) also determined that glycerine jelly

progressively increases the size of the grains and that

measurements should be made immediately after grain

mounting. However, there is no discussion anywhere in the

literature on either the processing or the mounting medium

altering the sculptural elements or their distribution on spores.

As stated by Large and Braggins (1990), exine morphology is

relatively unaffected by the various chemical treatments.

Table 1 summarises the processing and mounting media used

in this study.

Results

Biostratigraphic outcomes. Due to the size of the Australian

continent, spore-pollen biostratigraphic schemes were initially

developed that related to the present day geographical location

of the basins studied and the palynologists involved. This

resulted in distinctive spore-pollen biostratigraphies for the

Early Cretaceous in Western Australia (Backhouse, 1978,

1988; Balme, 1957,1964), northeastern Australia (Burger 1973,

1986; Norvick and Burger, 1975) and southeastern Australia

(Dettmann, 1986; Dettmann and Douglas, 1976; Dettmann and

Playford, 1969; Morgan et al., 1995). Although the stratigraphic

distribution of many key taxa is the same Australia-wide, the

ranges of many other species that are biostratigraphically

D.E. Seegets-Villiers and B.E. Wagstaff

useful, as noted by Morgan et al. (1995), vary. The pan-

Australian Mesozoic palynostratigraphy published by Helby et

al. (1987), and its latest update by Partridge (2006),

acknowledged some of these range variations, but did not

utilise others. Also, of importance for this study is the fact that

the order of stratigraphic occurrences of some taxa in Partridge

(2006) varies when compared to the most recent biostratigraphic

scheme of Dettmann (1986) for Victoria. Recently, Wagstaff et

al. (2012) defined subzones of the Coptospora paradoxa Zone,

the youngest zone in this study.

In this study, the coastal outcrop sections examined

encompassed the Foraminisporis wonthaggiensis Zone to just

above the base of the overlying Cyclosporites hughesii Zone

of Helby et al. (1987). The first appearance of the indicator of

the base of this zone, i.e. Foraminisporis asymmetricus and/or

the first appearance of the angiosperm species Clavatipollenites

hughesii (this study), marks the top of the coastal section. The

wells examined encompass a section that starts with the upper

part of the Crybelosporites striatus Zone and almost reaches

the top of the overlying Coptospora paradoxa Zone. Although

this study does not cover the entire Lower Cretaceous in the

Gippsland Basin, due to lack of continuous stratigraphic

section, enough section has been examined to allow some

conclusions to be drawn in relation to the spore-pollen ranges

and the biostratigraphic usefulness of taxa. This study

confirms the order of first appearances of P. notensis and P.

parvispinosus in the Gippsland Basin as indicated by

Dettmann (1986) and P. grandis as shown in Dettmann and

Douglas (1976). Using the zones of Helby et al. (1987), P.

notensis first appears at the base of the Foraminisporis

wonthaggiensis Zone, and Pilosisporites parvispinosus first

appears within this zone. The first appearance of P. grandis

defines the base of the P. grandis Subzone (Wagstaff et al.,

2012) of the Coptospora paradoxa Zone while bothP. notensis

and P. parvispinosus become extinct, in the Gippsland Basin,

just above this event. This study has confirmed that the order

of first and last occurrences up section is as shown in figure 2,

and as such rejects some of the suggestions of Partridge (2006)

for the Gippsland Basin.

Taxonomic outcomes. The diagnosis of P. grandis and P.

parvispinosus basically conform to those of Dettmann (1963),

except for minor variations as discussed below. However, P.

notensis shows a range of sculptural variations (size, type,

distribution of sculptural elements) in which the holotype is

only representative of one morphological variant. It is enigmatic

as to why these variations have not been described before as a

close inspection of the type slide P17611 as used by Cookson

and Dettmann (1958) revealed that some of these variations

were present on specimens that occur in this slide, but were not

included in the original description. Also, an examination of the

descriptions and the photo plates in the literature suggests that

previous workers had encountered this variation, as evidenced

by published images representing these different types.

The species of Pilosisporites are discussed in

geochronologic order from oldest to youngest.

Genus Pilosisporites Delcourt and Sprumont, 1955

TYPE SPECIES (by original designation): Pilosisporites

trichopapillosus (Thiergart, 1949) Delcourt and Sprumont,

1955 (Early Cretaceous, Germany).

Spore-pollen zones

FommfnJsporfs

wan thaggiansis

Cycfosporites

tiugftesii

Crybetasp antes

striatus n

tr.

CoptOSpprm paradoxa

hisses

MfflRM

rmia

Gippsland {This study)

potfortaaspom

Fommtnisporls

wonthaggtmnmi*

CydoMporitps

tHighmmii

Cryboiosporttmm

mtrtmivs

Coptospora part*? cure

Pfitmopotienttas

pannoaua

Gippsland {Partridge £006)

DIclyotosporitBB spedosus Zone

Coptospora paradoxa

Lower Upper

Pfrirrr ap alien! t&s

pnnnnnu;;

Gippsland {Dettmann, 1986;

Dettmann and Douglas, 1979)

CryoeiospoF i f*?s

Btylasus

CydoSp

Lower

writes htrgticsli

Middle

1 MJbxon*

Upoer

Crybytospprftes

mtrttHu* aubiorie

yrl

1/*

9&ii

tosporttea spec

ioauB

Is wontfiagglartsls

ry aspo

Ites styfosus

Murompora ft

y p s

C I

__,_

os spo a parvi sp nc

f>0

Ctovati ottenitos^hu hosii *

.... R

it . u n& spor tes ctyus

p ap pa

■ Pifostf&porftcs gt&ncffs

par p

PerotrttBtes^ubalus *

■ Pitfmopollpntte# pannosua

Cowtal —CttotK Well sections

Figure 2. Spore-pollen zones and age indicator species in the Gippsland Basin. Green shaded areas represent the stratigraphy recorded in this

study, with the ranges of taxa encountered shown in black and the ranges of taxa not recorded shown in brown. Some ranges are inferred from

previous work of Dettmann (1986) and Dettmann and Douglas (1979).

Morphological variation in Pilosisporites

Figure 3. Scale bar Figs A-H = 10 pm; Figs I-J = 12 pm, EFR = England Finder Reference, DIC = Differential interference contrast. A

Pilosisporites notensis Holotype Slide P1761 EFR Q37/4. Photograph taken using DIC. B Pilosisporites notensis Flat Rocks Slide P252131 EFR

T31/1. Specimen with long apical spines. Photograph taken using DIC. C, D Pilosisporites notensis Wombat-3 Slide P252136 #2 EFR P28/0. E,

F, H Pilosisporites notensis Shack Bay Slide P252135 EFR H21/1. G Pilosisporites notensis Flat Rocks Slide P252132 EFR H37/1 Photograph

taken using DIC. I Pilosisporites notensis Flat Rocks Slide P252132 EFR F39/4. Sculpture predominantly spines. Photograph taken using DIC.

J Pilosisporites notensis Flat Rocks P252133 EFR E21/1. Sculpture cones and curved spines.

D.E. Seegets-Villiers and B.E. Wagstaff

Figure 4. Range of sculptural elements recorded in Pilosiporites notensis.

Pilosisporites notensis Cookson and Dettmann, 1958 emend.

Figure 3 A-J

Synonymy : 1958 Pilosisporites notensis Cookson and

Dettmann, p. 102, pi. 15, figs 1, 3

1963 Pilosisporites notensis Cookson and Dettmann;

Dettmann, p. 37-38, pi. 4, figs 1-5; p. 33 fig 4D

1964 Pilosisporites notensis Cookson and Dettmann;

Balme, p.74, pi. VII, fig. 10

1969 Pilosisporites notensis Cookson and Dettmann;

Dettmann and Playford, pi. 11, fig. 2

1973 Pilosisporites notensis Cookson and Dettmann;

Burger, pi. 2, fig. 1

1974 Pilosisporites notensis Cookson and Dettmann;

Burger, pi. 13 fig 11

1976 Pilosisporites notensis Cookson and Dettmann;

Norvick and Burger, pi. 18, fig 17

non 1978 Pilosisporites notensis Cookson and Dettmann;

Backhouse, p.18, PI. 2, fig. 1

1980 Pilosisporites notensis Cookson and Dettmann;

Burger, p. 52, pi. 6, fig. 5

1986 Pilosisporites notensis Cookson and Dettmann;

Dettmann, fig. 6M

1987 Pilosisporites notensis Cookson and Dettmann;

Helby et al., fig. 20N

1988 Pilosisporites notensis Cookson and Dettmann;

Backhouse, pi.15 fig. 2

2012 Pilosisporites notensis Cookson and Dettmann;

Wagstaff et al., pl.II fig. 2

Emended diagnosis: Trilete spores, with strongly convex distal

and almost flat proximal surface. Amb triangular with concave

to nearly straight sides and broadly rounded apices. Laesurae

straight, length 2 / 3 to 3 4 of spore radius, with raised,

membraneous lips. Exine 1.5-3.5 pm thick, ornamented by

straight-sided or curved cones and/or spines ( sensu Punt et al.,

1994) of 0.5-3.5 pm basal diameter and 1-6.5 pm height. This

sculpture co-occurs with rare other sculpture types including

baculae, clavae, pilae, tuberculae and verrucae (fig. 4).

Sculptural elements consistently arranged along laesurae

margins and either covering the entire spore surface or

restricted to the apical areas. Sculpture elements of equal size

and distribution on proximal and distal surface.

Size: Equatorial diameter 52 (80) 125 pm (136 specimens),

polar diameter 52 (71) 91 pm (14 specimens).

Remarks and comparison : Specimens generally conform to the

original descriptions (Cookson and Dettmann, 1958; Dettmann,

1963). However, observations on a large number of specimens in

this study have revealed differences regarding the distribution

and type of sculptural elements (figs. 3, 4). The most common

morphological extreme of P. notensis exhibits short sculptural

elements (up to 3 pm high) and base diameters of between 0.5-3

pm. This group shows sculpture distributed either over the entire

spore surface as illustrated by Backhouse (1988) or mainly

restricted to apical areas where they are closely packed (fig. 3G).

The second morphological extreme, shows elongate sculptural

elements (up to 6.5 pm), with a narrow basal diameter of as little

as 1 pm. Sculptural elements are either distributed over the entire

distal and proximal surface (figs. 3E, F, H) or primarily restricted

to and closely packed together in the apical areas of the spore

(fig. 3B) as in the original description of P. notensis by Cookson

and Dettmann (1958) and as occurs in the holotype (fig. 3A).

The specimens of P. notensis with long apically distributed

spines, superficially resemble P. ingramii Backhouse (1988) in

that they possess long spines, and in some instances, a single

row of small spines on the laesurae. However, the grain size in

general exceeds that of P. ingramii and sculptural elements are

in general shorter. Between each morphological end member

there is a continual spectrum of distribution of sculptural

elements on the amb of the spores ranging from primarily the

apex with rare inter-apical occurrences, to increasing frequency

of elements inter-apically to spores in which the sculpture

extends over the entire surface of the amb. Sculptural element

size and distribution is independent of the overall size of the

grain.

Two other species of Pilosisporites bear a resemblance to P.

notensis. Pilosisporites trichopapillosus (Thiergart) Delcourt

and Sprumont, 1955 is recorded as also having sculpture

variation in which the long spines were restricted to apical areas

or covering the entire surface of the grain (Couper, 1958).

However, there is no suggestion that there are any sculptural

elements other than spines on this species (e.g. Couper, 1958;

Singh, 1964; Archangelsky and Llorens, 2005). Pilosisporites

verus (Delcourt and Sprumont) emend. Archangelsky and

Llorens, 2005 has sculptural elements that include spines with

broad bases and acute apices that are sometimes curved, with

lesser numbers of cones, bacula, granules and mameliform

processes. The sculptural elements are often concentrated at the

apical areas (Singh, 1964) but the spines in P. verus are longer,

i.e. 5.5-11 pm (Archangelsky and Llorens, 2005), than those

that occur in P. notensis. The equatorial diameter of both these

species is also within the lower end of the size of P. notensis.

Morphological variation in Pilosisporites

Figure 5. Scale bar Figs A-E and G-H = 10 pm; Fig. F=25 pm. EFR = England Finder Reference. A, B, F Pilosisporites grandis Holotype Slide

P22098 EFR N20/0. C, D Pilosisporites grandis Boundary Creek -1A core 285.5m Slide P252137 EFR Q14/4. E Pilosisporites parvispinosus

Holotype Slide P21997 EFR E37/3. G Pilosisporites parvispinosus Flat Rocks Slide P252134 EFR L41/3. H Pilosisporites parvispinosus Flat

Rocks Slide P252134 EFR Q31/1.

D.E. Seegets-Villiers and B.E. Wagstaff

Distribution (this study): Base of the Foraminisporis

wonthaggiensis Zone to within the lower part of the P. grandis

subzone of the C. paradoxa Zone. This study found that both end

members co-occur in the older part of the succession with their

First Appearance Datum (FAD) at the base of the F.

wonthaggiensis Zone. However, in the uppermost part of the

range of P. notensis, i.e. in the C. striatus and C. paradoxa zones,

only the shorter spined forms occur (figs. 3C, D). P. notensis is

never abundant or common in the upper part of its range.

Pilosisporites parvispinosus Cookson and Dettmann, 1958

Figure 5E, G, H

Synonymy 1958 Pilosisporites notensis Cookson and Dettmann,

p. 102, pi. XV fig. 2

1963 Pilosisporites parvispinosus Dettmann, p. 38, pi. IV

figs. 6-8; p. 33 fig 4F

1980 Pilosisporites parvispinosus Dettmann; Burger, p.

52, pi. 5 fig 9; pi. 6 figs 1-3

non 2012 Pilosisporites parvispinosus Dettmann;

Lebedeva and Pestchevitskaya, pi. 1 fig. 7

2012 Pilosisporites parvispinosus Dettmann; Wagstaff et

al., pi. II fig. 7

Diagnosis: see Dettmann (1963).

Size: Equatorial diameter 78 (85) 106 pm (11 specimens).

Remarks: Representatives of this species usually conform to

the diagnosis by Dettmann (1963), but show, in rare cases, a

slightly thicker exine (2-3.5 pm compared to Dettmann’s

(1963) 2-3 pm). The spinulate sculpture occasionally has

spinules with a wider base diameter (1-2 pm) compared to

Dettmann’s 1 pm) and a sculpture height of (1-2.5 pm). Rarely

present are longer apical sculptural elements.

Distribution (this study): Upper part of the Foraminisporis

wonthaggiensis Zone to Coptospora paradoxa Zone,

specifically the lower part of the Pilosisporites grandis subzone.

Pilosisporites grandis Dettmann, 1963

Figures 5A-D, F

Synonymy: 1963 Pilosisporites grandis Dettmann, pp. 38-39,

pi. V, figs 1-3; p. 33 fig 4E

1969 Pilosisporites grandis Dettmann; Dettmann and

Playford, pi.11 fig. 1

1980 Pilosisporites grandis Dettmann; Burger, p. 52, pi. 6,

fig. 4

2012 Pilosisporites grandis Dettmann; Wagstaff et ah, p.

69, pi. 2 fig. 2

non 2014 Pilosisporites grandis Dettmann; Takeshi and

Vijaya, fig. 5E

Size : Equatorial diameter 85 (92) 100 pm. (4 specimens).

Remarks: As discussed in Wagstaff et ah (2012) the grains

encountered in the Gippsland Basin are smaller than those

originally described by Dettmann (1963) from the Great

Artesian and Otway Basins. Dettmann (1963) gave the

equatorial diameter of the species as 100 (117) 142 pm. The

specimen illustrated by Burger (1980) from the Surat Basin

(Burger, 1980, plate 6, fig. 4) is also smaller (appearing to be

approximately 90 pm) than the size range given by Dettmann

(1963). We accept the statement of Burger (1980) in the plate

(pi. 6, fig. 4) description that the species is primarily identified

by its “dense and regular distribution of spines on the exine”.

Distribution (this study): Coptospora paradoxa Zone,

specifically Pilosisporites grandis to Cicatricosisporites

cuneiformis subzones.

Discussion

The genus Pilosisporites is unique in Australian palynology in

that all four species of the genus are biostratigaphically useful.

In spite of the morphological diversity of P. notensis, and the

size variations of P. grandis and P. parvispinosus, this fact is

not altered. However, this variation is, in the case of P.

notensis, so extreme that it warrants further investigation.

Hughes (1989) made an in-depth critique of the concept of

species and indicated that over time the addition of new

specimens identified as a species can cause the original

definition of the species to become broader causing the taxon

to become an ever larger cluster with irregular limits rather

than the neat concept of its generator. This is due to more

specimens being examined that encompass a greater span of

geologic time, geography (other localities) or both. Hughes

(1989, p. 12) also stated that “From the moment of publication,

it is customary for appropriate other newly discovered material

to be ‘identified with’, or attributed to, the existing species as

far as possible, because new species are in general erected

with caution and even with reluctance.” This leads over time to

the alteration of the meaning or scope of definition of the

totality of specimens included in the species under the name.

This seems to be the case for P. notensis where an examination

of the published images from across eastern Australia from

different localities and different spore-pollen zones, shows a

variation in morphology. However, temporal and geographic

components cannot be the only reasons for the variation in the

current study. The Flat Rocks and Kilcunda localities are both

small stratigraphic sections that were sampled (70 and 38

respectively) with high resolution. At each locality the full

range of morphological variation of P. notensis occurred, even

in many instances in a single sample. As such, the variation

cannot be attributed to geologic time or geographic differences

and needs to be explained.

Dettmann (1963, 1986) suggested no modern equivalents

for the Victorian Pilosisporites spore species, but did consider

them to be produced by ferns. Variation of the spore sculpture

within a species is not unknown in extant ferns. Parks et al.

(2000) found that the monolete fern spore Cystopteris fragilis

(L.) Bernhardi 1805 that occupied different microclimates and

substrates in Scotland had three distinctive sculptural types:

echinate, rugose and a “smooth” type that appeared to be

slightly granular using SEM. The plants producing these

distinctive spore types occupied different microclimates and

substrates (Parks et al., 2000). Parks et al. (2000) concluded

that the fern C. fragilis was polyploid with several populations

of variants. Beck et al. (2011) indicated that ploidy level could

Morphological variation in Pilosisporites

also control the size of spores with diploids, triploids, and

tetraploids of fern species in the genus Astrolepis, Benham

and Windham 1992 having statistically different and non¬

overlapping spore diameter distributions. Diverse ploidy levels

were also used to explain marked spore size differences in the

tropical American fern genus Stigmatopteris Christensen

1909 (Tryon and Tryon, 1982). These morphological variations

pose an interesting dilemma for palynologists as Otto and

Whitten (2000) estimate that the frequency of polyploidy is

particularly high in extant ferns (41.7%).

Polyploidy is the existence of genetically related taxa

(Thompson and Lumaret, 1992) with three or more basic

chromosome sets in their cell nuclei as opposed to the usual

two (Bennet, 2004). Polyploidy is heritable (Comai, 2005) and

allows homosporous ferns to create and maintain genetic

variation in spite of the effects of self-fertilisation due to their

monoecious gametophtye stage (Klekowski and Baker, 1966).

Klekowski and Baker (1966) further suggested that polyploidy

provides selective advantages to homosporous ferns allowing

them to maintain, create and release genetic variability in spite

of producing homozygous sporophytes. Page (2002) indicates

that the entire process of polyploidy in ferns results in the

creation of morphological and ecological novelty providing a

rapid route for species evolution and adaption and increasing

the plants’ vigour. The resulting taxa have the ability to adapt

to a broader range of ecological conditions compared to their

parents and in an actively evolving flora in a changing

environment higher ploidy derivatives more often find niches

for ecological success. Comai (2005) provided evidence that

polyploidy increases the diversity and plasticity of a species,

and contributes to its adaptive potential in the arctic where

polyploids are able to rapidly adapt to new niches and are able

to efficiently invade newly deglaciated areas due to hybrid

vigour. However, these polyploids are often considered to be

static entities in which gene silencing results in isolated

populations (Haufler, 2008). Thus, the picture for polyploids is

not favourable in the long term. Their ability to establish and

flourish during periods of environmental change when new

niches are opened is well recognised, but polyploids face

evolutionary difficulties as gene selection is inefficient due to

the multiple copies and they are often evolutionary dead-ends

(Arrigo and Barker, 2012).

High levels of niche variablity would be a major feature of

the Early Cretaceous of the Gippsland Basin. The evolving rift

between Australia and Antarctica (Duddy, 2003) and its fast

flowing braided river system (Vickers-Rich et al., 1997) would

have been adynamic unstable landscape. The palaeogeographic

setting of southeastern Australia was significant in that it was

near or inside the Antarctic Circle (Rich and Vickers-Rich,

2000; Wagstaff et al., 2013). Therefore, in spite of the high

carbon dioxide levels characteristic of the Cretaceous

Greenhouse, and warm and equable global climates (Spicer

and Corfield, 1992), it is suggested that in the Aptian-Albian

this region would still have been cold (Gregory et al., 1989)

and in winter would have experienced months of darkness.

The short biostratigraphic ranges of the species of

Pilosisporites that occur in the Cretaceous successions in

Australia suggest each species is very intolerant of climate/

environmental variability. The inability of P. ingramii to

migrate from Western Australia to the east implies high levels

of endemism, as does the inability of P. parvispinosus , P.

notensis and P. grandis to migrate into Western Australia. As

the climate became drier in the Gippsland Basin (Wagstaff et

al., 2013) the extinction of both P. parvispinosus and P.

notensis in the Albian and the evolution of P. grandis just

prior to this event, seems to suggest that the two former species

could not cope with these changed conditions. The variation

that occurs in the morphology of P. notensis could represent a

fern species that had undergone polyploidy thus allowing it to

inhabit a range of niches in an unstable braided fluvial setting.

The fact that the full range of morphotypes can occur in one

sample appears to indicate that the fern is occasionally

abundant/diverse in a riparian setting that encompasses a

range of sub-environments.

Conclusions

This study on spores in the genus Pilosisporites in Lower

Cretaceous strata in the Gippsland Basin has three outcomes.

The biostratigraphic usefulness of the three species is

confirmed with the following ranges. Pilosisporites notensis

first appears at the base of the Foraminisporis wonthaggiensis

Zone, and Pilosisporites parvispinosus within the zone. The

first appearance of Pilosisporites grandis defines the base of

the P. grandis subzone of the Coptospora paradoxa Zone

while both P. notensis and P. parvispinosus become extinct

just above this event. Taxonomically, the descriptions of P.

grandis and P. parvispinosus have extended size ranges

compared to the original descriptions. P. notensis shows not

only variation in size from the original description but also a

division into shorter and longer sculptural element types, and

within these, variation in the distribution of the sculpture on

the amb. This study suggests that the inability of the eastern

and western Australian species of Pilosisporites to migrate

across the continent, in combination with their short geological

ranges and the size and sculptural variation exhibited by the

three species examined from the Gippsland Basin, may be

evidence of polyploid reproductive strategies in the ferns of

this genus.

Acknowledgements

We would like to thank Professor Patricia Vickers-Rich and Dr

Thomas Rich who funded the majority of the research on the

outcrop samples. BHP Billiton funded the research at Kilcunda.

We would like to thank Lakes Oil N.L. for providing some of

the samples from Wombat-3. The Department of Primary

Industries allowed sampling of the wells and we would

particularly like to thank Terry Smith for providing help with

accessing the cuttings and core. We would like to thank David

Pickering from Museum Victoria for locating and loaning us

the type slides from their collection. We appreciate the

comments provided by the editor and two anonymous reviewers

that helped improve the text. ARC Linkage Grant LP0989203

funded the research on the well sections with industry partners

Lakes Oil N.L., Nexus Energy and Geotrack International.

D.E. Seegets-Villiers and B.E. Wagstaff

References

Alroy, J. 2014. Taxonomic occurrences of Pilosisporites

trichop apillosus recorded in the Paleobiology Database.

Fos si 1 works .http: //fossi 1 works. org.

Archangelsky A., Llorens, M., 2005. Palinologfa de la Formation

Kachaike, Cretacico Inferior de la Cuenca Austral, provincia de

Santa Cruz. II Esporas. Ameghiniana 42(2): 311-328.

Arrigo, N., Barker, M.S., 2012. Rarely successful polyploids and their

legacy in plant genomes. Current Opinion in Plant Biology 15: 1-7.

Backhouse, J., 1978. Palynological zonation of the Late Jurassic and

Early Cretaceous sediments of the Yarragadee Formation central

Perth Basin, Western Australia. Geological Survey of Western

Australia Report 7: 1-52.

Backhouse, J., 1988. Late Jurassic and Early Cretaceous palynology of

the Perth Basin, Western Australia. Geological Survey of Western

Australia Bulletin 135: 1-233.

Balme, B.E., 1957. Spores and pollen grains from the Mesozoic of

Western Australia. Commonwealth Scientific and Industrial

Research Organization Coal Research Section, Technical

Communication 25: 1-48.

Balme, B.E., 1964. The palynological record of Australian pre-

Tertiary floras. Pp.49-80, in: Cranwell, L.M., (ed). Ancient Pacific

Floras. University of Hawaii Press, Honolulu.

Bennet, M.D., 2004. Perspectives on polyploidy in plants - ancient

and neo. Biological Journal of the Linnean Society 82: 411-423.

Beck, J.B., Windham, M.D., Pryer, K.M., 2011. Do asexual polyploidy

lineages lead short evolutionary lives? A case study from the fern

genus Astrolepis. Evolution 65: 3217-3229.

Bryan, S.E., Constantine, A.E., Stephens, C.J., Ewart, A., Schon,

R.W., Parianos, J., 1997. Early Cretaceous volcano-sedimentary

successions along the eastern Australian continental margin:

Implications for the break-up of eastern Gondwana. Earth and

Planetary Science Letters 153: 85-102.

Burger, D., 1973. Spore-zonation and sedimentary history of the

Neocomian, Great Artesian Basin, Queensland. Special

Publications Geological Society of Australia 4: 87-118.

Burger, D., 1980. Palynology of the Lower Cretaceous in the Surat

Basin. Australian Bureau of Mineral Resources Bulletin 189: 1-106.

Burger, D., 1986. Palynology, cyclic sedimentation, and

palaeoenvironments in the late Mesozoic of the Eromanga Basin.

Pp. 53-70. in: Gravestock, D.I., Moore, P.S., Pitt, G.M., (eds).

Contributions to the Geology and Hydrocarbon Potential of the

Eromanga Basin. Special Publication Geological Society of

Australia 12. Geological Society of Australia, Sydney.

Clarke, L.J., Jenkyns, H.C., 1999. New oxygen isotope evidence for

long-term climatic change in the Southern Hemisphere. Geology

27(8): 699-702.

Comai, L., 2005. The advantages and disadvantages of being

polyploid. Nature Reviews Genetics 6: 836-846.

Cookson, I. Dettmann, M. E., 1958. Some trilete spores from Upper

Mesozoic deposits in the eastern Australian region. Proceedings

of the Royal Society of Victoria 70: 95-128.

Couper, R.A., 1958. British Mesozoic microspores and pollen grains: A

systematic and stratigraphic study. Palaeontographica Abt. B 103:

75-179.

Dettmann, M.E., 1963. Upper Mesozoic microfloras from south¬

eastern Australia. Proceedings of the Royal Society of Victoria 77:

1-148.

Dettmann, M.E., 1986. Early Cretaceous palynoflora of subsurface

strata correlative with the Koonwarra Fossil Bed, Victoria. Memoir

of the Association of Australasian Palaeontologists 3: 79-110.

Dettmann, M.E., Douglas, J.G., 1976. Mesozoic palaeontology.

Special Publications Geological Society of Australia 5: 164-169.

in: Douglas, J.G., Ferguson, J.A. (eds). Geology of Victoria

Dettmann, M.E., Playford, G, 1969. Palynology of the Australian

Cretaceous: a review. Pp. 174-210. in: Campbell, K.S.W. (ed).

Stratigraphy and Palaeontology, Essays in Honour of Dorothy

Hill. Australian National University Press, Canberra.

Duddy, I.R., 2003. Mesozoic. Pp. 239-286. in Birch, W.D., (ed).

Geology of Victoria. Special Publication Geological Society of

Australia 23.

Faegri, K.,Iverson, J., 1975. Textbook of Pollen Analysis. Munksgaard,

Copenhagen, Denmark. 295 pp.

Frakes, L.A., Francis, J.E., Syktus, J.I., 1992. Climate Modes of the

Phanerozoic. Cambridge University Press. 274pp.

Goswami, S., Meena, K.L., Das, M., Guru, B.C., 2008. Upper

Gondwana palynoflora of Mahanadi Master Basin, Orissa, India.

Acta Palaeobotanica 48(2): 171-181.

Gregory, R T, Douthitt, C.B., Duddy, I.R., Rich, P.V., Rich, T.H., 1989

Oxygen isotopic composition of carbonate concretions from the

Lower Cretaceous of Victoria, Australia: implications for the

evolution of meteoric waters on the Australian continent in a

paleopolar environment. Earth and Planetary Science Letters

92(1): 27-42.

Helby, R., Morgan, R., Partridge, A.D., 1987. A palynological zonation

of the Australian Mesozoic. Memoir of the Association of

Australasian Palaeontologists 4: 1-94.

Hughes, N.F., 1989. Fossils as Information: New Recording and

Stratal Correlation Techniques. Cambridge University Press,

Cambridge, UK. 136 pp.

Haufler, C.H., 2008. Species and speciation. Pp 303-331. in: Ranker,

T.A., Haufler, C.H. (eds). Biology and Evolution of Ferns and

Lycophytes, Cambridge University Press, Cambridge, UK.

Klekowski, E.J., Baker, H.G., 1966. Evolutionary significance of

polyploidy in the Pteridophya. Science 153: 305-307.

Morgan, R. 1980. Palynostratigraphy of the Australian Early and

Middle Cretaceous. Memoirs of the Geological Survey of New

South Wales, Palaeontology 18: 1-153.

Morgan, R., Alley, N.F., Rowett, A.I., White, M.R., 1995. Chapter 6—

Biostratigraphy. Pp. 95-101 in: Morton, J.G.G., Drexel, J.F. (eds).

The Petroleum Geology of South Australia: Vol. 1: Otway Basin.

Mines and Energy South Australia Report 95/12.

Norvick, M.S., Burger, D., 1975. Palynology of the Cenomanian of

Bathurst Island, Northern Territory, Australia. Bureau of Mineral

Resources Geology and Geophysics Bulletin 151: 1-169.

O’Sullivan, P.B., Mitchell, M.M., O’Sullivan, A.J., Kohn, B.P,

Gleadow, A.J.W., 2000. Thermotectonic history of the Bassian

Rise, Australia: implication for the breakup of eastern Gondwana

along Australia’s southeastern margins. Earth and Planetary

Science Letters 182: 31-47.

Otto, S.P., Whitten, J., 2000. Polyploid incidence and evolution.

Annual Review of Genetics 34:401-37.

Page, C.N., 2002. Ecological strategies in fern evolution: a neotropical

overview. Review of Palaeobotany and Palynology 119: 1-33.

Parks, J.C., Dyer, A.F., Lindsay, S., 2000. Allozyme, spore and frond

variation in some Scottish populations of the ferns Cystopteris

dickieana and Cystopteris fragilis. Edinburgh Journal of Botany

57(1): 83-105.

Partridge, A.D., 2006. Jurassic - Early Cretaceous spore-pollen and

dinocyst zonations for Australia, in: Monteil, E. (co-ord),

Australian Mesozoic and Cenozoic Palynology Zonations -

updated to the 2004 Geologic Time Scale. Geoscience Australia

Record 2006/23 (CD only).

Pestchevitskaya, E.B., 2007. Lower Cretaceous biostratigraphy of

northern Siberia: palynogical units and their correlation

significance. Russian Geology and Geophysics 48: 941-959.

Morphological variation in Pilosisporites

Pestchevitskaya, E.B., 2008. Lower Cretaceous palynostratigraphy and

dinoflagallate cyst palaeoecology in the Siberian palaeobasin.

Norwegian Journal of Geology 88: 279-286.

Punt, W., Blackmore, S., Nilsson, S., Le Thomas, A., 1994. Glossary of

pollen and spore terminology. LPP Contributions Series No.l,

Utrecht.

Rich, T.H., Vickers-Rich, P., 2000. The Dinosaurs of Darkness.

University of Indiana Press, Bloomington. 222 pp.

Royer, D.L., 2006. C0 2 -forced climate thresholds during the

Phanerozoic. Geochiemica et Cosmochimica Acta 70: 5665-5675.

Singh, C., 1964. Microflora of the Lower Cretaceous Mannville Group,

East Central Alberta. Alberta Research Council Bulletin 15: 1-239.

Sluyter, A., 1997. Analysis of maize ( Zea mays subsp. mays) pollen

normalizing the effects of microscope-slide mounting media on

diameter determinations. Palynology 21: 35-39.

Spicer, R.A., Corfield, R.M., 1991. A review of terrestrial and marine

climates in the Cretaceous with implications for modelling the

“Greenhouse Earth.” Geological Magazine 129(2): 169-180.

Thompson, J.D., Lumaret, R., 1992. The evolutionary dynamics of

polyploidy plants: origins, establishment and persistence. Tree 7(9):

302-307.

Tiwari, R.S., Tipathi, A., 1995. Palynological assemblages and absolute

age relationship of Intertrappean beds in the Rajmahal Basin, India.

Cretaceous Research 16: 53-72.

Tryon, R.M., Tryon, A.F., 1982. Ferns and Allied Plants with Special

Reference to Tropical America Springer-Verlag, New York, USA.

857 pp.

Vickers-Rich, P, Rich, T. H., Constantine, A., 1997. The polar

dinosaurs of southeastern Australia. Pp. 253-257. in: Wolberg, D.

L. , Stump, E., Rosenberg, G. D. (eds), Dinofest International

Proceedings. Academy of Natural Sciences, Philadelphia.

Wagstaff, B.E., McEwen Mason, J., 1989. Palynological dating of

Lower Cretaceous coastal vertebrate localities, Victoria,

Australia. National Geographic Research 5(1): 54-63.

Wagstaff, B.E., Gallagher, S.J., Norvick, M.S., Cantrill, D.J., Wallace,

M. W., 2013. High latitude Albian climate variability: Palynological

evidence for long-term drying in a greenhouse world.

Palaeogeography, Palaeoclimatology, Palaeoecology 386: 501-511.

Wagstaff, B.E., Gallagher, S. J., Trainor, J.K., 2012. A new subdivision

of the Albian spore-pollen zonation of Australia. Review of

Palaeobotany and Palynology 171: 57-72.

Waldman, M., 1971. Fish from the Lower Cretaceous of Victoria,

Australia with some comments palaeoenvironement. Special

Papers in Palaeontology 9: 1-124.

Willcox, J. B., Colwell, J. B., Constantine, A., 1992. New ideas on

Gippsland Basin regional tectonics. Pp. 93-110. in: Barton, C. M.,

Hill, K. A., Abele, C., Foster, J., Kempton, N. (eds). Energy,

Economics and Environment: Gippsland Basin Symposium.

Australian Institute of Mining and Metallurgy.

Zhang, J., Liu, H-t, Wu, B-w, 2008. Early Cretaceous spore and pollen

assemblages from the Zhangqiang Depression in the Zhangwu

Basin, Liaoning Province. Acta Micropalaeontologica Sinica

2008 (Issue 2): 196-203.

Memoirs of Museum Victoria 74:93-96 (2016) Published 2016

ISSN 1447-2546 (Print) 1447-2554 (On-line)

http://museumvictoria.com.au/about/books-and-journals/journals/memoirs-of-museum-victoria/

What is ‘Pseudo’ in Pseudotribosphenic Teeth?

Alistair R. Evans 1 - 2

Abstract

Keywords

1 School of Biological Sciences, Monash University, VIC 3800, Australia (alistair.evans@monash.edu)

2 Geosciences, Museum Victoria, Melbourne, VIC 3001, Australia

Evans, A.R. 2016. What is ‘Pseudo’ in Pseudotribosphenic Teeth? Memoirs of Museum Victoria 74: 93-96.

The discovery of a ‘pseudotribosphenic’ lower tooth row in 1982, with a basin anterior to the trigonid rather than

posterior, caused a large stir in mammalian palaeontology. This indicated that a tooth shape of equivalent complexity to

the tribosphenic tooth form could evolve more than once. The upper tooth predicted to occlude with the pseudotribosphenic

molar was reconstructed with a ‘pseudoprotocone’ to occlude with the pseudotalonid basin. Here I discuss the relative

merits of naming the major upper lingual cusp of pseudotribosphenic molars as ‘protocone’ due to its likely similar

developmental and functional relations as the protocone of tribosphenic molars. The use of a different name implies

greater morphological distance between tribosphenic and pseudotribosphenic upper molars than is perhaps warranted, and

likely exaggerates the perception of the difficulty in evolving both tribospheny and pseudotribospheny. The choice between

the evolution of the alternative forms of tribospheny may in fact be related to the degree of anterior-posterior bias in lower

molar development - tribospheny with a posterior bias, while pseudotribospheny with an anterior one.

tribosphenic, pseudotribosphenic, Shuotherium, protocone, pseudoprotocone.

Introduction

‘Tribosphenic’ was the term Simpson (1936) coined for the

basal tooth type of all extant therian mammals, from its dual

functions of grinding (‘tribo’) and shearing (wedge or ‘sphen’).

The key structures of this tooth form are the occluding blades

leading from the main cusps (forming a W-shaped ectoloph on

the upper molar, and a disconnected W on the lower molar),

and the mortar-and-pestle crushing of the protocone on the

lingual side of the upper molar into the talonid basin that

sits at the posterior of the lower molar behind the elevated

trigonid (fig. la). For decades, the complexity and intimate

relationships between these teeth led workers to the conclusion

that it would be ‘almost inconceivable’ that such a tooth

shape could have evolved more than once in the history of

mammals (Simpson, 1936:797). After Simpson’s work,

Patterson (1956) outlined the stages of evolution of the

tribosphenic molar. Based on afunctional analysis of occluding

crests, Crompton (1971) detailed a scenario for the evolution of

the tribosphenic dentition from pre-tribosphenic forms. The

importance of the tribosphenic form in the evolutionary

history of mammals was emphasised by Tom Rich’s graduate

advisor, Malcolm McKenna (1975), using it to diagnose a

clade of crown therians (Tribosphenida).

The single origin of the tribosphenic form began to look

more doubtful with the discovery of Shuotherium dongi by

Chow and Rich (1982), in which a small basin was positioned

at the anterior of the elevated, triangular trigonid of the lower

molars (fig. lb). Chow and Rich (1982) termed this basin the

pseudotalonid, in analogy to the shape and function of the true

talonid. For this to be analogous to the talonid, it must be a

crushing basin that receives a protocone-like structure. Chow

and Rich (1982) predicted that the upper molars of Shuotherium

would have such a cusp, which they termed the

‘pseudoprotocone’. This prediction was borne out by the

discovery by Wang et al. (1998) of an upper molar of

Shuotherium with a lingual cusp in general agreement with

the predicted position and shape.

The purpose of this short note is to discuss the usefulness

of the conventions currently used for naming cusps in the

pseudotribosphenic dentition, and the potential for names to

colour our interpretation of evolutionary scenarios. Here I will

consider what does ‘pseudo’ mean, and which parts of teeth

may consistently be called ‘pseudo’?

Cusp Development

In an embryo, the future tooth surface begins as the interface

between epithelium and mesenchyme cell layers in the tooth

germ. Soon after the initiation of the tooth germ, the primary

enamel knot forms. The enamel knot is the main signalling

centre of the tooth, expressing dozens of genes. Certain gene

products of the enamel knot prevent proliferation of the

epithelium adjacent to the knot, and the proliferation of

surrounding epithelium continues. This differential

proliferation bends the epithelium-mesenchyme interface,

creating a local topological maximum that is the site of a

A.R. Evans

a) Tribosphenic b) Pseudotribosphenic

Fig. 1. Comparison of tribosphenic (a) and pseudo-tribosphenic (b) morphology for upper (top) and lower (bottom) molars, with basins and cusps

labelled according to the nomenclature proposed here. The main upper cusps of both forms are labelled protocone, paracone and metacone, while

the structures associated with the pseudotalonid basin on the lower are suffixed with ‘pseudo’, (a) upper is Peramus and lower is unidentified

lower from Crompton (1971), redrawn from Wang et al. (1998); (b) upper and lower are based on Pseudotribos, redrawn from Luo et al. (2007).

future cusp. At some distance from the primary enamel knot,

additional knots can form, called secondary enamel knots,

that also produce local maxima in the folded interface. The

result is the general topography of the tooth represented by the

epithelial-mesenchyme interface. After folding at each cusp is

completed, mineralisation commences, starting at the cusp tip,

with dentine deposited from the interface towards the interior

of the tooth, and enamel on the outer surface.

While it is conceivable that the developmental-genetic

process may exist such that tooth cusps are in some way

encoded in a gene or genes, there are no unique gene signatures

in any single cusp that has been investigated. In fact, because

of the pleiotropy of genes (the effect of a gene on many

phenotypic features) and the network nature of gene expression

and signalling pathways, most tooth features including cusps

are interlinked in development by shared genes and signalling

pathways. In this sense, the cusps are not independent at the

level of developmental processes (Kangas et al., 2004). The

spacing and timing of each enamel knot controls the relative

position and height of the resulting cusps. This mechanism

appears to be conserved among therian mammals (placentals:

Jernvall and Thesleff, 2000; marsupials: Moustakas et al.,

2011). Therefore, we cannot identify a cusp based on any

particular gene or specific combination of genes, i.e., there is

no ‘protocone gene’. However, there may be a gene or genes

that control aspects such as lingual bias in growth of the upper

tooth. The increase in such a signal may produce sufficient

space for a cusp, which would then be called a ‘protocone’.

Some genes are known to affect cusp formation.

Ectodysplasin ( Eda ) is a tumor necrosis factor (TNF) gene that

is expressed in the developing tooth (Kangas et al., 2004;

Harjunmaa et al., 2014). When the EDA protein is absent, as is

the case in the spontaneous null mutant in the mouse called

Tabby, the molars are simpler and smaller, but when EDA is

overexpressed in the epithelium, they are more complex

compared to the wild type (Kangas et al., 2004; Harjunmaa et

al., 2012). Fine-tuning of the amount of ectodysplasin generates

intermediate tooth shapes, and replicates the order in which

these cusps appeared in evolution (Harjunmaa et al., 2014).

What is ‘Pseudo’ in Pseudotribosphenic Teeth?

Cusp Homology

The lack of specific genes for each cusp and the lability of the

developing tooth to changes in gene products such as EDA

appear to be somewhat at odds with the palaeontological

perspective, which tends to view the positioning and relative

size of major cusps as highly conserved and very stable.

Evolutionary change appears very gradual compared to the

havoc that can be wreaked by the modification in a single gene

like Eda. This implicit view has led to the use of presence/

absence or shape of tooth features as cladistics characters for

phylogenetic reconstruction. If the developmental process

were so labile, then there would be no phylogenetic signature

in tooth cusp patterns at all. The phylogenetic signature in

teeth at high taxonomic level is relatively low, presumably due

to high degrees of homoplasy (such as the repeated evolution

of the hypocone; Jernvall et ah, 1996), but still can be

informative at lower levels.

The use of tooth characteristics in phylogenetic

reconstruction assumes homology among cusps. Homologies

of cusps among tooth forms, and even between upper and

lower teeth, have been proposed for over a century (Osborn,

1888). While it is now clear that it is very unlikely that there is

a simple relationship of ‘homology’ among cusps, naming

conventions have at least in part been based on criteria of

homology. Wang et al. (1998) proposed these to be morphology,

topographic position and occlusal relationships. Based on

differences in topographic position, the lower basins of

tribosphenic and pseudotribosphenic teeth are justified in their

divergent names.

Protocone and the Meaning of ‘Pseudo’

The major lingual cusp on upper molars of pseudotribosphenic

dentitions has been analogised to the protocone, given its

similarity in position, shape and inferred function to the

protocone in tribosphenic dentitions. However, because it

occludes with the basin on the anterior of the lower, the prefix

‘pseudo’ has been used to indicate that it is in some way

different from the standard protocone.

Comparisons between the position and shape of the

protocone and pseudoprotocone reveal a reasonable

concordance between them (fig. 1; see also Wang et al., 1998,

fig. 6). They both fulfil the same function of supporting crests

that occlude with the lingual crests of the talonid basin. Wang

et al. (1998) suggest that Shuotherium was not able to closely

approximate the buccal surface of the pseudoprotocone with

the lingual surface of the pseudotalonid basin, and so may not

be able to ‘crush’ or ‘grind’ as many, but not all, tribosphenic

molars can do (Crompton and Sita-Lumsden, 1970).

Regardless, their overall functional relationships remain the

same, although they are mirrored in the anterior-posterior

axis. In what ways are these lingual cusps different? Since we

currently understand that there is not a unique gene signature

that could distinguish these two, and they are in the same

position of the tooth with approximately the same shape, we

could conclude that there is no major difference in their

development or function. Therefore, I propose that there is no

need to use the qualifier of ‘pseudo’ for the large lingual cusp

on the upper molars of pseudotribosphenic teeth, and that it be

called ‘protocone’.

In the hypothetical upper pseudotribosphenic molar,

Chow and Rich (1982) named the posterobuccal cusp

metacone, using the same topological convention as

tribosphenic molars (as did Luo et al., 2007 for the new

pseudotribosphenic mammal Pseudotribos robustus). Wang

et al. (1998) label this cusp the ‘pseudometacone’, which

occludes between the pseudohypoconid of the opposing

lower tooth and the protoconid of the tooth posterior to it.

The ‘pseudometacone’ of the pseudotribosphenic teeth has

an equivalent position and shape to the metacone of

tribosphenic teeth. The justification for the ‘pseudo’

designation is likely because of its different occlusal

relationships with the lower compared to the tribosphenic

metacone, which occludes in the space between the

hypoconid and protoconid of the same lower tooth. An

equivalent difference in occlusal relationships exists for the

paracone, and so following the same convention it would be

the ‘pseudoparacone’. This shows an inconsistent use of the

‘pseudo’ prefix in exactly what is different or ‘pseudo’ about

the feature. I propose here that the ‘pseudo’ be used only for

the new topographical structure, the anterior basin of the

lower molar, and its associated cusps and crests, such as the

pseudohypoconid and pseudohypolophid (fig. lb).

Importance of Names

Why is it important to reconsider the naming of this cusp, in

what looks like a purely nomenclatural discussion? The term

pseudoprotocone implies some substantive difference from

the protocone, and suggests major developmental and/or

functional distinctions between these cusps.

In order to evolve a tribosphenic-like tooth from a basal

reversed-triangle tooth, three features must be added: a basin

on the lower tooth, a lingual cusp on the occluding upper, and

an additional buccal cusp (either paracone or metacone) also

on the upper. The biggest difference between evolving a

tribosphenic tooth and a pseudotribosphenic tooth is whether

the basin is anterior or posterior. This will affect the shape of

the protocone, paracone and metacone, but the protocone and

two buccal cusps must still be present. From a developmental

perspective, then, the protocone is essentially the same for the

two tooth forms.

Anterior-posterior Bias

Using ‘pseudo’ gives the impression of substantial difference

in shape and function from tribosphenic, while in fact the

differences are relatively minor. It is likely only the anterior-

posterior bias in the lower molar that makes the difference.

Recent developmental experiments show an inherent bias in

the morphogenesis of mouse molars, such that a posterior

extension is more likely than an anterior one (Harjunmaa et

al., 2014; Luo, 2014). It is likely that such a bias existed in the

ancestor to all modern toothed mammals. This begs the

question of whether pseudotribosphenic mammals had an

anterior bias rather than a posterior one. How labile may this

anterior-posterior bias be? Could a switch in the bias have

A.R. Evans

changed several times in the history of mammals? Depending

on the postulated evolutionary relationships among

tribosphenic and pseudotribosphenic mammals, this switch

may have occurred once or several times (Luo et al., 2007;

Rich and Vickers-Rich, 2010).

The origin and evolution of anterior-posterior

developmental bias in lower molar development relative to the

upper appears to be a bigger question than the convergence of

the tribosphenic-like form itself. If a lower molar has a

posterior bias in producing a basin, then it can occlude with a

nascent lingual cusp that can later evolve to become a

protocone. A basin produced by an anterior developmental

bias can also occlude with a nascent protocone.

Currently there are no obvious molecular signals that may

produce this anterior-posterior differential bias in tooth

development, but this is a significant line of enquiry for

future research.

Conclusion

While the tribosphenic tooth is an intricate, precisely-

occluding device (Evans and Sanson, 2003; Evans and Sanson,

2006), equivalent structures have evolved a number of times,

at least in tribosphenic and pseudotribosphenic mammals. But

the difficulty of evolving such a shape may have been

overestimated, and is perhaps exaggerated by the

‘pseudoprotocone’ terminology.

Acknowledgements

I would like to thank Tom Rich for his enthusiasm and

dedication to Mesozoic mammal research in Australia, and

inspiration to generations of Australian palaeontologists and

biologists. I also thank Erich Fitzgerald for inviting me to

contribute to this volume to honour Tom Rich's contributions

to vertebrate palaeontology. Thanks to Zhe-Xi Luo and an

anonymous reviewer for suggestions that refined the ideas

presented here and improved the manuscript.

References

Chow, M., and Rich, T.H.V. 1982. Shuotherium dongi, n. gen. and sp.,

a therian with pseudo tribosphenic molars from the Jurassic of

Sichuan, China. Australian Mammalogy 5: 127-142.

Crompton, A.W. 1971. The origin of the tribosphenic molar. Pp. 65-87

in: Kermack, D.M. and Kermack, K.A. (eds). Early Mammals.

Academic Press: London.

Crompton, A.W., and Sita-Lumsden, A. 1970. Functional significance

of the therian molar pattern. Nature 227: 197-199.

Evans, A.R., and Sanson, G.D. 2003. The tooth of perfection:

functional and spatial constraints on mammalian tooth shape.

Biological Journal of the Linnean Society 78: 173-191.

Evans, A.R., and Sanson, G.D. 2006. Spatial and functional modeling

of carnivore and insectivore molariform teeth. Journal of

Morphology 267: 649-662.

Harjunmaa, E., Kallonen, A., Voutilainen, M., Hamalainen, K.,

Mikkola, M.L., and Jernvall, J. 2012. On the difficulty of

increasing dental complexity. Nature 483: 324-327.

Harjunmaa, E., Seidel, K., Hakkinen, T., Renvoise, E., Corfe, I.J.,

Kallonen, A., Zhang, Z.-Q., Evans, A.R., Mikkola, M.L., Salazar-

Ciudad. I., Klein, O.D., and Jernvall, J. 2014. Replaying

evolutionary transitions from the dental fossil record. Nature 512:

44-48.

Jernvall, J., and Thesleff, I. 2000. Reiterative signaling and patterning

during mammalian tooth morphogenesis. Mechanisms of

Development 92: 19-29.

Jernvall, J., Hunter, J.P, and Fortelius, M. 1996. Molar tooth diversity,

disparity, and ecology in Cenozoic ungulate radiations. Science

274: 1489-1492.

Kangas, A.T., Evans, A.R., Thesleff, I., and Jernvall, J. 2004.

Nonindependence of mammalian dental characters. Nature 432:

211-214.

Luo, Z.-X. 2014. Tooth structure re-engineered. Nature 512: 36-37.

Luo, Z.X., Ji, Q., and Yuan, C.X. 2007. Convergent dental adaptations

in pseudo-tribosphenic and tribosphenic mammals. Nature 450:

93-97.

McKenna, M.C. 1975. Toward a phylogenetic classification of the

Mammalia. Pp. 21-46 in: Luckett, W.P. and Szalay, F.S. (eds),

Phylogeny of the Primates. Plenum Publishing Corporation: New

York.

Moustakas, J.E., Smith, K.K., and Hlusko, L.J. 2011. Evolution and

development of the mammalian dentition: insights from the

marsupial Monodelphis domestica. Developmental Dynamics

240: 232-239.

Osborn, H.F. 1888. The evolution of mammalian molars to and from

the tritubercular type. American Naturalist 22: 1067-1079.

Patterson, B. 1956. Early Cretaceous mammals and the evolution of

mammalian molar teeth. Fieldiana Geology 13: 1-105.

Rich, T., and Vickers-Rich, P. 2010. Pseudotribosphenic: the history of

a concept. Vertebrata PalAsiatica 48: 336-347.

Simpson, G.G. 1936. Studies of the earliest mammalian dentitions.

Dental Cosmos 78: 791-800, 940-953.

Wang, Y.Q., Clemens, W., Hu, Y.M., and Li, C.K. 1998. A probable

pseudo-tribosphenic upper molar from the late Jurassic of China

and the early radiation of the Holotheria. Journal of Vertebrate

Paleontology 18: 777-787.

Memoirs of Museum Victoria 74:97-105 (2016) Published 2016

ISSN 1447-2546 (Print) 1447-2554 (On-line)

http://museumvictoria.com.au/about/books-and-journals/journals/memoirs-of-museum-victoria/

The upper dentition and relationships of the enigmatic Australian Cretaceous

mammal Kollikodon ritchiei

Rebecca Pian 1 - 2 - 3 , Michael Archer 1 *, Suzanne J. Hand 1 , Robin M.D. Beck 1 ’ 4 and Andrew Cody 5

1 PANGEA Research Centre, School of Biological, Earth and E nvironmental Sciences, University of New South Wales,

Sydney, New South Wales 2052, Australia (m.archer@unsw.edu.au)

2 Division of Paleontology, American Museum of Natural History, New York, NY 10024, USA (rpian@amnh.org)

3 Department of Earth and E nvironmental Sciences, Columbia University, New York, NY 10026 USA

4 School of E nvironment & Life Sciences, University of Salford, Salford M5 4WT, England

5 The National Opal Collection, 119 Swanston St, Melbourne, Victoria 3000, Australia

*To whom correspondence should be addressed. E-mail: m.archer@unsw.edu.au

Abstract Pian, R., Archer, M., Hand, S.J., Beck, R.M.D. and Cody, A. 2016. The upper dentition and relationships of the enigmatic

Australian Cretaceous mammal Kollikodon ritchiei. Memoirs of Museum Victoria 74: 97-105.

Mesozoic mammals from Australia are rare, so far only known from the Early Cretaceous, and most are poorly

represented in terms of dentitions much less cranial material. No upper molars of any have been described. Kollikodon

ritchiei is perhaps the most bizarre of these, originally described on the basis of a dentary fragment with three molars. Here

we describe a second specimen of this extremely rare taxon, one that retains extraordinarily specialised upper cheekteeth

(last premolar and all four molars). Each molar supports rows of bladeless, rounded cuspules many of which exhibit apical

pits that may be the result of masticating hard items such as shells or chitin. Reanalysis of the phylogenetic position of this

taxon suggests, based on a limited number of apparent synapomorphies, that it is an australosphenidan mammal and

probably the sister group to Monotremata. This reanalysis also supports the view that within Monotremata, tachyglossids

and ornithorhynchids diverged in the early to middle Cenozoic.

Keywords New South Wales, Mesozoic, Albian, Australosphenida, Monotremata, Ausktribosphenidae.

Introduction

Mesozoic mammals from Australia were unknown until the

description in 1985 of the holotype and only known specimen

of the monotreme Steropodon galmani (Archer et al., 1985), a

partial right dentary with all three molars, from the Early

Cretaceous (Albian) Griman Creek Formation at Lightning

Ridge, New South Wales. This was followed by discovery at

Lightning Ridge of Kollikodon ritchiei (Flannery et al., 1995),

known from a single partial right dentary with three molars in

situ and alveoli for the posterior two premolars and a fourth

molar. Since then, additional Early Cretaceous mammals have

been described from sites in southern Victoria and New South

Wales. These include: from the Aptian Flat Rocks locality in

Victoria, the ausktribosphenids Ausktribosphenos nyktos (Rich

et al., 1997) and Bishops whitmorei (Rich et al., 2001a), the

monotreme Teinolophos trusleri (Rich et al., 1999, 2001b), and

the multituberculate Corriebaatar marywaltersae (Rich et al.,

2009); from the Albian Dinosaur Cove locality in Victoria, the

partial humerus of (the possible monotreme) Kryoryctes

cadburyi (Pridmore et al., 2005); and from Lightning Ridge in

New South Wales a very large, mammal-like isolated tooth

(Clemens et al., 2003). Most recent phylogenetic analyses have

placed ausktribosphenids and monotremes within a larger

Gondwanan radiation termed Australosphenida (Luo et al.,

2001, 2002, 2007a; Martin and Rauhut, 2005; Rougier et al.,

2011; Wood and Rougier, 2005), together with the early Jurassic

(Toarcian) (Cuneo et al., 2013) South American Asfaltomylos

(Rauhut et al., 2002) and Henosferus (Rougier et al., 2007), and

the middle Jurassic (Bathonian) Ambondro mahabo (Flynn et

al., 1999) from Madagascar. Some authors, however, have

questionedthe inclusion of monotremes within Australosphenida

(Pascual et al., 2002; Rich et al., 2002; Rowe et al., 2008;

Woodburne, 2003).

Archer et al. (1985) described Steropodon galmani as a

plesiomorphic, toothed monotreme in the monotypic family

Steropodontidae, an assignation that has been widely accepted

(Kielan-Jaworowska et al., 1987; Musser, 2006; Phillips et al.,

2009). Inclusion of Teinolophos trusleri within Monotremata is

also uncontested, despite some debate over placement within

the stem or crown group (Phillips et al., 2009; Rich et al., 2001b;

Rowe et al., 2008). Flannery et al., (1995) described Kollikodon

ritchiei as a possible monotreme, placing it in its own monotypic

family, the Kollikodontidae. This assignation has proved more

controversial, with suggestions that it may be a basal

R.Pian, M. Archer, S.J. Hand, R.M.D. Beck & A. Cody

mammaliaform rather than a monotreme and as such more

appropriately placed outside crown-group Mammalia (Musser,

2003). This controversy rejects in part the limited morphological

information available on the basis of the previously only known

specimen, a partial dentary with three highly autapomorphic,

bunodont molars. Discovery of an additional specimen of K.

ritchiei, a partial maxilla with one premolar and four molars,

now provides significant additional information about the

structure and likely evolutionary relationships of this enigmatic

taxon. This specimen also represents the first maxilla with teeth

known for any Australian Mesozoic mammal.

Here we describe this specimen and test the evolutionary

relationships of K. ritchei via phylogenetic analysis based on a

comprehensive morphological character matrix.

Materials and Methods

Measurements of the specimen were taken to the nearest 0.01

mm using a Leica M205 C microscope and integrated Leica

DFC290 camera. Tooth lengths were measured along the long

axis of the molar row. Widths were measured from the widest

transverse points across the tooth, perpendicular to the long axis.

Kollikodon ritchiei was added to a revised version of the

character matrix of Luo et al. (2011), which is in itself a

modified version of earlier matrices (Luo et al., 2001, 2002,

2007b; Luo & Wible, 2005). Revisions to the matrix were

made on the basis of corrections and criticisms by Woodburne

et al. (2003), Rougier et al. (2007), Rowe et al. (2008) and

Phillips et al. (2009). The final matrix included 104 taxa and

438 characters. 62 multistate characters representing plausible

morphoclines were ordered. The topologies of the consensus

trees derived through the ordered and unordered analyses

were very similar, and as such only the ordered analyses are

included here (Wiens, 2001).

Maximum parsimony analyses were conducted using the

computer program PAUP* (Phylogenetic Analysis Using

Parsimony and Other Methods) version 4.0bl0 (Swofford,

2002). 1,000 heuristic replicates were initially carried out,

saving 10 trees per replicate, followed by a second heuristic

search within the trees obtained from the first search. Zero-

length branches were then collapsed. Strict and 50% majority-

rule consensus trees were derived from the most parsimonious

trees recovered for each analysis. Bootstrap analysis was used

to assess nodal support. To calculate bootstrap values, 250

bootstrap replicates were run, with a time limit of 60 seconds

per replicate.

Bayesian analyses were conducted using Lewis’s (2001)

Mk likelihood model for discrete morphological data in the

program MrBayes 3.2.1 (Ronquist et al., 2012). Applied

assumptions included scoring of only parsimony informative

characters and gamma distribution that permits rate variation

across different characters. Two independent runs of four

Monte Carlo Markov chains (one cold and three heated) were

run for 5,000,000 generations with trees sampled every 500

generations. Convergence was confirmed by average standard

deviation of split frequencies of less than 0.05. The first 25%

of samples were discarded as “burn-in” and remaining samples

used to construct the 50% majority-consensus.

Systematic Palaeontology

Mammalia Linnaeus, 1758

Australosphenida Luo, Cifelli and Kielan-Jaworowska, 2001

Kollikodontidae Flannery, Archer, Rich and Jones, 1995

Kollikodon Flannery, Archer, Rich and Jones, 1995

Kollikodon ritchiei Flannery, Archer, Rich and Jones, 1995

Holotype. AM F96602 (Australian Museum Palaeontological

Collection, Sydney, Australia,), right dentary fragment

preserving ml-3 and alveoli for two premolars and m4.

Referred specimen. Opalised right skull fragment preserving

part of the maxilla, which retains the posterior premolar (possibly

P4) and Ml-4, and possibly also part of the palatine (Figs 1-2). A

35pm voxel Xradia microCT data file of the specimen has been

lodged with the Museum of Victoria in Melbourne. Detailed 3D

prints of this specimen can be made from the scan data. Solid

casts taken from a mould of the complete upper dentition are also

available; one (AM F140201) is registered in the collections of

the Australian Museum. Although the original specimen, which

is a natural glass cast without internal structure of any kind, is

less informative than the microCT scan data (given that it reveals

structures in undercut areas not visible via conventional

microscopy) and no more informative than 3D prints and hard

casts, this specimen is available for further examination as part

of the National Opal Collection, on application to its Director,

Andrew Cody (andrew@codyopal.com).

Locality and age. Griman Creek Formation; Early Cretaceous

(Middle Albian) (Flannery et al., 1995). The type locality is

claim 30226, Moonshine area of the Cocoran opal field,

Lightning Ridge, New South Wales, Australia (Flannery et al.,

1995). The new skull fragment described here comes from an

unnamed mine on the Cocoran opal field.

New diagnosis ofclade containing Kollikodon and monotremes.

Kollikodon ritchiei and definitive monotremes

(omithorhynchids, tachyglossids, Steropodon and Teinolophos )

differ from other groups variously regarded as australosphenidans

(ausktribosphenids, Ambondro, Asfaltomylos and shuotheriids),

in so far as they are known, in having no paraconid on the first

lower molar, an extremely abrupt discontinuity in size between

the ultimate premolar and the first molar in the upper and lower

dentitions, and in the presence of an enlarged dentary canal.

Revised specific diagnosis. Kollikodon ritchiei is a large (by

Mesozoic standards) mammal (estimated body mass

approximately 1935 g based on ml area (Legendre, 1986)) that

differs from definitive monotremes that retain a functional

dentition in exhibiting the following combination of features:

bunodont molars with no vertical blades (lophs or crests) of any

kind; broadly crescentic upper molars with unique cusp

arrangement; reduced or absent posterior cingula/cingulids on

all molars; markedly convex curve of the buccal edge of the

upper and (to a lesser extent) lower molar rows.

The upper dentition and relationships of the enigmatic Australian Cretaceous mammal Kollikodon ritchiei

Figure 1. Kollikodon ritchiei right maxillary fragment and molar row preserving Px, Ml, M2, M3 and RM4. Stereopair occlusal view, ant,

anterior; bucc, buccal.

Description of the upper dentition and cranial fragment

The new specimen is a right cranial fragment comprising part

of the maxilla and possibly part of the palatine (Figs 1-2). The

maxilla preserves the root of the zygomatic arch and part of

the palatal shelf. The palatine may form part of the posterior

section of this shelf, although no sutures are evident. The

infraorbital canal is exposed on the anterodorsal edge of the

broken maxilla. No complete edges can be identified around

the preserved portion of the palate.

Within the maxilla, the posterior premolar (actual

homology with other mammalian premolars is unknown) and

all four upper molars, Ml-4, are preserved in situ. No inference

can be made about teeth anterior to the posterior premolar

because the maxilla is missing anterior to that point. As in the

lower dentition, there is a stark discontinuity in size between

the premolar and the molar row. The double-rooted,

comparatively simple premolar aligns with the median row of

cusps on the molars. The four fully bunodont molars are

characterized by rows of low, rounded, dome-like cusps that

are separated from each other by arcuate grooves of varying

depth. None of the cusps is subtended by blades although the

enamel edges of pits in the apices of many of the cusps may

have provided horizontal arcuate blades that assisted in

segmenting food during transverse mastication.

The occlusal plane of the upper molar row is

anteroposteriorly convex, corresponding to the concavity of

the occlusal plane of the lower molar row. Although the lingual

margin of the upper tooth row is more or less rectilinear, the

buccal margin is strongly convex, reflecting differences in the

width of the individual molars, with M2-3 being the widest.

When the holotype is placed in centric occlusion with the

upper dentition, the buccal margins of the upper molars

markedly overhang the buccal margin of the lower molars,

resulting in a strongly anisodontic bite. Considering the

molars, M3 has the largest total occlusal surface followed in

descending order by M2, M4 and Ml (Table 1).

100

R.Pian, M. Archer, S.J. Hand, R.M.D. Beck & A. Cody

Figure 2. Kollikodon ritchiei right maxillary fragment and molar row preserving Px, Ml, M2, M3 and M4. (A) Stereopair oblique-occlusal view.

(B) Stereopair lingual view. (C) Stereopair buccal view, ant, anterior.

The upper dentition and relationships of the enigmatic Australian Cretaceous mammal Kollikodon ritchiei

101

Although each molar is distinctly different from the others,

there are common features and meristic trends that progress

posteriorly along the molar row. The molars are transversely

wide. Many of the cusps have pits or depressions at their

apices. The arrangement of cusps on each molar could be

interpreted as forming either a series of two arcuate transverse

rows of cusps, three variably-longitudinal rows of cusps, or

two central anteromedial cusps ringed by a perimeter of 2 to 3

buccal cusps and 1 to 2 lingual cusps. Given the morphological

distinctiveness of each molar, none of these interpretations

applies equally well to all of the individual teeth.

On M2-4 the anterobuccal and anterolingual corner cusps

are anteriorly displaced compared to the anteromedial cusp,

resulting in a concave anterior margin and a convex posterior

margin of the crown. In contrast, Ml has convex margins on

both the anterior and posterior sides of the crown making it

unlike the otherwise crescentic M2-4. The distinction between

the lingual cusps becomes less evident posteriorly such that in

M4 only one anteroposteriorly elongate lingual cusp is

apparent. Ml is the only molar that exhibits a basal anterior

cingulum. In contrast to the holotype in which it can be clearly

seen that each lower molar is double-rooted, the number of

roots for each upper molar and the internal structure of the

maxilla are unclear. There appears to be no preservation of any

anatomical feature within the glass structure of the maxilla.

The apical pits invite speculation that these may have

originally been areas of thin or even absent enamel that, once

breached to expose softer dentine, served to increase the

transverse cutting capacity of the otherwise blade-less molars.

Functionally they would have acted as entrapment devices to

help immobilize items being transversely sheared. Unfortunately,

the amorphous glass constituency of the crowns does not enable

differentiation of enamel and dentine hence this possibility

cannot be tested. Alternatively the pits may be the result of

apical concussive pressure caused by compression against wide,

flat, hardened surfaces such as mollusc shells or crustacean

chitin. If the result of compression, it is perhaps surprising that

all except two (the hemicircular pits in the posterobuccal and

posteromedial cusps of Ml) are nearly circular even at the

extreme buccal edge of the dentition such as the anterobuccal

cusp of M4. If the two hemicircular pits at the posterior margin

of Ml were also originally circular, interproximal wear could

have removed the posterior halves of these pits. However, if this

is the explanation for the hemicircularity, the lack of

corresponding loss of tooth material from the anterior flank of

M2 as well as the lack of posterior wear on the posterior

premolar invite a more detailed analysis of potentially unique

occlusal mechanics in this strange group of mammals.

Phylogenetic Analysis

Maximum parsimony analysis of our morphological character

matrix, with selected multistate characters ordered, recovered

6048 most parsimonious trees (tree length = 2339; consistency

index = 0.35). A simplified 50% majority-rule consensus is given

in Fig. 3A, with dotted lines indicating nodes that collapse in the

strict consensus. In the 50% majority-rule consensus tree,

Kollikodon ritchiei is placed within Australosphenida, forming a

Table 1. Dimensions of the upper right dentition of the original

specimen of Kollikodon ritchiei represented by AM F140201 (in mm).

Tooth

Length

Width

4.49

3.58

6.86

8.77

8.33

13.24

7.74

15.20

5.74

12.01

polytomy with the monotreme Teinolophos trusleri and another

monotreme clade comprising Steropodon galmani, Tachyglossus

aculeatus, Obdurodon dicksoni and Ornithorhynchus anatinus.

The K. ritchiei/ monotreme clade is weakly supported, with a

bootstrap value of 53% and it collapses in the strict consensus.

There is, however, strong support (bootstrap value 80%) for the

clade comprising S. galmani , Ta. aculeatus, Ob. dicksoni and Or.

anatinus. Under strict consensus the australosphenidan clade

collapses, as does the clade placing K. ritchiei and Te. trusleri

with the other monotremes. When K. ritchiei is excluded from

analyses, however, Australosphenida is retained in the strict

consensus and is reasonably well supported compared to other

higher-level clades, with a bootstrap value of 69%; monophyly

of the monotremes is also strongly supported, with a bootstrap

value of 86%.

Bayesian analysis of the same matrix using the Mk model

and a gamma distribution to model rate heterogeneity between

characters (Fig. 3B) resulted in a similar topology to the

maximum parsimony analysis; specifically, K. ritchiei is placed

within Australosphenida as sister to monotremes. Bayesian

posterior probability (BPP) support for a monotrem dKollikodon

clade was moderate at 0.83. Support for Australosphenida was

considerably stronger, with a BPP of 0.95. Unlike in the

maximum parsimony analysis, the position of Te. trusleri was

retained in the Bayesian analysis, with a support value of 0.75.

Strikingly, in contrast to the maximum parsimony tree, Or.

anatinus and Ta. aculeatus were sister-taxa to the exclusion of

Ob. dicksoni , with relatively high BPP of 0.94.

The monotrem dKollikodon clade (as resolved in the

maximum parsimony analysis) is supported by a single

synapomorphy that can be scored for K. ritchiei: the abrupt

disjunction in size between premolars and molars, with the

molars being significantly larger than the premolars in both

groups (character 448: 0 => 1). This character state change

optimizes as a synapomorphy of this clade regardless of

whether accelerated or delayed transformation is assumed,

and occurs along no other branch on the tree (i.e. it has a

consistency index of 1).

Comparisons with other putative australosphenidans are

equally difficult because the molar morphology of K. ritchiei

is so autapomorphic with no resemblance to any of the

tribosphenic or pseudotribosphenic morphologies exhibited

by species of Ambondro, Henosferus, Asfaltomylos,

Ausktribosphenos, Bishops , Pseudotribos or Shuotherium. On

the other hand, features noted above that appear to group K.

102

R.Pian, M. Archer, S.J. Hand, R.M.D. Beck & A. Cody

Thrinaxodon

Massetognathus

Probainognathus

Tritylodontids

Pachygenelus

Adelobasileus

Sinoconodon

Morganucodon

Megazostrodon

Haldanodon

Castorocauda

Hadrocodium

Fruitafossor

Shuotherium

Pseudotribos

Ambondro

Asfaltomylos

Ausktribosphenos

Bishops

Kollikodon

Steropodon

Tachyglossus

Obdurodon

Ornithorhynchus

Teinolophos

Gobiconodon

Repenomamus

Amphilestes

Yanoconodon

Jeholodens

Trioracodon

Priacodon

Haramiyavia

Plagiaulacidans

Cimolodontans

Tinodon

Trechnotheria

i- Thrinaxodon

Probainognathus

Tritylodontids

Pachygenelus

Adelobasileus

- Sinoconodon

1.00r Morganucodon

r Probainogm

I-™

\^\ PaChi

0.93

0.57

PL'

Megazostrodon

_r Haldanodon

Castorocauda

1 00

- Hadrocodium

Fruitafossor

Haramiyavia

Shuotherium

1.00 L Pseudotribos

■ Ambondro

- Asfaltomylos

Ausktribosphenos

7"- Bishops

Kollikodon

Teinolophos

0 7 g, , Steropodon

Obdurodon

0 85 ][ Ornithorhynchus

0.941— Tachyglossus

Jeh

°'®r- Amphilestes

~I_r Gobiconodon

r- Repenomamus

7^ 0 85 _r

0.99L

inodon

Trechnotheria

Plagiaulacidans

Cimolodontans

Figure 3. Results of phylogenetic analyses of 104 taxa and 438 characters, including 62 ordered characters. Kollikodon ritchiei is highlighted in

bold, Australosphenida is indicated by blue shading. 68 taxa are collapsed in the clade Trechnotheria. (A) Maximum parsimony analysis 50%

majority-rule consensus. Dashed lines indicate nodes that collapse under strict consensus of 6048 most parsimonious trees, each with a tree length

of 2336 and consistency index (Cl) of 0.35. Numbers below branches indicate bootstrap values (only those above 50% reported). (B) Bayesian

analysis 50% majority-rule consensus of the post burn-in trees. Nodal support values indicate Bayesian posterior probabilities above 50%.

ritchiei with monotremes are not shared with species in these

other australosphenidan taxa.

Discussion

The specimen described here represents the first maxillary

dentition of any Australian Mesozoic mammal. It can be

confidently referred to Kollikodon ritchiei on the basis of

similarities in the bunodont dentition and other aspects of

molar crown morphology, compatible size, and a close occlusal

fit with the holotype. This specimen confirms that K. ritchiei

is highly autapomorphic and in fact even more unusual than

originally appreciated, with the crescent-shaped upper molars

and cusp arrangement being unique among mammaliaforms

known to date. The original description suggested that K.

ritchiei had at least four lower molars (Flannery et al., 1995).

Based on the presence of four molars in the upper dentition,

four molars were probably also present in the lower dentition.

The numerous apical pits present on the upper molars

combined with the highly bunodont morphology, curvature of

the cusp rows on each crown and marked convexity of the

molar row as a whole suggest an unusual form of occlusion.

None of the cusps is subtended by primary blades, suggesting

that whatever food was eaten was probably crushed although

the circular pits, however they were formed, may have helped

immobilize hard food items that were being masticated in

whichever direction comminution took place.

The known material of K. ritchiei preserves very few

features that are phylogenetically informative and this taxon is

therefore scored for only 26 of 438 characters in our matrix.

The upper and lower cheektooth dentition is well preserved, as

are some aspects of mandibular morphology, but the material

composition of the maxilla (transparent glass with no internal

features preserved) and its missing margins have resulted in

retention of very little additional information about cranial

morphology. Many dental characters in the modified matrix of

Luo et al. (2011) relate to the presence or absence of specific

cusps, as well as their relative position and sizes (if present).

The majority of these characters cannot be scored for K.

ritchiei because the homologies of most of the cusps

(particularly on the upper molars) are unclear. The difficulties

of applying tribosphenic terminology to even tribosphenic-

like monotreme teeth is also an ongoing point of contention

(Woodburne, 2003). A further stumbling block is the limited

The upper dentition and relationships of the enigmatic Australian Cretaceous mammal Kollikodon ritchiei

103

fossil record of other Mesozoic australosphenidans, because

no upper dentitions or maxillae have yet been published.

Nevertheless maximum parsimony and Bayesian methods of

phylogenetic analysis support the inclusion of K. ritchiei

within crown-group Mammalia, in contrast to recent

suggestions that it may have been a stem-mammaliaform

(Musser, 2003). Both methods also place K. ritchiei as sister

group to definitive monotremes within Australosphenida,

albeit with varying degrees of support.

It should be emphasized that only one feature was

recovered as synapomorphic for a Kollikodon/ monotreme

clade in all analyses: the marked disjunction in size between

the posterior premolar and the first molar. This feature occurs

in all toothed monotremes that are represented by adequate

fossils, i.e. Steropodon galmani (Archer et al., 1985) and the

Miocene ornithorhynchid Obdurodon dicksoni (Archer et al.,

1992), in addition to K. ritchiei. As such it appears to be an

unambiguous, uncontradicted synapomorphy for this clade.

The same condition appears to be present in the extant platypus

Ornithorhynchus anatinus (Green, 1937) but, because of

significant uncertainties about its vestigial dentition, it was

scored as unknown in our matrix. Presence of an enlarged

mandibular canal was also identified in the original description

of K. ritchiei (Flannery et al., 1995) as an apomorphy linking

this taxon with definitive monotremes, possibly indicating

sensory elaboration at the front of the face such as occurs in

ornithorhynchids which have many electroreceptors in the

dermis covering the bill.

In contrast to the autapomorphic and extremely bunodont

K. ritchiei , all known toothed monotremes show remarkable

similarity in molar morphology spanning a significant period

of time from the Early Cretaceous S. galmani and T. trusleri

to the Paleocene Monotrematum sudamericanum and the

three Oligocene/Miocene species of Obdurodon (Musser,

2006; Pascual et al., 2002; Pian et al., 2013). The only molar

features that have been identified as characteristic of

monotremes that are also present in K. ritchei include the

presence of very large and wide talonids, and absence of a

cusp in the paraconid position on the first lower molar (Long

et al., 2002). Definitive monotremes that retain an adult

dentition share a number of additional dental apomorphies,

including prominent shelf-like anterior and posterior cingulids

on the lower molars and high transverse, loph-like blades on

the trigonids and talonids present as either a single blade or as

a V-shaped blade. In definitive monotremes for which the

upper dentition is known, this pattern is mimicked on the

upper molars with transverse V-shaped blades. Although the

dentition of Or. anatinus is vestigial and deciduous, the same

pattern appears to be present in the molar remnants found in

juveniles (Green, 1937; Woodburne & Tedford, 1975).

In contrast, Kollikodon ritchiei lacks any traces of vertical

transverse blades, cingula or cingulids. Furthermore, although

the cusps of the lower molars of K. ritchiei can be tentatively

homologised with those of toothed monotremes (Flannery et

al., 1995), we are unable to do this with any confidence for the

cusps of the upper molars. However, it is possible that the

bunodont form of the molars in K. ritchiei evolved from a

relatively more plesiomorphic transverse blade system of the

kind seen in species of Steropodon, Teinolophus,

Monotrematum and Obdurodon. Based on the results of our

formal phylogenetic analyses and pending discovery of

morphologically annectant taxa, we suggest the most

parsimonious hypothesis is that K. ritchiei is a highly

autapomorphic sister-taxon to definitive monotremes.

While in both the maximum parsimony and Bayesian

analyses, Kollikodon fell outside crown-group Monotremata, it

was closer to definitive monotremes than other

australosphenidans. Whether Kollikodon itself should be

considered a monotreme is ultimately dependent on how the

clade Monotremata is defined. A Kollikodon plus definitive

monotreme clade is supported by the following apomorphies:

presence of a partially enlarged mandibular canal, the marked

disjunction in size between the last premolar and the first

molar, large and wide talonids, and absence of a cusp in the

position of a paraconid on the first lower molar. Tentatively we

suggest that, despite these potential synapomorphies, the

otherwise highly autapomorphic and character-ambiguous

Kollikodon should be regarded as a sister group of Monotremata.

One striking difference between our maximum parsimony

and Bayesian analyses is the relationships suggested between

the living platypus Ornithorhynchus anatinus, the fossil

platypus species of Obdurodon, and the living short-beaked

echidna Tachyglossus aculeatus. The maximum parsimony

analysis weakly supported an Ornithorhynchus plus

Obdurodon clade to the exclusion of Tachyglossus, whereas

the Bayesian analysis recovered a strongly supported

Ornithorhynchus plus Tachyglossus clade to the exclusion of

Obdurodon. The latter topology implies that tachyglossids

evolved from a semi-aquatic, billed platypus-like ancestor,

potentially relatively late in the Cenozoic. Further evidence in

support of this hypothesis comes from molecular-based

divergence dates, which estimate that Ornithorhynchus and

tachyglossids diverged 19-48 million years ago (Meredith et

al., 2011; Phillips et al., 2009), and also from estimates of

myoglobin net surface charge in T. aculeatus which suggest

an amphibious ancestry (Mirceta et al., 2013). To date, the

fossil record has provided little additional data bearing on

this issue because all known fossil tachyglossids are

edentulous and obviously ‘echidna-like’. There are no pre-

Pleistocene cranial remains of tachyglossids known with the

exception of Megalibgwilia robustus (also known as

“ Zaglossus” robustus - , see Flannery and Groves, 1998;

Griffiths et al., 1991; Musser, 2006). The single specimen of

M. robustus is commonly presumed to be Miocene in age.

However, there is significant uncertainty about this dating

and a Pliocene age may be more likely (Musser, 2006). If the

latter is the case, there are no pre-Pliocene tachyglossids

known. The phylogenetic analyses presented here also support

this hypothesis in that Obdurodon, Tachyglossus and

Ornithorhynchus form a clade to the exclusion of Steropodon

and Teinolophos, although the precise relationships within

this clade differ depending on the method of analysis. No

support was found for the recent hypothesis (Rowe et al.,

2008) that T. trusleri and S. galmani are crown-group

monotremes closer to Ornithorhynchus than to tachyglossids.

104

R.Pian, M. Archer, S.J. Hand, R.M.D. Beck & A. Cody

Acknowledgements

We thank Dr Chris Telford for assistance in providing advice

about appropriate molding materials that have enabled us to

produce casts of the specimen that is the focus of this research.

We also thank two anonymous referees for their helpful

suggestions about how to improve this contribution.

References

Archer, M., Flannery, T. F., Ritchie, A. and Molnar, R. E. 1985. First

Mesozoic mammal from Australia - an Early Cretaceous monotreme.

Nature 318 (6044): 363-366. doi: 10.1038/318363a0.

Archer, M., Jenkins Jr, F. A., Hand, S. J., Murray, R and Godthelp, H.

1992. Description of the skull and non-vestigial dentition of a

Miocene platypus ( Obdurodon dicksoni n. sp.) from Riversleigh,

Australia and the problem of monotreme origins. Pp.15-27 in:

Augee, M. (ed.). Platypus and Echidnas. Royal Zoological Society

of New South Wales: Sydney.

Clemens, W. A., Wilson, G. P. and Molnar, R. E. 2003. An enigmatic

(synapsid?) tooth from the Early Cretaceous of New South Wales,

Australia. Journal of Vertebrate Paleontology 23: 232-237.

Cuneo, R., Ramezani, J., Scasso, R., Pol, D., Escapa, I., Zavattieri, A. M.

and Bowring, S.A. 2013. High-precision U-Pb geochronology and

a new chronostratigraphy for the Canadon Asfalto Basin, Chubut,

central Patagonia: Implications for terrestrial faunal and floral

evolution in Jurassic. Gondwana Research 24: 1267-1275.

doi:10.1016/j.gr.2013.01.010.

Flannery, T. F., Archer, M., Rich, T. H. and Jones, R. 1995. A new

family of monotremes from the Cretaceous of Australia. Nature

377 6548: 418-420. doi:10.1038/377418a0.

Flannery, T. F. and Groves, C. P. 1998. A revision of the genus Zaglossus

(Monotremata, Tachyglossidae), with description of new species

and subspecies. Mammalia 62: 367-396. doi:10.1515/

mamm.1998.62.3.367.

Flynn, J. J., Parrish, J. M., Rakotosamimanana, B., Simpson, W. F. and

Wyss, A. R. 1999. A Middle Jurassic mammal from Madagascar.

Nature 401: 57-60.

Green, H. L. H. H. 1937. The development and morphology of the teeth

of Ornithorhynchus. Philosophical Transactions of the Royal

Society B: Biological Sciences 228: 367-420. doi:10.1098/

rstb.1937.0015.

Griffiths, M., Wells, R. T. and Barrie, D. J. 1991. Observations on the

skulls of fossil and extant echidnas (Monotremata: Tachyglossidae).

Australian Mammalogy 14: 87-101.

Kielan-Jaworowska, Z., Crompton, A. W. and Jenkins Jr, F. A. 1987.

The origin of egg-laying mammals. Nature 326: 871-873.

Legendre, S. 1986. Analysis of mammalian communities from the Late

Eocene and Oligocene of southern France. Palaeovertebrata 16:

191-212.

Lewis, P. 2001. A likelihood approach to estimating phylogeny from

discrete morphological character data. Systematic Biology 50: 913—

925.

Long, J. A., Archer, M., Flannery, T. F. and Hand, S. J. 2002. Prehistoric

mammals of Australia and New Guinea: one hundred million years of

evolution. University of New South Wales Press: Sydney.

Luo, Z.-X., Chen, P, Li, G. and Chen, M. 2007a. A new eutriconodont

mammal and evolutionary development in early mammals. Nature

446: 288-293. doi: 10.1038/nature05627.

Luo, Z.-X., Cifelli, R. L. and Kielan-Jaworowska, Z. 2001. Dual origin

of tribosphenic mammals. Nature 409: 53-7. doi: 10.1038/35051023.

Luo, Z.-X., Ji, Q. and Yuan, C.-X. 2007b. Convergent dental adaptations

in pseudo-tribosphenic and tribosphenic mammals. Nature 450:

93-97. doi: 10.1038/nature06221.

Luo, Z.-X., Kielan-Jaworowska, Z. and Cifelli, R. L. 2002. In quest for a

phylogeny of Mesozoic mammals. Acta Palaeontologica Polonica

47: 1-78.

Luo, Z.-X. and Wible, J. R. 2005. A Late Jurassic digging mammal and

early mammalian diversification. Science 308: 103-107.

doi: 10.1126/science. 1108875

Luo, Z.-X., Yuan, C.-X., Meng, Q.-J. andJi,Q. 2011. A Jurassic eutherian

mammal and divergence of marsupials and placentals. Nature, 476:

442-445. doi: 10.1038/nature10291

Martin, T. and Rauhut, O. W. M. 2005. Mandible and dentition of

Asfaltomylos patagonicus (Australosphenida, Mammalia) and the

evolution of tribosphenic teeth. Journal of Vertebrate

Paleontology 25: 414-425.

Meredith, R. W., Janecka, J. E., Gatesy, J., Ryder, O. A, Fisher, C. A,

Teeling, E. C., ... Murphy, W. J. 2011. Impacts of the Cretaceous

terrestrial revolution and KPg extinction on mammal

diversification. Science 334: 521-524. doi:10.1126/

science.1211028.

Mirceta, S., Signore, A. V, Burns, J. M., Cossins, A. R., Campbell, K.

L., and Berenbrink, M. 2013. Evolution of mammalian diving

capacity traced by myoglobin net surface charge. Science 340:

1234192. doi:10.1126/science.l234192.

Musser, A. M. 2003. Review of the monotreme fossil record and

comparison of palaeontological and molecular data. Comparative

Biochemistry and Physiology Part A 136: 927-942. doi:10.1016/

S1095-6433.

Musser, A. M. 2006. Furry egg-layers: monotreme relationships and

radiations. Pp. 927-942 in: Merrick, J.R., Archer, M., Hickey,

G.M. and Lee, M. S. Y. (eds). Evolution and Biogeography of

Australasian Vertebrates. Auscipub: Oatlands. 942 pp.

Pascual, R., Goin, F. J., Balarino, L. and Udrizar Sauthier, D. D. 2002.

New data on the Paleocene monotreme Monotrematum

sudamericanum, and the convergent evolution of triangulate

molars. Acta Palaeontologica Polonica 47: 487-492.

Phillips, M. J., Bennett, T. H. and Lee, M. S. Y. 2009. Molecules,

morphology, and ecology indicate a recent, amphibious ancestry

for echidnas. Proceedings of the National Academy of Sciences of

the United States of America 106: 17089-17094. doi:10.1073/

pnas.0904649106.

Pian, R., Archer, M. and Hand, S. J. 2013. A new, giant platypus,

Obdurodon tharalko os child, sp. nov. (Monotremata,

Ornithorhynchidae), from the Riversleigh World Heritage Area,

Australia. Journal of Vertebrate Paleontology 33: 1255-1259. do

i: 10.1080/02724634.2013.782876.

Pridmore, P. A., Rich, T. H. Vickers-Rich, P., and Gambaryan, P. P.

2005. A tachyglossid-like humerus from the Early Cretaceous of

south-eastern Australia. Journal of Mammalian Evolution 12:

359-378. doi:10.1007/sl0914-005-6959-9.

Rauhut, O. W. M., Martin, T., Ortiz-Jaureguizar, E. and Puerta, P.

2002. A Jurassic mammal from South America. Nature 416: 165-

168. doi:10.1038/416165a.

Rich, T. H., Flannery, T. F., Trusler, P., Kool, L., van Klaveren, N. A.

and Vickers-Rich, P. 2001a. A second tribosphenic mammal from

the Mesozoic of Australia. Records of the Queen Victoria Museum

110: 1-9.

Rich, T. H., Flannery, T. T. F., Trusler, P., Kool, L., van Klaveren, N.

A. and Vickers-Rich, P. 2002. Evidence that monotremes and

ausktribosphenids are not sistergroups. Journal of Vertebrate

Paleontology 22: 466-469.

Rich, T. H., Vickers-Rich, P., Constantine, A., Flannery, T. F., Kool, L.

and van Klaveren, N. 1997. A tribosphenic mammal from the

Mesozoic of Australia. Science 278: 1438-1442. doi: 10.1126/

science.278.5342.1438.

The upper dentition and relationships of the enigmatic Australian Cretaceous mammal Kollikodon ritchiei

105

Rich, T. H., Vickers-Rich, R, Constantine, A., Flannery, T. F., Kool, L.

and van Klaveren, N. 1999. Early Cretaceous mammals from Flat

Rocks, Victoria, Australia. Records of the Queen Victoria Museum

106: 1-35.

Rich, T. H., Vickers-Rich, P., Flannery, T. F., Kear, B. P., Cantrill, D.

J., Komarower, P., Kool, L., Pickering, D., Trusler, P., Morton, S.,

van Klaveren, N. and Fitzgerald, E. M. G. 2009. An Australian

multituberculate and its palaeobiogeographic implications. Acta

Palaeontologica Polonica 54: 1-6. doi:10.4202/app.2009.0101.

Rich, T. H., Vickers-Rich, P., Trusler, P., Flannery, T. F., Cifelli, R.,

Constantine, A., Kool, L., van Klaveren, N. 2001b. Monotreme

nature of the Australian Early Cretaceous mammal Teinolophos.

Acta Palaeontologica Polonica 46: 113-118.

Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D. L., Darling, A.,

Hohna, S., Larget, B., Lui, L., Suchard, M. A. and Huelsenbeck, J.

P. 2012. MrBayes 3.2: efficient Bayesian phylogenetic inference

and model choice across a large model space. Systematic Biology

61: 539-542. doi:10.1093/sysbio/sys029.

Rougier, G. W., Apesteguia, S. and Gaetano, L. C. 2011. Highly

specialized mammalian skulls from the Late Cretaceous of

South America. Nature 479: 98-102. doi:10.1038/naturel0591.

Rougier, G. W., Martinelli, A. G., Forasiepi, A. M. and Novacek, M. J.

2007. New Jurassic mammals from Patagonia, Argentina: a

reappraisal of australosphenidan morphology and interrelationships.

American Museum Novitates 3566: 1-54. doi:10.1206/0003-

0082(2007)507[l:njmfpa]2.0.co;2.

Rowe, T., Rich, T. H., Vickers-Rich, P., Springer, M. and Woodburne,

M. O. 2008. The oldest platypus and its bearing on divergence

timing of the platypus and echidna clades. Proceedings of the

National Academy of Sciences of the United States of America

105: 1238-1242. doi:10.1073/pnas.0706385105.

Swofford, D. L. 2002. PAUP*. Phylogenetic Analysis Using

Parsimony (*and Other Methods). Sinauer Associates:

Sunderland, Massachusetts.

Wiens, J. J. 2001. Character analysis in morphological phylogenetics:

problems and solutions. Systematic Biology 50: 689-699.

Wood, C. B. and Rougier, G. W. 2005. Updating and recoding enamel

microstructure in Mesozoic mammals: in search of discrete

characters for phylogenetic reconstruction. Journal of

Mammalian Evolution 12: 433 -460.

Woodburne, M. 2003. Monotremes as pretribosphenic mammals.

Journal of Mammalian Evolution 10: 195-248.

doi:10.1023/B:JOMM.0000015104.29857.f0.

Woodburne, M. O., Rich, T. H. and Springer, M. S. 2003. The

evolution of tribospheny and the antiquity of mammalian clades.

Molecular Phylogenetics and Evolution 28(2): 360-385.

doi:10.1016/S1055-7903(03)00113-l.

Woodburne, M. O. and Tedford, R. H. 1975. The first Tertiary

monotreme from Australia. American Museum Novitates 2588:

1 - 12 .

Memoirs of Museum Victoria 74:107-116 (2016) Published 2016

ISSN 1447-2546 (Print) 1447-2554 (On-line)

http://museumvictoria.com.au/about/books-and-journals/journals/memoirs-of-museum-victoria/

Mysticetes baring their teeth: a new fossil whale, Mammalodon hakataramea ,

from the Southwest Pacific

R. EWAN FORDYCE 1 ’ 2 * (http://zoobank.org/urn:lsid:zoobank.org:author:311048BF-4642-412E-B5DA-E01B8C03B802) AND

Felix G. Marx 1 ’ 3 (http://zoobank.org/urn:lsid:zoobank.org:author: 1791C478-33A7-4C75-8104-4C98C7B22125)

1 Department of Geology, University of Otago, PO Box 56, Dunedin 9054, New Zealand (ewan.fordyce@otago.ac.nz)

2 Departments of Vertebrate Zoology and Paleobiology, National Museum of Natural History, Smithsonian Institution,

Washington DC 20560, USA

3 Department of Geology and Palaeontology, National Museum of Nature and Science, Tsukuba, Japan (felix.marx@

otago.ac.nz)

* To whom correspondence should be addressed. E-mail: ewan.fordyce@otago.ac.nz

http://zoobank.Org/urn:lsid:zoobank.org:pub:7A2CAF55-70DC-4561-AA3D-86FA72C721E6

Abstract Fordyce, R.E. and Marx, F.G. 2016. Mysticetes baring their teeth: a new fossil whale, Mammalodon hakataramea, from

the Southwest Pacific. Memoirs of Museum Victoria 74: 107-116.

A small, toothed fossil cetacean from Hakataramea Valley (South Canterbury, New Zealand) represents a new Late

Oligocene species, Mammalodon hakataramea. The new material is from the Kokoamu Greensand (Duntroonian Stage,

about 27 Ma, early to middle Chattian) of the Canterbury Basin, and thus about 2 Ma older than the only other species

included in this genus, Mammalodon colliveri (Late Oligocene, Victoria, Australia). The anterior pedicle of the tympanic

bulla is not fused to the periotic and resembles that of Delphinidae in basic structure. The teeth show extreme attritional

and/or abrasive wear, which has obliterated the crowns. Like Mammalodon colliveri, M. hakataramea was probably

raptorial or a benthic suction feeder.

Keywords systematics, evolution, stratigraphy, anatomy. New Zealand, Osedax.

Introduction

New Zealand is a notable source of fossil cetaceans (whales,

dolphins) in the Southwest Pacific, with specimens ranging

in age from late Middle Eocene to Pleistocene. Rocks of the

southern Canterbury Basin (Field et al., 1989), in particular,

have produced rare Eocene and more common Late Oligocene

to earliest Miocene cetaceans. These mid-Cenozoic fossils

include representatives of the archaeocete family

Basilosauridae, putative stem neocetes (Kekenodontidae)

and diverse odontocetes, such as kentriodontids and other

putative delphinoids, waipatiids (e.g. Otekaikea Tanaka and

Fordyce, 2014, 2015), squalodontids and Squalodelphis- like

taxa. Mysticetes are represented by a diverse assemblage

comprising Eomysticetidae (e.g. Tohoraata Boessenecker

and Fordyce, 2014a), basal Balaenopteroidea, and enigmatica

(e.g. Horopeta Tsai and Fordyce, 2015). The key cetacean¬

bearing units - the Kokoamu Greensand (see below) and the

Otekaike Limestone - and the localities in and around the

Waitaki Valley that expose them were reviewed in the

aforementioned articles.

Here, we name and describe a new species of the toothed

mysticete Mammalodon, based on specimen OU 22026 from

the southern Canterbury Basin. The fossil is distinct from the

hitherto monotypic Mammalodon colliveri (Late Oligocene,

~23.9-25.7 Ma) of Victoria, Australia (Fitzgerald, 2010), and is

probably close to 27 Ma. Of note, the fossil includes a tympanic

bulla with a well-preserved anterior pedicle, otherwise poorly

described for archaeocetes and archaic Neoceti.

Specimen OU 22026 was listed by Fordyce (1991: 1312) as

Mammalodon sp. and was later mentioned, but not named, in

an abstract (Fordyce and Marx, 2011). Recently, we included

OU 22026 in a total-evidence phylogenetic analysis of extant

and fossil Mysticeti (Marx and Fordyce 2015: fig. 2; see fig. 4

here), which identified it as sister to Mammalodon colliveri,

with Janjucetus hunderi immediately adjacent (basal).

Together, these three species form an expanded

Mammalodontidae, which in turn are closely related to a

diverse range of aetiocetids described only from the North

Pacific. OU 22026 was also sampled for isotopes by Clementz

et al. (2014), who reported 5 13 C and 5 18 0 values for structural

bone carbonate from the bulla that are inconsistent with

(filter-)feeding low in the food chain.

108

R.E. Fordyce & F.G. Marx

15-

□ uatemaiy

'up.

Lwh X.

gravel

Reference section:

Hakataramea [Haughs 1 ]

quarry

Symbols:

up Lwh = upper Whainga roan

lo Lwh = lower Whaingaroan

diffuse shell bed.

■ Lentipecten &

brachiopds

-1 m bioturbated unconformity

"—EM = Earthquakes Marl

early Oligocene

Fig. 1, Locality map and stratigraphy of the Sisters Creek-Homestead Creek area of Hakataramea Valley. The stratigraphic column, right, is

modified from that of Tsai and Fordyce (2015) for “Haughs’ Quarry”, which provides the nearest detailed column to Sisters Creek. Mammalodon

hakataramea came from about the horizon identified as the diffuse shellbed with Lentipecten hochstetteri and brachiopods.

Definitions and terminology

Anatomical terminology follows Mead and Fordyce (2009),

unless indicated.

Methods

All elements were uncovered by hand scraping from soft

matrix. The bulla was recovered fractured but still naturally

associated; it was cleaned and consolidated with cyanoacrylate

region by region. For photography, the bulla, teeth and skull

roof of M. hakataramea and the bulla of Mammalodon colliveri

were coated with sublimed ammonium chloride. Images of the

bulla of M. hakataramea are composites, stacked (using Adobe

Photoshop) from multiple shots at varying foci. Photography

used a Nikon 105 mm micro lens with a D700 or D800 (M.

hakataramea) or D70 ( M. colliveri ) camera body.

Institutional abbreviations

NMV P, Museum Victoria Palaeontology Collection,

Melbourne, Australia. OU, fossil collection in Geology

Museum, University of Otago, Dunedin, New Zealand.

Systematic Palaeontology

Cetacea Brisson, 1762

Neoceti Fordyce and Muizon, 2001

Mysticeti Gray, 1864

Mammalodontidae Mitchell, 1989

Mammalodon Pritchard, 1939

Emended diagnosis of Mammalodon. Small-sized mysticetes

differing from chaeomysticetes in having teeth. Differ from all

toothed mysticetes except Janjucetus in having a foreshortened,

dorsoventrally tall rostrum, a linguiform anterior border of the

supraorbital process, a triangular wedge of the frontal separating

the ascending process of maxilla from the posterolateral margin

of the nasal, a roughly horizontal dorsal profile of the braincase

(relative to the lateral edge of the rostrum) and posteriorly

reclined mandibular cheek teeth; further differ from all other

toothed mysticetes except Janjucetus and Chonecetus in having

a distinctly V-shaped fronto-parietal suture in dorsal view; from

Llanocetus and two previously coded, undescribed archaic

mysticetes (ChM PV4745; OU GS10897) in having both

relatively and absolutely smaller posterior cheek teeth with

proximally fused roots, and an inner posterior prominence of

the tympanic bulla that is subequal to the outer prominence in

posterior view; from all aetiocetids in having a more elongate

intertemporal region and an anteroposteriorly broader coronoid

process of the mandible; from Aetiocetus and Fucaia in lacking

a medially expanded lacrimal and a dorsoventrally constricted

mandible, and in having more robust cheek teeth with distally

separate roots; from Chonecetus in having a broader, less

anteriorly-thrust supraoccipital bearing a well-developed

external occipital crest, and in lacking a parasagittal cleft on the

dorsal surface of the parietal; from Aetiocetus in having a

clearly heterodont dentition and closely-spaced posterior cheek

teeth with well-developed enamel ridges on both the labial and

lingual sides of the crown; from Morawanocetus in having a

much more robust postorbital process of the frontal; from

Ashorocetus in having a less steeply inclined supraoccipital

shield and a somewhat more anteromedially oriented

basioccipital crest; and from the enigmatic Willungacetus in

having a clearly marked orbitotemporal crest extending

posteriorly on to the intertemporal constriction and a rounded

(rather than triangular), less anteriorly-thrust supraoccipital

bearing a well-developed external occipital crest. Finally,

Mammalodon differs from the only other described

mammalodontid, Janjucetus , in having a rostrum with a bluntly

rounded apex and a gently convex lateral profile in dorsal view,

Mysticetes baring their teeth: a new fossil whale, Mammalodon hakataramea, from the Southwest Pacific

109

coalesced alveoli for the upper incisors, a gracile, foreshortened

and dorsoventrally flattened premaxilla, an anteriorly expanded

nasal, a transversely narrow, linguiform ascending process of

the maxilla extending posteriorly as far as the nasal, a more

anteriorly directed orbit, a more laterally oriented postorbital

process, a transversely convex dorsal profile of the parietals

with no salient sagittal crest, a more laterally oriented nuchal

crest, a broadly rounded apex of the supraoccipital shield in

dorsal view, an anterior portion of the tympanic bulla that is

squared (rather than obliquely truncated) in ventral view, an

inner posterior prominence of the tympanic bulla that is

subequal to the outer prominence in posterior view, a straight

and comparatively gracile mandible bearing large mental

foramina, and three upper and four lower molars, all of which

(at least in adult specimens) are affected by heavy occlusal wear.

Remarks. Comparisons of Mammalodon with Ashorocetus

eguchii, Willungacetus aldingensis and Chonecetus sookensis

are currently hampered by the poor state of preservation of the

available material; none of the latter, for example, includes the

tympanic bulla or teeth. Ashorocetus is currently known only

from the posterior portion of a braincase preserving little

surface detail (Barnes et al. 1995). Willungacetus and

Chonecetus sookensis are based on somewhat more complete,

but still highly fragmentary crania having lost their rostra, most

or all of the ear bones and much of the (basicranial) surface

detail (Pledge, 2005; Russell, 1968). Until the discovery of

better material, the comparisons made here are necessarily

provisional. The use of occlusal wear as a potential diagnostic

character may be queried, as this feature may primarily correlate

with age or the foraging environment. Nevertheless, the extreme

wear present in the two species of Mammalodon is unusual

amongst fossil cetaceans as a whole, and highly so in the context

of toothed mysticetes in particular. If tooth wear in Mammalodon

is primarily linked to the manner of occlusion and/or food

preferences, it may record a valid character that can be used for

diagnostic and cladistic purposes. Without evidence to the

contrary, we therefore retain it here as part of the diagnosis.

Mammalodon hakataramea Fordyce and Marx sp. nov.

Zoobank LSID. http://z 00 bank. 0 rg/urn:lsid:z 00 bank. 0 rg:act:

6D960230-3799-4597-BAAD-2537772B99A6

Figs. 2, 3

Holotype. OU 22026 - dorsal part of braincase, comprising

much of the supraoccipital and parts of the parietals and

squamosals, preserved with the original dorsal surface down,

and the bioeroded ventral surface upwards; left tympanic bulla

lacking the posterior process; five teeth with little or no

remnants of the crown. All elements were closely associated,

with no other fossil cetacean remains nearby.

Type locality. Open flat bed of Sisters Creek, 70-80 m

downstream from a prominent limestone bank directly north of

Riverside farmhouse, McHenrys Road, Hakataramea Valley,

South Canterbury (Fig. 1). Field number REF 13-10-87-2. Grid

reference: latitude 44 deg 38 min 30.5 sec, longitude 170 deg 38

min 45.0 sec, or NZMS260 map 140: 232 158. The Geoscience

Society of New Zealand fossil record number is I40/f400. The

locality is 2 km NNW of the informally-named “Haughs’

Quarry,” as shown by Tanaka and Fordyce (2015).

Horizon and age. OU 22026 is from a massive, bioturbated,

calcareous section of the Kokoamu Greensand, where it was

associated with sparse macrofossils including scattered

pectinids ( Lentipecten hochstetteri ) and terebratulid

brachiopods. Lentipecten hochstetteri and the benthic

foraminiferan Notorotalia spinosa indicate the Duntroonian

Stage. Judging from a comparable section in the Greensand at

Haughs’ Quarry, about 2 km to the SSE (fig. 1, right; also, see

Tsai and Fordyce, 2015), the diffuse shellbed of pectinids and

brachiopods is low in the Duntroonian, probably near the base.

The Duntroonian is dated as 25.2-27.3 Ma (Raine et al. 2015)

and OU 22026 is presumed close to 27 Ma, or early Chattian

(Vandenberghe et al., 2012).

Diagnosis. Differs from M. colliveri in having smaller teeth,

an anteroposteriorly longer supraoccipital and a parabolic

nuchal crest that lacks an abrupt anterolateral curve in dorsal

view, as well as in having a tympanic bulla with a more

distinct interprominential notch, a straight medial margin, an

anterolaterally more inflated outer lip, and a deeper

involucrum bearing less developed oblique sulci (without

adjacent nodules).

Etymology. Hakataramea, a Maori name for the valley where

the holotype was collected. Haka, a dance; taramea, a sharp-

spined herb, “spear-grass” (Apiaceae: Aciphylla squarrosa ),

with sweet-smelling gum from the flower stalks. The

name may commemorate a specific incident (Reed and

Dowling, 2010).

Description

Ontogenetic stage. The specimen is probably a mature adult

because of the extreme wear that has mostly obliterated the

tooth crowns (fig. 1A-E). The parieto-occipital suture is open

along parts of the nuchal crest, but this condition is also seen in

adult modern baleen whales (e.g. Balaenoptera acutorostrata,

Miller, 1924: plate 4; Balaenoptera borealis, Andrews, 1916:

plate 41; Megaptera novaeangliae. True 1904: plates 29, 32).

Skull roof. The dorsal roof of the skull (fig. 2 F, G; table 1) is

represented by the ventrally eroded, thin parietals, the

supraoccipital and, at the posterolateral margins, probably the

dorsalmost portions of both squamosals. There is no distinct

Table 1. Measurements of the skull roof of OU 22026, +/- 0.5 mm.

Skull roof, anteriormost parietal to posteriormost

supraoccipital, midline

+134.5

Length of parietals on vertex

+45.0

Length of supraoccipital, midline

+94.0

Width, outer margins of nuchal crest,

+154.5

posteriormost preserved points

(= posterolateral extremities of skull roof)

110

R.E. Fordyce & F.G. Marx

Cf M.cqlfiveri _

J nuchal crest

M. colliveri

dorsal lip of foramen magnum

Fig. 2. A-G, I, holotype of Mammalodon hakataramea, OU 22026; all material coated with sublimed ammonium chloride and lit from upper

left. A-E, individual isolated teeth in labial or lingual view. F, G, I, holotype skull roof of Mammalodon hakataramea. F, I, dorsal view, anterior

towards the top; G, oblique dorsolateral view from the left, anterior towards the lower left. H, holotype skull of Mammalodon colliveri Pritchard,

NMV P199986, dorsal view, not coated with sublimed ammonium chloride. H and I are shown at the same scale to compare the differences in

size and profile between the two Mammalodon species. The two dashed lines show the position of the apex of the nuchal crest and the dorsal lip

of the foramen magnum in M. colliveri ; M. hakataramea is aligned with the upper line.

Mysticetes baring their teeth: a new fossil whale, Mammalodon hakataramea, from the Southwest Pacific

111

medial crest,

anterior pedicle

prominential ridge

lateral plate, anterior pedicle |\/|

a orsal face, anterior pedicle

broken anterior process of malleus

dorsai face, anterior pedicle -

suture for periotic

Fig. 3. A-F, L, M, holotype left tympanic bulla of Mammalodon hakataramea , OU 22026; all views show the bulla coated with sublimed

ammonium chloride and lit from upper left. A, dorsal. B, ventral. C, posterior. D, anterior. E, lateral. F, medial. G-K, holotype right tympanic

bulla of Mammalodon colliveri Pritchard, NMV P199986, with views mirrored for ease of comparison with M. hakataramea-, bulla is coated

with sublimed ammonium chloride and mirrored views show lighting from upper right. G, dorsal. H, ventral. I, medial. J, lateral. K, posterior.

L, M, enlarged views of left tympanic bulla of Mammalodon hakataramea. to show anterior pedicle. L, slightly dorsomedial posterior view. M,

slightly posterior dorsomedial view.

112

R.E. Fordyce & F.G. Marx

interparietal. The dorsal periosteal surfaces are damaged by

patchy bioerosion, which in two places has also led to the

perforation of the supraoccipital. Enough remains to see that

the supraoccipital is longer, from its apex to the margin of

foramen magnum, than in M. colliveri (fig. 2H, I). The gently

concave supraoccipital is raised little (~ 3 mm) above the

parietals, and forms a thin-edged nuchal crest with a parabolic

profile and a smoothly rounded apex in dorsal view (fig. 2F, G).

By contrast, the crest in M. colliveri is more robust, with

abruptly curved anterolateral corners that markedly overhang

the parietals (fig. 2H). A lateral or oblique view (fig. 2G) shows

the nuchal crest gently convex anteriorly but markedly

steepening posteriorly, as if descending toward the posterior

margin of the temporal fossa. Anteriorly, the supraoccipital has

a small, flattened dorsal apex, passing backwards into a short

but well-developed external occipital crest. Posteriorly, the

supraoccipital is raised and thickened in the midline, with the

adjacent surfaces steepening bilaterally; in M. colliveri , such

features are developed near the foramen magnum.

What remains of the parietals suggests that the fused bones

form a wide and smoothly rounded intertemporal region

without any salient sagittal or parasagittal crests, contrasting

with the narrower, dorsally tabular, condition in M. colliveri.

Irregular parasagittal grooves could result from bioerosion, but

are more likely to be sulci associated with parietal foramina.

Poorly-preserved irregularities in the bone surface posteriorly

(fig. 2G) may represent the parieto-squamosal sutures.

Tympanic bulla (fig. 3A-F, L-M; table 2). The left bulla is

slightly crushed, with the outer lip a little compressed ventrally.

The anterior pedicle has been distorted post-mortem and

rotated ventromedially, so that the suture for the anterior bullar

facet of the periotic is steeply dipping, rather than sub¬

horizontal. The posterior process, conical process and the

anterolateral crest of the outer lip are lost.

In dorsal view (fig. 3A), the bulla has a straight medial

profile, a bluntly rounded, slightly squared apex and an inflated

outer lip, with the bone attaining its greatest width at the level

of the sigmoid process. The two posterior prominences or

lobes are separated by the conspicuous, near-perpendicular

interprominential notch (fig. 3A, B). In dorsal or ventral view,

the outer prominence is sharp, passing dorsally into a marked

Table 2. Measurements of left bulla of OU 22026, +/- 0.5 mm

Length, apex adjacent to Eustachian outlet

to apex of outer posterior prominence

56.5

Length, parallel with medial face

55.5

Width, maximum, immediately below sigmoid cleft

38.0

Depth of involucrum, maximum, at anterior margin

of broken base of inner posterior pedicle

25.5

Depth, tip of sigmoid process to ventral surface with

bulla sitting in stable position

38.0

Length, apex adjacent to Eustachian outlet to apex of

sigmoid process

38.5

prominential ridge {sensu Tsai and Fordyce, 2015). The inner

prominence is more smoothly rounded and does not extend as

far posteriorly, which may indicate an anteromedial in situ

orientation of the long axis of the bulla as seen, for example, in

Janjucetus hunderi. In posterior view (fig. 3C, L) a strong,

slightly oblique ridge crosses the inner prominence to reach

the interprominential notch; there is no ridge on the outer

prominence. Ventrally (fig. 3B), the interprominential notch

passes into a median furrow 22-23 mm long, about half the

length of the bulla.

The Eustachian outlet forms a shallow, anteromedially

oriented notch (fig. 3A, M). The adjacent portion of the

involucrum is obliquely flat medially and excavated laterally.

Posteriorly, the involucrum rises and widens via an abrupt,

obliquely oriented step at mid-length. Oblique striae that cross

the involucrum are finer than in M. colliveri (fig. 3F) and not

separated by tubercles laterally. The otherwise smooth dorsal

surface anteriorly on the involucrum was probably covered by a

lobe of the peribullary sinus; this smooth bone extends posteriorly

20+ mm, at least to the oblique “step” in the dorsal profile, and

possibly to the level of the prominent sub-vertical postmortem

crack (fig. 3M). Further posteriorly, the elevated involucral

surface is considerably rougher, suggesting that the peribullary

sinus may not have extended over the involucrum here. In medial

view, the involucrum has a horizontal zone of irregular fine

creases, probably marking tendinous connections to the

basioccipital crest (fig. 3F, M). A large irregular depression on

the posteriormost portion of the involucrum (fig. 3M) is probably

a collapsed cluster of galleries formed by the osteophagous

siboglinid worm Osedax (see Boessenecker and Fordyce, 2014b

for similar occurrences in other New Zealand Oligocene

Cetacea). In posterior view, the involucrum is more prolonged in

a dorsolateral-ventromedial plane than the sub-cylindrical

involucrum of M. colliveri (compare figs. 3C, D and 3K). An

oblique (slightly lateral) dorsal view, not figured here, shows a

slight concavity in the medial profile of the involucrum - much

less than in M. colliveri, in which this concavity is pronounced.

The outer lip preserves a sharp crest at the Eustachian

outlet and becomes anterolaterally inflated as it passes back

towards the anterior pedicle. The anterolateral corner of the

tympanic bulla formed by this inflated portion is more rounded

than in M. colliveri (compare fig. 3B, H). The crest of the outer

lip is broken to reveal the tympanic cavity in dorsal view. The

floor of the latter is smooth and has a marked transverse saddle

about level with the anterior pedicle, behind which the

tympanic cavity deepens markedly, and narrows. The cavity

both undercuts the involucrum and, at its posterior limit, rises

dorsally to excavate it below the inner posterior pedicle.

The anterior pedicle has a narrow, anteroposteriorly long

junction with the outer lip (fig. 3A, M). The junction is cracked,

and it is uncertain if a groove for the chorda tympani nerve

was present. In dorsal or dorsomedial view (fig. 3A, M), the

anterior pedicle has three elongate faces roughly perpendicular

to each other. The original structure and orientations are

interpreted thus: a lateral plate that descends to the outer lip;

an elongate sub-oval dorsal face with a shallow grooved suture

for the periotic; and a descending medial crest, with a groove

that presumably contributes to the origin of the tensor tympani

Mysticetes baring their teeth: a new fossil whale, Mammalodon hakataramea, from the Southwest Pacific

113

muscle (following Tsai and Fordyce 2015). A posterior view

(fig. 3L) shows the silhouetted cross section of the pedicle,

with the three surfaces bounding an elongate V- to U-shaped

ventral groove. On the dorsal face, the suture for the anterior

bullar facet of the periotic is a long, shallow groove (fig. 3L,

M); accordingly, the fovea epitubaria was probably shallow

and flat, rather than saddle-shaped. The dorsal face of the

pedicle does not show any clear region of fusion with the

anterior margin of the mallear fossa, although the lateral edge

of the posterior apex has lost a sub-mm area of surface bone

which might indicate fusion.

Adjacent to the anterior pedicle, the outer lip bears a shallow

vertical groove, but no obvious lateral furrow (fig. 3E). The

mallear ridge is prominent, oriented obliquely and most raised

at its mid-length. Posteriorly, at the inner margin of the sigmoid

process, the ridge passes into a tiny projection representing the

broken anterior process of the malleus (fig. 3M). In anterior or

posterior view (fig. 3D, L), the dorsal profile of the sigmoid

process has three indistinctly separate faces: a medial one,

probably marking the proximity of the malleus; a dorsal one,

possibly apposing the sigmoid fossa of the squamosal; and a

lateral one. The enrolled posterior lip of the sigmoid process

overhangs a sigmoidal cavity {sensu Tsai and Fordyce, 2015)

delimited by a low oblique ridge presumably for the tympanic

sulcus (fig. 3L). In lateral view (fig. 3E), the sigmoid process is

bounded ventrally by a damaged, anteroventrally oblique

sigmoid cleft. The conical process is lost. A strong prominential

ridge ( sensu Tsai and Fordyce, 2015) lateral to the elliptical

foramen is matched by a thickened ridge within the tympanic

cavity. Judging from well-preserved thin flanges, the elliptical

foramen was patent as a narrow opening about 5 mm deep.

Teeth (fig. 2A-E; table 3). Five teeth are represented by roots,

with one tooth (fig. 2B) retaining a tiny possible dentine

remnant of the crown. The other four teeth are worn, exposing

sections through the infilled pulp cavity. The worn, sub-ovate

surfaces curve down both labiolingually and mesiodistally.

Wear exposes lines of arrested growth (growth-layer groups, as

commonly used for Cetacea), marked by alternating lighter and

darker coloured bands in the outer biomineral, which is

presumed to be cementum. Occlusal surfaces were examined

under high magnification, but no wear patterns (grooves,

striations) were apparent. Tooth A of Figure 2 is laterally

compressed and conical. Tooth B is conical, with a double

distal tip, perhaps representing two fused roots. Tooth C is

laterally compressed. Tooth D is grooved on the labial and

lingual faces, perhaps representing two fused roots. Tooth E

has two closely approximated large roots that taper and

converge distally, with a small third root between the two

larger adjacent to the occlusal surface. Tooth E is the only one

that can reasonably be identified to position, as a posterior

premolar or molar.

Discussion

Mammalodon hakataramea is one of relatively few toothed

mysticetes to be formally described, and only the third member

of the highly unusual and seemingly rather localised

mammalodontids (fig. 4). Bianucci et al. (2011), however,

mentioned a potential record of this family from the

Mediterranean. Mammalodon hakataramea is also the first

new species of archaic toothed mysticete from the New Zealand

region to be formally named as such. Other New Zealand

Oligocene Cetacea likely also represent archaic Mysticeti, but

have either been misidentified or remain undescribed. One

example of such material is “ Squalodon ” serratus Davis, 1888,

an isolated cheek tooth that may have belonged to an aetiocetid

(Fordyce, 2008). Another is an Early Oligocene specimen

described by Keyes (1973) as a “proto-squalodont”, but

identified as a basal mysticete by Marx and Fordyce (2015)

based on as-yet undescribed portions of the skull (OU

GS10897). Nevertheless, toothed mysticetes from New Zealand

are rare, and only a handful of potential candidates have been

recovered during Fordyce’s field programme of 30 years.

The holotype of Mammalodon hakataramea shows two

features that are noteworthy in terms of structure and/or

function. First, the anterior pedicle of the bulla reveals details

rarely preserved in basal mysticetes: the long, thin lateral plate

merging with the outer lip, the dorsal face with the suture for

the anterior bullar facet of the periotic and the medial crest, all

of which bound a ventral groove. These structures are readily

homologised with the anterior pedicle, or accessory ossicle, in

extant Delphinidae, e.g. Tursiops truncatus and Globicephala

melas. In the delphinids, the lateral plate descends to the

groove for the chorda tympani. The dorsal face (with a faintly

grooved suture for the periotic) is short and arched

anteroposteriorly to match the saddle-shaped fovea epitubaria.

The medial plate is inflated and nodular (thus partly closing

the ventral groove), and contributes to the origin for the tensor

tympani (see Mead and Fordyce, 2009: fig. 25W). A ventrally-

grooved anterior pedicle and unfused bulla/periotic contact

also occur in at least one kekenodontid and one eomysticetid at

OU, albeit in pedicles broken from the bulla. In the putative

gulp-feeding Late Oligocene mysticetes Mauicetus parki and

Horopeta umarere (both Chaeomysticeti), the dorsal face of

the anterior pedicle is partly fused posteriorly to the periotic.

In addition, the medial ridge is not developed in these taxa, but

extends dorsally up the medial face of the periotic. Accordingly,

there is no ventral groove.

Table 3. Measurements of teeth of OU 22026, +/- 0.5 mm

Tooth of Fig. 2A, maximum length of root

9.5

Tooth of Fig. 2A, maximum diameter

4.0

Tooth of Fig. 2B, maximum length of root

21.0

Tooth of Fig. 2B, maximum diameter

5.5

Tooth of Fig. 2C, maximum length of root

16.5

Tooth of Fig. 2C, maximum diameter

7.0

Tooth of Fig. 2D, maximum length of root

26.0

Tooth of Fig. 2D, maximum diameter

7.0

Tooth of Fig. 2E, maximum length of root

23.5

Tooth of Fig. 2E, maximum diameter, mesiodistal

11.0

114

R.E. Fordyce & F.G. Marx

Zygorhiza kochii

100

Waipatia maerewhenua

pat

hen

Janjucetus hunderi

- Mammalodon colliveri

Mammalodon hakataramea _

Fucaia goedertorum -

Physeter macrocephaius

Mammalodontidae

Chonecetus $ooken$is

Morawanocetus yabukii

OCPC 1178

Aetiocetus cotyialveus

Aetiocetus poiydentatus

Aetiocetus weltoni _

ChM PV4745

OUGS10897

Llanocetus denticrenatus

Micro mysticetus rothauseni -

Yamatocetus canalkulatus

Eomysticetus whitmorei

- Waharoa ruwhenua ,

I- Mi

6 1 iool_T ~~

100 1

Chaeomysticeti

20 10

i . i

Middle 1 Late 1

Eocene

Early 1 Late 1

Oligocene

Early 1 Middle 1 Late

Miocene

Pli.

Pis.

Aetiocetidae

Eomysticetidae

crown Mysticeti

time fMa)

Fig. 4. Relevant detail of phylogeny from Marx and Fordyce (2015: Fig. 2), showing relationships of Mammalodon hakataramea and other

Mysticeti basal to the crown group. Pli., Pliocene; Pis., Pleistocene.

Secondly, the worn occlusal surfaces of the teeth curve

down on to the mesiodistal and labiolingual faces of the roots.

The similar shape and surface detail amongst the teeth suggest

that the wear is not post-mortem bioerosion, but was likely

caused by abrasion. Phylogenetic bracketing (fig. 4) implies

that M. hakataramea actually had functional tooth crowns on

robust teeth in a short rostrum, like its sister taxon M. colliveri.

The rounded worn surfaces in M. hakataramea contrast with

the more clearly planar wear characterising the cheek tooth

crowns in the holotype of M. colliveri (see Fitzgerald, 2010).

Nevertheless, planar attrition cannot be ruled out as a factor

earlier in the ontogeny of OU 22026, and it is possible that the

rounded wear surfaces reflect old age. Extensive attritional

wear might have removed much of the tooth crown, as in M.

colliveri , until proper occlusion, and thus further attrition,

became impossible. Abrasive wear, conversely, could have

continued through on-going contact of the teeth with food or

ingested sediment.

That OU 22026 survived despite having lost functional

tooth crowns argues against tooth-assisted filter feeding, as

was also argued by Fitzgerald (2010). This conclusion is

consistent with the isotopic values reported by Clementz et al.

(2014), which indicate that M. hakataramea fed higher in the

food chain than typical filter feeding mysticetes. We agree

with Fitzgerald (2010) that the extensive wear in Mammalodon

is more easily reconciled with suction feeding than with

raptorial feeding, which depends on functional teeth for

grasping and/or processing large prey; nevertheless, facultative

durophagy or raptorial sarcophagy cannot be ruled out.

Cetacean ecology during the Oligocene was rather

different from today. Modern seas are dominated by one

clade of mysticetes - the gulp-feeding rorquals

(Balaenopteridae) - and two clades of echolocating

odontocetes: the deep-diving, suction-feeding beaked whales

(Ziphiidae) and the ecologically rather plastic dolphins

(Delphinidae). Like the modern species, some of New

Zealand’s Late Oligocene baleen whales (e.g. Mauicetus

parki, Horopeta umarere), probably filter-fed by skimming or

gulping. The long- and narrow-jawed eomysticetids, with no

modern equivalents, could have been skimmers or suction

feeders, but probably not gulp feeders. The small-toothed

mysticetes from the wider Australasian region were rather

disparate, and probably included both raptorial and (benthic)

suction feeders, such as Mammalodon colliveri. Given their

overall similarity and shared extreme tooth wear, M. colliveri

and M. hakataramea were perhaps suction feeders. Late

Oligocene odontocetes were all echolocators, with

assemblages dominated by platanistoid dolphins in contrast

to the dominant delphinids of modern seas. These platanistoids

included both clearly heterodont taxa, such as shark-toothed

dolphins with a robust dentition suitable for crushing food,

and near-homodont forms that - like modern dolphins -

probably swallowed with minimal processing. Remarkably,

kekenodontid archaeocetes coexisted, until about 26 Ma, with

cetaceans of “modern” feeding habit, presumably snap¬

feeding raptorially and without the benefit of echolocation.

Acknowledgements

This work has its origins in a chance meeting between Thomas

H. Rich and R. Ewan Fordyce in 1974, in the office of GA

Tunnicliffe at Canterbury Museum. Fordyce was then a

Mysticetes baring their teeth: a new fossil whale, Mammalodon hakataramea, from the Southwest Pacific

115

student, considering New Zealand-based postgraduate studies

in some mix of zoology and geology; Rich was assessing the

potential to find fossil terrestrial mammals in New Zealand.

Unsurprisingly, Rich raised the topic of vertebrate

palaeontology and, further, he persisted by letter to encourage

Fordyce. In turn, Fordyce started doctoral studies on New

Zealand fossil Cetacea, and ultimately took a job at University

of Otago where he developed a vertebrate palaeontology

research programme, leading to finds such as that reported

here. Thank you, Tom Rich. We also thank the following for

their help in this project: Michael Brosnan, landowner, for

allowing excavation; the McKenzie family for field

accommodation; Andrew Grebneff and Craig Jones for field

work; Andrew Grebneff for preparing the specimen; O.R

Singleton and Neil Archbold for access to Mammalodon

colliveri-, Erich Fitzgerald for discussions on Mammalodon.

We thank the reviewers, Travis Park and Erich Fitzgerald, for

their constructive critiques. Research was supported by a

University of Otago PhD scholarship and a Japan Society for

the Promotion of Science postdoctoral scholarship to FG

Marx, by National Geographic Society field grants 3542-87

and 3657-87 to RE Fordyce, and by a Monash University

Postdoctoral Fellowship which supported Fordyce’s early

study on Mammalodon colliveri.

References

Andrews, R. C. 1916. Monographs of the Pacific Cetacea. II. The sei

whale ( Balaenoptera borealis Lesson). 1. History, habits, external

anatomy, osteology, and relationship. Memoirs of the American

Museum of Natural History 1:289-388.

Barnes, L. G., Kimura, M., Furusawa, H. and Sawamura, H. 1995.

Classification and distribution of Oligocene Aetiocetidae

(Mammalia; Cetacea; Mysticeti) from western North America

and Japan. Island Arc 3:392-431.

Bianucci, G., M. Gatt, R., Catanzariti, S., Sorbi, C. G., Bonavia, R.,

Curmi, and A. Varola. 2011. Systematics, biostratigraphy and

evolutionary pattern of the Oligo-Miocene marine mammals from

the Maltese Islands. Geobios 44: 549-585.

Boessenecker, R. W., and Fordyce, R. E., 2014a. A new eomysticetid

(Mammalia: Cetacea) from the Late Oligocene of New Zealand

and a reevaluation of L Mauicetus’ waitakiensis. Papers in

Palaeontology doi: 10.1002/spp2.1005: 1-34.

Boessenecker, R. W. and Fordyce, R. E. 2014b. Trace fossil evidence

of predation upon bone-eating worms on a baleen whale skeleton

from the Oligocene of New Zealand. Lethaia DOI: 10.1111/

let.12108.

Brisson, M. J. 1762. Regnum animale in classes IX distributum sive

synopsis methodica. Editio altero auctior; Theodorum Haak,

Leiden, Netherlands.

Clementz, M. T., Fordyce, R. E., Peek, S. L. and Fox, D. L. 2014.

Ancient marine isoscapes and isotopic evidence of bulk-feeding

by Oligocene cetaceans. Palaeogeography, Palaeoclimatology,

Palaeoecology 400:28-40.

Davis, J. W. 1888. On fossil fish remains from the Tertiary and

Cretaceo-Tertiary formations of New Zealand. Transactions of

the Royal Dublin Society, series 2 4:1-56.

Field, B. D., Browne, G.H. and Davy, B. W. 1989. Cretaceous and

Cenozoic sedimentary basins and geological evolution of the

Canterbury Region, South Island, New Zealand. New Zealand

Geological Survey basin studies 2:1-94.

Fitzgerald, E. M. G. 2010. The morphology and systematics of

Mammalodon colliveri (Cetacea: Mysticeti), a toothed mysticete

from the Oligocene of Australia. Zoological Journal of the

Linnean Society 158:367-476.

Fordyce, R. E. 1991. A new look at the fossil vertebrate record of New

Zealand; pp. 1191-1316 in P. V. Rich, J. M. Monaghan, R. F. Baird,

and T. H. Rich (eds). Vertebrate palaeontology of Australasia.

Pioneer Design Studio and Monash University, Melbourne.

Fordyce, R. E. 2008. Fossil mammals; pp. 415-428 in Winterbourn,

M. J., Knox, G. A., Burrows, C. J. and Marsden, I. (eds). Natural

history of Canterbury. University of Canterbury Press,

Christchurch.

Fordyce, R. E., and Marx, F.G. 2011. Toothed mysticetes and

ecological structuring of Oligocene whales and dolphins from

New Zealand. Geological Survey of Western Australia, Record

2011/9:33.

Fordyce, R. E., and Muizon, C. de. 2001. Evolutionary history of

whales: a review; pp. 169-234 in Mazin, J.-M. and Buffrenil, V. de

(eds), Secondary adaptation of tetrapods to life in water.

Proceedings of the international meeting, Poitiers, 1996. Verlag

Dr Friedriech Pfeil, Miinchen.

Gray, J. E. 1864. On the Cetacea which have been observed in the seas

surrounding the British Islands. Proceedings of the Zoological

Society of London 1864 (2): 195-248.

Keyes, I. W. 1973. Early Oligocene squalodont cetacean from Oamaru,

New Zealand. New Zealand Journal of Marine and Freshwater

Research 7:381-390.

Marx, F. G., and Fordyce, R.E. 2015. Baleen boom and bust: a

synthesis of mysticete phylogeny, diversity and disparity. Royal

Society Open Science 2:DOI 10.1098/rsos. 140434.

Mead, J. G., and Fordyce, R.E. 2009. The therian skull: a lexicon with

emphasis on the odontocetes. Smithsonian Contributions to

Zoology 627:1-248.

Miller, G. S. 1924. A pollack whale from Florida presented to the

National Museum by the Miami Aquarium Association.

Proceedings of the United States National Museum 66(9): 1-15.

Mitchell, E. D. 1989. A new cetacean from the Late Eocene La Meseta

Formation, Seymour Island, Antarctic Peninsula. Canadian

Journal of Fisheries and Aquatic Science 46:2219-2235.

Pledge, N. S. 2005. A new species of early Oligocene cetacean from

Port Willunga, South Australia. Memoirs of the Queensland

Museum 51:123-133.

Pritchard, G. B. 1939. On the discovery of a fossil whale in the older

tertiaries of Torquay, Victoria. The Victorian Naturalist 55:151-159.

Raine, J. I., Beu, A.G., Boyes, A.F., Campbell, H.J., Cooper, R.A.,

Crampton, J.S., Crundwell, M.P, Hollis, C.J., and Morgans,

H.E.G. 2015. Revised calibration of the New Zealand Geological

Timescale: NZGT2015/1. GNS Science Report 2012/39: 1-53.

Reed, A. W., and Dowling, P. 2010. Place names of New Zealand.

Penguin, Auckland, 502 pp.

Russell, L. S. 1968. A new cetacean from the Oligocene Sooke

Formation of Vancouver Island, British Columbia. Canadian

Journal of Earth Science 5:929-933.

Tanaka, Y., and Fordyce, R.E. 2014. Fossil dolphin Otekaikea marplesi

(latest Oligocene, New Zealand) expands the morphological and

taxonomic diversity of Oligocene cetaceans. PLoS One 9(9):

el07972.

Tanaka, Y., and Fordyce, R.E. 2015. A new Oligo-Miocene dolphin

from New Zealand: Otekaikea huata expands diversity of the

early Platanistoidea. Palaeontologia electronica 18.2.23A: 1-71.

True, F. W. 1904. The whalebone whales of the western North Atlantic

compared with those occurring in European waters with some

observations on the species of the North Pacific. Smithsonian

contributions to knowledge 33: 1-332.

116

R.E. Fordyce & F.G. Marx

Tsai, C.-H., and Fordyce, R.E. 2015. The earliest gulp-feeding

mysticete (Cetacea: Mysticeti) from the Oligocene of New

Zealand. Journal of Mammalian Evolution: DOI 10.1007/sl0914-

015-9290-0.

Vandenberghe, N., Hilgen, F.J., Speijer, R.R, Ogg, J.G., Gradstein,

F.M., Hammer, O., Hollis, C.J., and Hooker, J. J. 2012. Chapter 28

- The Paleogene Period; pp. 855-921 in Gradstein, F.M., Ogg, J.G.,

Schmitz, M., and Ogg, G. (eds). The Geologic Time Scale.

Elsevier, Boston.

Memoirs of Museum Victoria 74:117-136 (2016) Published 2016

ISSN 1447-2546 (Print) 1447-2554 (On-line)

http://museumvictoria.com.au/about/books-and-journals/journals/memoirs-of-museum-victoria/

A late Oligocene waipatiid dolphin (Odontoceti: Waipatiidae) from Victoria,

Australia

Erich M.G. Fitzgerald 1 ’ 2

1 Geosciences, Museum Victoria, GPO Box 666, Melbourne, Victoria 3001, Australia (efitzgerald@museum.vic.gov.au)

2 Department of Vertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington DC

20560, USA

Abstract Fitzgerald, E.M.G. 2016. A late Oligocene waipatiid dolphin (Odontoceti: Waipatiidae) from Victoria, Australia. Memoirs

of Museum Victoria 74: 117-136.

A partial odontocete skeleton comprising isolated teeth, forelimb elements, ribs, and vertebrae is described from

the upper Oligocene (Chattian) Jan Juc Marl of Jan Juc, Victoria, southeast Australia. Its dental and forelimb characters

most closely resemble those of the late Oligocene Waipatia and Sulakocetus from New Zealand and the Caucasus,

respectively; thus the Jan Juc odontocete is referred to an indeterminate species in the family Waipatiidae (Platanistoidea).

This specimen represents the first report of Waipatiidae in Australia, expands the taxonomic diversity of Australian

Oligocene Cetacea, and shows that Waipatiidae occurred in the Chattian cetacean assemblages of both Australia and New

Zealand.

Keywords Platanistoidea, Waipatiidae, dolphin, Paleogene

Introduction

The fossil record of Cetacea (whales and dolphins) in

Australia is meager: not through lack of Cenozoic marine

rock outcrop, which is widespread in southern Australia, but

rather a limited history of systematic research (Fitzgerald,

2004; Fordyce, 2006). Yet, the potential for improving this

meager record, and gaining broader insights into cetacean

evolution, have long been recognized by Thomas H. Rich

(Rich, 1976, 1999; Vickers-Rich and Rich, 1993; Rich in

Warne et al., 2003). Rich developed an awareness of the

potential for research on Australasian fossil Cetacea, first in

New Zealand during National Geographic Society-funded

fieldwork (Rich, 1975; Rich and Rich, 1982), and then in

Australia at the beginning of his career as Curator at the

National Museum of Victoria (now Museum Victoria) in

1974. In both instances, this nascent attention paid to fossil

Cetacea was encouraged by Dr Frank C. Whitmore, Jr., a

United States Geological Survey marine mammal

palaeontologist assigned to the National Museum of Natural

History (Eshelman and Ward, 1994). By November 1975,

Rich had produced a comprehensive inventory of the fossil

Cetacea in the Palaeontology Collection of Museum Victoria.

The following year (1976), Rich with the assistance of Ian

R. Stewart, collected a partially articulated incomplete fossil

cetacean skeleton from the Upper Oligocene Jan Juc Marl at

Jan Juc Beach, Victoria (Figs. 1 and 2). This specimen was

registered in 1978 as Museum Victoria Palaeontology

, fossil, systematics, taxonomy.

Collection (NMV P) 48861 and identified as a “squalodontoid?”

On 7 October 1987, F. C. Whitmore, Jr. examined some of the

homodont anterior teeth of NMV P48861, identifying the

specimen as a “delphinoid”. It was not until 2003 that the

preparation of NMV P48861 was commenced by the author,

resulting in a third (preliminary) attempt at identifying this

fossil as “?Eurhinodelphinidae” (Fitzgerald, 2004: 191).

The aims of this paper are to describe the informative

parts of the skeleton of NMV P48861, resolve its phylogenetic

relationships, and interpret its biogeographic significance.

Until now, the described late Oligocene cetacean assemblage

from Australia has consisted of a probable kekenodontid

archaeocete ( ‘Squalodon’ gambierensis Glaessner, 1955), two

species of toothed mysticete in the family Mammalodontidae

(Mammalodon colliveri Pritchard, 1939 and Janjucetus

hunderi Fitzgerald, 2006), and isolated teeth referred to the

enigmatic odontocete genus Prosqualodon (Fordyce, 1982;

Fitzgerald, 2004). In addition, unnamed odontocete remains

tentatively attributed to the Eurhinodelphinidae have been

described from the fluvio-lacustrine -Upper Oligocene Namba

Formation of northeast South Australia (Fordyce, 1983;

Fitzgerald, 2004). The allocation of NMV P48861 to the

odontocete clade Waipatiidae marks the first record of this

family in Australia, thereby increasing the family-level

diversity of cetaceans known locally from the Paleogene, and

expanding the record of Australian fossil Cetacea.

118

E.M.G. Fitzgerald

5: ^

.SJc'

(dS

Stage

Planktonic

Foraminifera

Bird

Rock

section

Calcareous

Nannofossils

13-

12-

A A

- 4 — <

Mlb

Globoquadrina dehiscens

Puebla

Clay

Discoaster druggi

NN2

\ 1

10-

23 Q 21 4 Ma - V

. —

23.0 Ma

i i

NN1

1 1 1

i i

1 1 1

Reticulofenestra bisecta

NP25

i i

1 1 1

1 1

Zygrhablithus bijugatus

1 1 1

Jan Juc

5 —

Marl

i i i

4 —

i i

24.8 Ma -

1 1 1

1 1

2 —

i i i

1 1

l l l

i i

NMV P48861

L L L

*4 - horizon

Figure 1. Locality map and stratigraphic section of the locality of NMV P48861, Waipatiidae gen. et sp. indet. Dates in bold text are based on

measured 87 Sr/ 86 Sr ratios in McLaren et al. (2009). Planktonic foraminiferal and calcareous nannofossil stratigraphy, and geochronology are

based on Gradstein et al. (2012). Planktonic foraminiferal data from Li et al. (1999), and calcareous nannofossil data from Siesser (1979). The

measured section at Bird Rock is based on McLaren et al. (2009).

Material and Methods

Preparation. Material was prepared at Melbourne Museum

primarily using pneumatic engravers and pin vises fitted with

tungsten carbide rod. Dilute (10%) acetic acid was used to

remove concretionary carbonate surrounding some bones.

Limited areas of resistant matrix were removed using an air-

abrasive machine. Bone was glued with cyanoacrylate and/or

40% Paraloid B-72 ethyl-methacrylate copolymer dissolved in

acetone. A dilute (3%) solution of Paraloid B-72 in acetone

was used as a consolidant.

Photography and measurement. Prior to preparation, archival

photographs showing the exposed bones in the sediment were

made using a 35 mm film Nikon EL SLR. A digital composite

of scans of these photographs is depicted in Fig. 2. All other

photographs were taken with a Nikon D90 DSLR camera and

a 60 mm micro lens. All measurements were made with

vernier calipers.

A late Oligocene waipatiid dolphin (Odontoceti: Waipatiidae) from Victoria, Australia

119

Anatomical terminology. Because all teeth were found isolated

their precise position in the tooth row is unknown, therefore

each tooth is numbered with Roman numerals (I-IX) in

ascending order to indicate its estimated relative position in

the tooth row from most anterior (I) to most posterior (IX).

Due to uncertain homology with the cusps of other mammals,

the term denticle is used instead of cusp for each major

projection on the crown. Denticles (d) are coded as main (md),

anterior (a, numbered away from the md: adl, ad2, etc.), or

posterior (p, numbered away from the md: pdl, pd2, etc.)

following Marx et al. (2015: 16). Postcranial terms follow

Flower (1885) and Schaller (2007).

Institutional abbreviations. LACM, Natural History Museum

of Los Angeles County, Los Angeles; MLP, Museo de La Plata,

La Plata, Argentina; NMV C, Mammalogy Collection,

Museum Victoria, Melbourne; NMV P, Palaeontology

Collection, Museum Victoria, Melbourne; OU, Geology

Museum, University of Otago, Dunedin; USNM, National

Museum of Natural History, Washington, DC.

Systematic Palaeontology

Cetacea Brisson, 1762

Odontoceti Flower, 1865, sensu Flower, 1867

Platanistoidea Gray, 1863, sensu Muizon, 1987

Waipatiidae Fordyce, 1994

Gen. et sp. indet.

“...a primitive eurhinodelphinid odontocete.” (Fitzgerald, 2004:

184)

Referred material. NMV P48861, incomplete skeleton

consisting of: nine isolated teeth; fragments of one cervical and

12 thoracic vertebrae; parts of 16 ribs; left incomplete scapula,

humerus, radius, ulna, two metacarpals, and phalanx; right

(fragmentary) scapula, humerus, radius, ulna, metacarpal, and

phalanx; and fragments of two presumed carpals plus three

phalanges (Figs. 2-11; Tables 1-2). Collected by Thomas H.

Rich and Ian Stewart, 1976.

Locality. Shore platform in intertidal zone, immediately north

of Bird Rock (a prominent stack), western end of Jan Juc Beach,

Jan Juc, Victoria, southeast Australia; near latitude 38° 20'

58" S, longitude 144° 18' 10" E (Fig. 1).

Horizon and age. NMV P48861 was collected as a single large

block (dimensions ~850x520x300 mm) of massive light grey

friable silty sandy glauconitic marl forming the lowermost ~2

m of the Jan Juc Marl exposed at Bird Rock (Unit BR 1 in

Section 4 of Abele, 1979: 23-25) (Fig 1). The sparse associated

macrofossils include molluscs ( Dosinia , Limopsis chapmani,

Notocallista, Ennucula, cf. Tellina, and Turritellidae indet.: T.

A. Darragh, pers. comm. 3 July 2015), bryozoans ( Otionellina

and cf. Lunulites rutella: R. Schmidt, pers. comm. 3 July 2015),

and teleost fish bones.

Table 1. Measurements in mm of NMV P48861, Waipatiidae gen. et

sp. indet.: teeth.

Tooth

crown

height

crown

anteroposterior

length

crown

labiolingual

width

maximum

root

length

10.4+

6.6

5.8

36.6+

7.3+

5.9

4.7

31.8+

III

14.0

6.0

5.0

43.0

12.0

5.7

5.2

27.8+

8.2+

5.9

4.0

22.3+

10.1+

6.2

4.5

23.9+

VII

10.1+

9.1

4.7

21.4+

VIII

8.5+

10.3

6.5

23.2

8.4+

10.4

6.2

21.0

Table 2. Measurements in mm of NMV P48861, Waipatiidae, gen. et

sp. indet.: forelimb elements. Dimensions adapted from Uhen (2004).

Measurements rounded to nearest 0.5 mm. + symbol denotes

measurements of the preserved dimension of an incomplete element.

Scapula

left

right

maximum preserved height

134.0+

maximum preserved length

183.0+

neck of scapula width

46.0

depth of glenoid fossa

8.0

Humerus

left

right

maximum length

147.0

150.0

maximum width of proximal end

66.0+

65.5+

maximum width of shaft

56.0

minimum width of shaft

40.0

42.0

maximum width of distal end

38.0

37.0

maximum transverse diameter of proximal end 69.0

transverse diameter of shaft at mid-length

28.0

transverse diameter of distal end

26.0

25.0

Ulna

left

right

maximum length

141.0+

172.0+

shaft length

99.0+

122.5

olecranon length

68.0+

75.0+

maximum width across olecranon

70.0

64.5+

width of shaft at mid-length

34.0

33.0

maximum width of distal end

41.5+

42.5

Radius

left

right

maximum length

143.0

shaft length

118.0

maximum width of proximal end

28.0+

29.0+

width of shaft at mid-length

35.0

maximum width of distal end

33.5+

120

E.M.G. Fitzgerald

left scapula

cervical

vertebra

left

phalanx

^ left

humerus

right ^

metacarpal

left

metacar pa Is

right

humerus^

-- impression or

position of bone

□

vertebra

forelimb element

□

Figure 2. The original distribution of elements in matrix prior to preparation, NMV P48861, Waipatiidae gen. et sp. indet. Top, block of matrix

enclosing bones as collected in field, prior to preparation. Bottom, tracing of bone outlines in matrix prior to preparation.

A late Oligocene waipatiid dolphin (Odontoceti: Waipatiidae) from Victoria, Australia

121

Figure 3. Anterior teeth I—III of NMV P48861, Waipatiidae gen. et sp. indet. Tooth I, presumed procumbent incisor in: A, labial; B, anterior; C,

posterior; and D, lingual views. Tooth II, anterior tooth in: E, posterior; F, labial; G, anterior; and H, lingual views. Tooth III, right upper anterior

tooth in: I, labial; J, posterior; K, anterior; and L, lingual views. Specimens whitened with ammonium chloride.

122

E.M.G. Fitzgerald

fjlj'//;

Figure 4. Right upper anterior/anterior cheek teeth IV-VI of NMV P48861, Waipatiidae gen. et sp. indet., in labial (A, E, I), lingual (B, F, J),

anterior (C, G, K), and posterior (D, H, L) views. A-D: tooth IV. E-H: tooth V. I-L: tooth VI. Specimens whitened with ammonium chloride.

A late Oligocene waipatiid dolphin (Odontoceti: Waipatiidae) from Victoria, Australia

123

Figure 5. Upper cheek teeth VII-IX of NMV P48861, Waipatiidae gen. et sp. indet., in labial (A, E, I), lingual (B, F, J), anterior (C, G, K), and

posterior (D, H, L) views. A-D: tooth VII, left upper anterior cheek tooth. E-H: tooth VIII, right upper posterior cheek tooth. I-L: tooth IX,

right upper posterior cheek tooth. See Material and Methods for abbreviations. Specimens whitened with ammonium chloride.

124

E.M.G. Fitzgerald

Right ribs

Left ribs

Figure 6. Ribs of NMV P48861, Waipatiidae gen. et sp. indet. in anterior view. 1: first right rib. 2: second left rib. 3: third left rib.

A late Oligocene waipatiid dolphin (Odontoceti: Waipatiidae) from Victoria, Australia

125

thin crest on anterior

margin of radius

scapular neck

head of humerus

— neck of humerus

insertion of M, infraspinatus

broken base of olecranon tuberosity

carpal facet

distal epiphysis

50 mm

supraspinous fossa

acromion

coracoid process

greater tubercle

insertion of M. suprasinpatus

deltoid tuberosity

deltoid

teres major fossa

scapular spine

Figure 7. The left forelimb bones of NMV P48861, Waipatiidae gen. et sp. indet. in lateral view. A: scapula. B: humerus. C: radius. D: ulna.

Specimens whitened with ammonium chloride.

126

E.M.G. Fitzgerald

Figure 8. Scapulae of NMV P48861, Waipatiidae gen. et sp. indet. Left scapula in: A, medial; and B, distal views. C: glenoid region of right

scapula in medial view. Specimens whitened with ammonium chloride.

A late Oligocene waipatiid dolphin (Odontoceti: Waipatiidae) from Victoria, Australia

127

ssS'f. '

head

humerus

neck of

humerus

deltoid

tuberosity

olecranon

facet

.‘V

radial angle

ulnar crest

distal crest

insertion of

M. infraspinatus

intertubercular

sulcus

greater

tubercle

insertion of

M. subscapularis

lesser

tubercle

,t:#v

head of

humerus

insertion of

M, supraspinatus

deltoid

-crest-

ulnar facet

radial facet

Figure 9. Humeri of NMV P48861, Waipatiidae gen. et sp. indet. Left humerus in: A, medial; B, anterior; C, posterior; and D, proximal views.

Right humerus in: E, lateral; and F, medial views. Specimens whitened with ammonium chloride.

128

E.M.G. Fitzgerald

distal epiphysis

50 mm

facet for olecranon

ligament

olecranon

thin crest on

anterior margin

of radius

Figure 10. Radius and ulnae of NMV P48861, Waipatiidae gen. et sp. indet. A: left radius in anterior view. B: right ulna in anterior view. C: right

ulna in medial view. D: left ulna in posterior view. Specimens whitened with ammonium chloride.

A late Oligocene waipatiid dolphin (Odontoceti: Waipatiidae) from Victoria, Australia

129

• *

SrT:

'>i V*

'■W

i ■

jM'.,

10 mm

Figure 11. Metacarpals and phalanx of NMV P48861, Waipatiidae gen. et sp. indet. A: left metacarpal. B: left metacarpal. C: right metacarpal.

D: left phalanx in lateral view. Specimens whitened with ammonium chloride.

Although planktonic foraminifera are rare in the Jan Juc

Marl and rarely age-diagnostic (Li et al., 1999), maximum and

minimum age constraints are available. 40 Ar/ 39 Ar dating of

Angahook Formation basalts underlying the Point Addis

Limestone (laterally equivalent to the Jan Juc Marl) at Aireys

Inlet gave an age of 28.7 ± 0.2 Ma (McLaren et al., 2009). The

oldest age of the Jan Juc Marl based on 87 Sr/ 86 Sr ratios measured

in brachiopods from the lowest 3 m of the Bird Rock section is

27.2 Ma (McLaren et al., 2009). Sphenolithus ciperoensis occurs

in the basal beds of the Jan Juc Marl at Bird Rock, marking the

base of calcareous nannofossil zone NP24 and therefore an age

of <29.62 Ma (Siesser, 1979; Gradstein et al., 2012). Together,

these data suggest the Jan Juc Marl in outcrop is no older than the

Rupelian-Chattian boundary, 28.1 Ma (McLaren et al., 2009).

The contact between the Jan Juc Marl and conformably

overlying Puebla Clay has long been considered to approximate

the Oligocene-Miocene boundary (Abele, 1979; Li et al., 1999;

McLaren et al., 2009). Zygrhablithus bijugatus is absent from

the top ~2.5 m of Jan Juc Marl in the Bird Rock section (Siesser,

1979), its last appearance datum within calcareous nannofossil

zone NP25 at 23.76 Ma (Gradstein et al., 2012). Siesser (1979)

also reported the last occurrence of Reticulofenestra bisecta

about 1 m below the Jan Juc Marl/Puebla Clay contact; the last

appearance datum of this species marking the top of zone

NP25 at 23.13 Ma (Gradstein et al., 2012). The first appearance

datum of Discoaster druggi marks the boundary between

calcareous nannofossil zones NN1 and NN2 (22.82 Ma), and

this species is first recorded in the beds above the Jan Juc Marl/

Puebla Clay contact (Siesser, 1979; Gradstein et al., 2012) (Fig.

1). The planktonic foram Globoquadrina dehiscens, the first

occurrence of which marks the base of zone Mlb (22.44 Ma) in

southern Australia, is first recorded in the basal Puebla Clay

(Li et al., 1999; McGowran et al., 2004; Gradstein et al., 2012).

The evidence from biostratigraphy shows that the Jan Juc Marl/

Puebla Clay contact is between 23.13 and 22.82 Ma, straddling

the Oligocene-Miocene boundary at 23.03 Ma (McLaren et al.,

2009) (Fig. 1). This is corroborated by 87 Sr/ 86 Sr ratios from the

basal Puebla Clay, which give a range of possible ages from

23.89-21.39 Ma (McLaren et al., 2009).

The age of the exposed Jan Juc Marl is therefore most

rigorously constrained to between about 28.10 and 22.82 Ma,

Chattian to earliest Aquitanian. NMV P48861 was collected

from the lowest beds in the Bird Rock section of the Jan Juc

Marl, stratigraphically below the last occurrence of Zygrhablithus

bijugatus, which has a last appearance datum of 23.76 Ma (Fig.

1). This constrains the age of NMV P48861 to between about

28.1 and 23.7 Ma, and therefore within the Chattian.

Diagnosis. An odontocete with: heterodont dentition including

at least one pair of procumbent apical teeth and small double-

rooted posterior cheek teeth with triangular crowns bearing

two or three posterior denticles; a small rod-like coracoid

process of the scapula; an elongated humerus bearing a strongly

salient deltoid tuberosity continuous with a distally-elongated

crest, and a distal end that is distinctly narrower

(anteroposteriorly) than the proximal end of the shaft; a long

and anteroposteriorly narrow radius bearing a transversely thin

crest on its anterior edge; and a well-developed hatchet-shaped

olecranon of the ulna. None of these characters represent

unambiguous synapomorphies of Waipatiidae, but this

combination of characters is found only in taxa assigned to that

clade (see Comparisons below).

Remarks on Platanistoidea. The concept of Platanistoidea used

here is that of Muizon (1987) with emendments by Fordyce (1994)

and Tanaka and Fordyce (2015a); namely that Platanistoidea

includes the living family Platanistidae plus the extinct clades

130

E.M.G. Fitzgerald

Squalodelphinidae, Waipatiidae, Otekaikea, and Squalodontidae.

This definition and taxonomic content of Platanistoidea has been

questioned (Lambert et al., 2014: 988): some recent analyses posit

both Squalodontidae and Waipatiidae as stem odontocetes

(Geisler et ah, 2011,2014; Lambert et ah, 2014,2015; Sanders and

Geisler, 2015); or platanistoids (Murakami et ah, 2012; Tanaka

and Fordyce, 2015a); or exclude squalodontids from Platanistoidea,

but include Waipatiidae in the latter (Tanaka and Fordyce, 2014).

The taxonomic content and phylogenetic position of

Squalodontidae (and the potentially related Prosqualodori) are

enduring problems in cetacean systematics recently reviewed by

Tanaka and Fordyce (2014: 27). Their hypothesis for the content

of Squalodontidae is followed here. For reviews of the taxonomic

content and phylogenetic position of other putative platanistoid

clades (i.e. Allodelphinidae, Dalpiazinidae) see Muizon (1988,

1991,1994), Fordyce (1994), Barnes (2006), Barnes and Reynolds

(2009), and Lambert et ah (2014).

Description

Ontogenetic age. The ossified and smooth articular surfaces

on the scapula and humerus, twinned with the distal epiphyses

of the radius and ulna not being fused, suggests that NMV

P48861 represents at least a sexually mature but physically

immature adult (Class V) according to the qualitative

developmental categories established by Perrin (1975) for the

delphinid Stenella attenuata.

Teeth. NMV P48861 is a heterodont odontocete, with evidence of

at least one pair of procumbent tusk-like anterior teeth. Six single-

rooted teeth (teeth I-VI: Figs. 3-4) and three double-rooted teeth

(teeth VII-IX: Fig. 5) are preserved in isolation. The relative

position of each tooth is identified with reference to Waipatia

maerewhenua (Fordyce, 1994; cast of the holotype OU 22095).

All teeth apart from a presumed tusked incisor (tooth I) and

conical anterior tooth (tooth II) are interpreted as upper teeth on

the basis of their strong lingual recurvature. The tusked incisor

(tooth I: Figs. 3A-D) has a broken crown exposing dentine and a

patent pulp cavity. The enamel-covered crown is subcircular in

cross section, lacks keels, and bears enamel with longitudinal

ridges on its lingual/posterolingual surface. The anterolingual

surface of the crown has a small pyriform wear facet (Fig. 3D).

The enamelocementum boundary extends further basally on the

lingual/posterolingual side of the crown. The elongate and gently

recurved root is missing most of its cementum, exposing dentine.

A conical anterior tooth (tooth II: Figs. 3E-H) has a crown

with an oval cross section, and an oblique apical wear facet on its

lingual aspect. When complete, the crown was probably

relatively short compared to the elongated root. The labial

surface of the crown is smooth, with a keeled posterior edge, and

fine ridges on its preserved posterolingual surface. Immediately

basal to the crown, the single root is slightly waisted, but then

becomes inflated in the anteroposterior and labiolingual planes

before tapering towards the root apex. The labial surface of the

apical one-quarter of the root has a median groove.

Two upper right caniniform anterior teeth (teeth III and

IV: Figs. 3I-L and 4A-D, respectively) bear a crown with a

single conical denticle and a worn crown apex. The crown is

recurved lingually and is somewhat labiolingually inflated at

its base. The anterior and posterior edges are strongly keeled,

and there are fine longitudinal ridges on the labial side of the

crown base. The lingual surface of the crown in tooth III has

diffuse longitudinal ridges (Fig. 3L). The single root

immediately basal to the crown is waisted such that there is a

distinct ‘neck’. Further towards the root apex the root is

labiolingually inflated, then tapers towards the root apex.

An upper right anterior tooth (tooth V: Figs. 4E-H) has a

crown with a single triangular denticle and a worn crown apex.

The relatively small crown is recurved lingually, bears a strongly

keeled posterior edge, and has fine ridges on its posterolabial

and posterolingual surfaces. The enamelocementum boundary

extends further basally at the posterior ends of both labial and

lingual sides of the crown. In labial and lingual views there is a

distinct ‘neck’ immediately basal to the crown. The single root

is strongly labiolingually inflated and bears a median groove on

the labial surface of its preserved apex (Fig. 4E).

An upper right anterior tooth (tooth VI: Figs. 4I-L) has a

crown with a single triangular denticle and a worn crown apex.

The crown is recurved lingually, bears strongly keeled anterior

and posterior edges, and fine ridges on its posterolabial and

lingual surfaces. The enamelocementum boundary extends

further basally at the anterior ends of both labial and lingual

sides of the crown. The crown of this tooth closely approximates

the morphology of the right upper anterior cheek teeth of

Waipatia maerewhenua. The incomplete (presumed) single

root is labiolingually inflated.

A double-rooted upper left anterior cheek tooth (tooth

VII: Figs. 5A-D) has a crown with a high triangular main

denticle (md) bearing keeled anterior and posterior edges, an

incipiently papillate anterolingual cingulum, three tiny

posterior denticles (pdl-3: Fig. 5B), indistinct ridges along

the base of its labial surface, and strong longitudinal ridges

along the base of its lingual surface. A distinct ‘neck’ occurs

basal to the enamelocementum boundary. The two parallel

roots are fused along their entire preserved length, recurved

posterodorsally, and labiolingually inflated at approximately

mid-length. The anterior root tapers strongly towards its apex

such that its preserved apical end is about half the diameter of

the posterior root.

A double-rooted upper right posterior cheek tooth (tooth

VIII: Figs. 5E-H) has a crown with a relatively low triangular

md and two small posterior denticles (pdl-2: Fig. 5E). The md

is heavily worn on its anterior edge and apex. The posterior

denticles are worn on their apices. The posterior edges of all

denticles bear strong keels. The labial surface of the crown

bears indistinct fluted ornament, whereas enamel on the lingual

surface is heavily ornamented with longitudinal ridges and

wrinkles arising from a basal papillate cingulum. A distinct

‘neck’ basal to the enamelocementum boundary can be seen in

labial and lingual views. The two parallel roots are fused for

about three-quarters of their length, recurved posterodorsally

(at an angle of ~60° to the axis of the crown), and strongly

labiolingually inflated in their basal half. Both roots taper

towards their apex, although the apical end of the anterior root

is less than half the diameter of the posterior root. A prominent

elongate swelling on the lingual aspect of the posterior root

probably represents a vestigial fused third root (Fig. 5F).

A late Oligocene waipatiid dolphin (Odontoceti: Waipatiidae) from Victoria, Australia

131

A double-rooted upper right posterior cheek tooth (tooth

IX: Figs. 5I-L) has a crown with a low triangular md and three

posterior denticles (Fig. 51). The posterior denticles decrease in

size away from the md (anteroposterior diameter = 5.4 mm):

pdl anteroposterior diameter = 2.2 mm; pd3 anteroposterior

diameter =1.3 mm. The main denticle plus pdl and pd2 have

heavily worn apices. Additionally, the anterior edge of the md

is worn (Fig. 5K). A distinct shear wear facet occurs on the

lingual surface of the crown at the level (anteroposteriorly) of

the notch between the md and pdl (Fig. 5J). The posterior

edges of all denticles are keeled. The labial surface of the

crown bears distinct ridged and fluted ornament. The enamel

on the lingual surface of the crown is more heavily ornamented

with longitudinal ridges and wrinkles arising from a strongly

papillate basal cingulum. This cingulum wraps around the

anterior and posterior edges of the crown base and on to the

antero- and posterolabial corners of the basal crown. The

crown of this tooth resembles the morphology of the third-to-

last upper cheek teeth of Waipatia maerewhenua. A clear

‘neck’ occurs basal to the enamelocementum boundary. The

two parallel roots are fused for about two-thirds of their length,

recurved posterodorsally (at an angle of ~50° to the axis of the

crown), and strongly labiolingually inflated in their basal half.

There is a prominent anterior bulge at the base of the anterior

root, and both roots are strongly tapered towards their apices.

Vertebrae. The fragmentary spinous processes, right halves of

the vertebral arch, and transverse processes of one cervical

(probably the seventh), and twelve thoracic vertebrae (first to

twelfth) are preserved (Fig. 2). Thoracic vertebrae 1-3 have

high and transversely flat spinous processes, with the spinous

process of thoracic vertebra 1 being approximately half the

width of those of thoracic vertebrae 2 and 3. The rest of the

preserved parts of the vertebrae are uninformative.

Ribs. Parts of 16 ribs, five right, eight left (five of which are

double-headed), and three indeterminate, are preserved (Fig.

6). A partial right rib 1 has a wide and flat shaft (29 mm

maximum and 10 mm minimum diameter proximally), which

increases in width distally (34 mm maximum diameter at

preserved distal end). Three left double-headed ribs (damaged

ventrally) are interpreted as ribs 2, 3, and a mid-series rib

(based on position in the sediment relative to the vertebral

column and comparisons with modern odontocetes, e.g.,

Platanista gangetica NMV C27417 and Delphinus delphis

NMV C24964), and are 262+, 322+, and 284+ mm in chord

length, respectively. Left ribs 2 and 3 are anteroposteriorly flat

and wide along their length (rib 2, 25 mm maximum and 11

mm minimum diameter at mid-shaft; rib 3, 19 mm maximum

and 9 mm minimum diameter at mid-shaft). The left mid-series

rib is narrower and more ovoid in cross-section (18 mm

maximum and 11.5 mm minimum diameter at mid-shaft).

Scapula. Both left and right scapulae are incomplete: the left

scapula lacks the dorsal margin (Figs. 7, 8A), and the right

scapula is represented by an uninformative fragment of dorsal

margin (Fig. 2) plus the coracoid process and approximately

half of the glenoid (Fig. 8C). Orientation of the scapula follows

Tanaka and Fordyce (2015a: 32) whereby the glenoid fossa is

ventral. The scapula is: fan-shaped, its anterior and posterior

edges forming an angle of about 100°; transversely thin

(especially in the middle of the infraspinous fossa); and, by

analogy with other odontocete scapulae (e.g., Benke, 1993;

Muizon, 1994), probably longer than high. Anteriorly, there are

two projections: the acromion and coracoid process.

Viewed laterally (Fig. 7), the long (80+ mm) acromion

projects anteroventrally, has a dorsoventrally high base, and does

not expand distally. In distal view (Fig. 8B), the acromion curves

gently laterally at its base, but more distally curves anteromedially.

The rod-like coracoid process arises from a robust base (8.5 mm

width, 12 mm height) ventromedial to the acromion. The

coracoid process is strongly recurved ventromedially, and long

relative to its transverse diameter (32 mm long; minimum and

maximum diameters of 5.7 mm and 7.6 mm, respectively, at mid¬

length). Viewed distally, the angle between the coracoid process

and acromion is about 40°. The coracoid process is distinctly

waisted about 10 mm from its distal apex, which is slightly

globular (Fig. 8C). The scapular neck is constricted. Distally, the

glenoid fossa has an oval outline, longer than wide (47 mm

length, 35 mm width).

In lateral view (Fig. 7), the base of the acromion is

continuous posterodorsally with the scapular spine, which

curves anterodorsally. Anteriorly, the preserved supraspinous

fossa is anteroposteriorly narrow. It is separated from the

anteroposteriorly broad infraspinous fossa by a ridge with a

tabular lateral surface (anteroposterior diameter 19 mm). The

infraspinous fossa has a smoothly undulating surface. Its

posterior edge is formed by a subtle convexity for the border

between the infraspinous and teres major fossae. The posterior

edge of the scapula has a gently concave profile in lateral view

(angle between posterior edge of the scapula and neck of the

scapula is ~140°). The medial surface of the scapular blade is

dominated by the broad V-shaped subscapularis fossa

(Fig. 8A).

Humerus. The left humerus is nearly complete (Figs. 7, 9A-D),

but the head of the right humerus is eroded (Fig. 9E). Surface

detail on both humeri is generally well preserved. The humerus

is relatively elongated (length >250% of maximum width), and

has a slightly transversely flattened shaft (minimum width of

shaft ~140% of its transverse diameter) (Table 2). The distal

end of the humerus is significantly narrower than the proximal

end (width of distal end of humerus ~57% of its proximal end).

The locations of some muscle attachments on the humerus

differ between odontocete families, and in some cases depart

from their homologues in terrestrial mammals. Notable here is

the insertion for M. deltoideus, which in terrestrial mammals

is a distinct deltoid tuberosity and/or crest (Flower, 1885;

Schaller, 2007). However, in odontocetes the deltoid tuberosity

varies in its relative size and position, and indeed may not be

present at all, hence M. deltoideus inserts on: a distinct deltoid

tuberosity and adjacent crest of humerus in Physeter (Berzin,

1972) and Kogia (Schulte and Smith, 1918); lateral surface of

the distal end of the humerus in Inia (Klima et al., 1980),

Pontoporia (Strickler, 1978), Neophocaena (Howell, 1927),

and Phocoena (Smith et al., 1976); anterior edge and lateral

surface of the humerus in Monodon (Howell, 1930); and the

132

E.M.G. Fitzgerald

anteroventral edge and adjacent lateral surface of the humerus

in Tursiops and Stenella (Benke, 1993). For this study, muscle

attachments are identified using a combination of the

aforementioned literature on odontocete myology, plus

artiodactyls (Nickel et al., 1986; Schaller, 2007).

The proximal end of the humerus is dominated by a

smooth, rounded, head that has a hemi-elliptical outline in

lateral view (Fig. 7), and represents about 30% of the length of

the humerus. Viewed proximally, the head of the humerus is

approximately the same size as the tubercles, from which it is

separated by a deep sulcus (Fig. 9D). In anterior and posterior

views the proximal edges of the head and lesser tubercle are at

approximately the same level, and a distinct neck separates the

head from the body of the humerus (Figs. 9B-C). Medial to

the head, the proximal surface of the lesser tubercle has a

distinct flattened region for insertion of the M. subscapularis.

A distinct intertubercular sulcus separates the lesser tubercle

from the anteriorly adjacent and relatively small greater

tubercle, which has a flattened area on its proximomedial

aspect for insertion of M. supraspinatus that marks a steep step

between the proximal surfaces of the two tubercles. The

insertion of the M. supraspinatus continues posterolaterally

into a deep pit and ventrolaterally angled flattened area.

Further distally, on the lateral surface of the humerus and

below the anterior edge of the head, is a proximodistally long

fossa for the insertion of M. infraspinatus, which terminates in

a deep pit (but not a patent foramen) at the level of the proximal

one-third of shaft length (Fig. 9E). The anterior edge of the

humerus is transversely thin and sigmoidal in lateral/medial

view. A strongly developed and proximodistally long (~40 mm

length) deltoid tuberosity occupies about half of the length and

the maximum width of the shaft. The apex of the deltoid

tuberosity is located within the proximal 65% of the humerus.

The deltoid crest of the humerus runs distally from the deltoid

tuberosity, becoming indistinct proximal to the radial angle

(Figs. 9E-F). Distally, the radial and ulnar facets have gently

undulating surfaces, are separated by a sharp distal crest, and

form an obtuse angle in lateral view (Fig. 7). A low ulnar crest

marks the transition from the distal part of the ulnar facet to its

pentagonal part on the posterior aspect of the humerus (Fig.

9C). Proximomedial to the latter feature is a small, flattened

olecranon facet for attachment of the olecranon ligament.

Radius. The left and right radii are nearly complete, but

somewhat crushed mediolaterally; and the right radius is

corroded and lacks some of its external surface (Figs. 7, 10A).

The shaft is narrow and elongated, in lateral view having a

gently convex anterior edge and slightly concave posterior edge

(Fig. 7). The distal epiphysis is incompletely fused to the shaft.

Proximally, the fovea of the head of the radius has a

quadrangular outline with a distinct concavity at its

anteromedial corner. The surface of the fovea (articular face for

the radial facet of the humerus) is posteromedially-tilted (Fig.

10A). Anteriorly, the shaft bears a thin crest that extends from

the head of the radius distally to the shaft’s mid-length (Fig.

10A). The distal half of the radius widens gradually towards the

distal epiphysis, which is wider than the proximal end. The

carpal facet has an angular distal profile in lateral view (Fig. 7).

Ulna. The left ulna is nearly complete, lacking the

anteroproximal region of the olecranon and part of the distal

end (including epiphysis) (Figs. 7, 10D). The right ulna lacks

the posterior edge of the olecranon, but is otherwise virtually

complete (Figs. 10B-C). The proximal and distal ends of the

ulna are robust (23 mm and 19 mm transverse diameter,

respectively) with the shaft being transversely thin at its mid¬

length (~11 mm); giving the shaft of the ulna a subtly hourglass¬

shaped outline in anterior and posterior views (Figs. 10B, D).

The olecranon projects proximally and posteriorly as a

transversely thin blade. Anteriorly, the olecranon bears a

rugose and proximodistally elongated facet for the olecranon

ligament, located proximal to the hourglass-shaped trochlear

notch (Fig. 10B). Posteriorly, the outer edge of the olecranon

has a rugose surface (Fig. 10D). In lateral view, the trochlear

notch forms a nearly 90° angle, with its vertical part being

transversely narrower (18 mm maximum transverse diameter)

than the horizontal part (22 mm maximum transverse diameter)

(Figs. 7, 10C). Anterodistal to the trochlear notch is a small

tuberosity that fits a notch in the posteroproximal edge of the

radius (Fig. 7). The distal half of the lateral surface of the shaft

bears numerous nutrient foramina of uncertain homology (Fig.

7). The interosseous and posterior borders of the shaft gradually

diverge towards the distal end, to which the ellipsoid epiphysis

is not fused.

Carpals. Two bone fragments (presumed carpals) are

uninformative and are not described.

Metacarpals. Three metacarpals were found in the sediment

during preparation of NMV P48861: two close to the distal end

of the left antebrachium (hence identified as left metacarpals),

and one close to the distal end of the right antebrachium (hence

identified as a right metacarpal) (Figs. 2, 11A-C). Each

metacarpal has: an approximately rhomboid outline, with

concave anterior and posterior edges; transversely convex

lateral surface; and a transversely flattened palmar surface. The

shorter left metacarpal (Fig. 11A) is relatively wide (20.5 mm

maximum width, 36 mm length) and ellipsoid in cross section

(6 mm transverse diameter, 16 mm wide at mid-length). The

longer left metacarpal (Fig. 11B) is elongated (18 mm maximum

width, 39 mm length) and more ovoid in cross section (8 mm

transverse diameter, 12 mm wide at mid-length). The right

metacarpal (Fig. 11C) is nearly identical in size and shape to

the longer left metacarpal. It is not possible to accurately

identify which position each metacarpal occupied in the manus.

Phalanges. Four phalanges were found in the sediment during

preparation of NMV P48861, although only one phalanx is

complete enough to merit description (Fig. 11D). It was found

close to the distal end of the left antebrachium (Fig. 2), and is

hence identified as a left phalanx. It is flattened transversely (5

mm transverse diameter, 12 mm wide at mid-length), and

relatively long (28.5 mm long, 18 mm width at proximal end).

This elongated form, and possession of a wider proximal than

distal (16 mm) end, suggests that this is a proximal phalanx. It

is hourglass-shaped in lateral/plantar views, with flat proximal

and distal ends.

A late Oligocene waipatiid dolphin (Odontoceti: Waipatiidae) from Victoria, Australia

133

Discussion

Comparisons. NMV P48861 differs from archaeocetes by

having relatively tiny heterodont cheek teeth and a humerus that

lacks a trochleated distal end, instead possessing distinct radial

and ulnar facets. NMV P48861 differs from mysticetes

(including toothed stem taxa) (e.g., Fucaia goedertorum (Barnes

and Furusawa in Barnes et al., 1995), LACM 131146; Kellogg,

1965; Benke, 1993; Boessenecker and Fordyce, 2015a) by

having an elongated and narrow rod-like coracoid process of the

scapula, and a humerus that is longer than the antebrachium.

NMV P48861 can be further differentiated from the toothed

mysticete clades: Llanocetidae and Mammalodontidae by

having smaller and lower-crowned cheek teeth lacking strongly

developed ridges on both labial and lingual surfaces of the

crown; and Aetiocetidae by having posterior cheek teeth with

more strongly developed ornament on the labial surface of the

crown. NMV P48861 is not a xenorophid, simocetid, mirocetid,

or agorophiid odontocete, differing by having smaller posterior

cheek teeth. In addition, the humerus of NMV P48861 is more

specialized than that of Mirocetus riabinini Mchedlidze, 1970

(Sanders and Geisler, 2015) by having: a less laterally-projecting

head; a less prominent deltoid crest distal to the deltoid

tuberosity; and distinct radial and ulnar facets on the distal end.

NMV P48861 differs from Prosqualodon by having: relatively

small posterior cheek teeth that lack strong nodular crown

ornament; an elongated coracoid process on the scapula; a

humerus with a straight posterior edge (viewed laterally) and a

strongly developed deltoid tuberosity; and a more elongated

antebrachium. NMV P48861 differs from crown odontocetes

other than Platanistoidea in lacking: homodont conical posterior

teeth, a distal end of the humeral shaft with an anteroposterior

width > to that of the proximal end of the shaft, and a strongly

developed coracoid process that enlarges distally. Although the

coracoid process of the scapula is reduced in eurhinodelphinids

to a rod-like form ( Xiphiacetus bossi Kellogg, 1925, USNM

11867, Muizon, 1994), NMV P48861 further differs from this

family by lacking a distinct crest between the infraspinatus

fossa and teres major fossa on the scapula.

Based on these comparisons and the character combinations

described above, NMV P48861 is assigned to the Platanistoidea.

Within Platanistoidea, NMV P48861 differs from all taxa

other than Squalodontidae and Waipatiidae in having

heterodont dentition and double-rooted posterior cheek teeth.

However, it differs from Squalodontidae by lacking large

robust teeth. NMV P48861 lacks the two scapular characters

proposed as synapomorphies of Platanistoidea: the acromion

positioned on the leading (anterior) edge of the scapula,

resulting in loss of the supraspinous fossa; and absence of the

coracoid process (e.g., Muizon, 1987, 1994). However, several

taxa hypothesized to be platanistoids possess both a

supraspinous fossa and a distinct coracoid process on the

scapula: the squalodontid Phoberodon arctirostris Cabrera,

1926 (Cozzuol and Humbert-Lan, 1989; Cozzuol, 1996; MLP

5-4); Otekaikea spp. (Tanaka and Fordyce, 2014, 2015a); and

Sulakocetus dagestanicus Mchedlidze, 1976 (Mchedlidze,

1984; Muizon, 1987). Hence, the scapular characters of

Muizon (1987) may be synapomorphies of a more exclusive

clade within Platanistoidea (i.e. Squalodelphinidae +

Platanistidae) and/or evolved independently in Prosqualodon

and Squalodon. NMV P48861 shares tusk-like anterior teeth

and a rod-like morphology of the coracoid process with

Otekaikea, but differs from that genus by having: more

strongly heterodont cheek teeth with lower, less conical

crowns bearing salient posterior denticles; a scapula with a

posteroventral border forming a 45° angle with the horizontal

in lateral view (cf. ~15° in Otekaikea)-, a more elongated

humerus (minimum anteroposterior width of shaft is <30%

humerus length); the dorsal edge of the head of the humerus

approximately level with the dorsal edge of the lesser tubercle;

an infraspinous fossa that does not terminate distally in a

distinct ovoid pit on the lateral surface of the humeral shaft;

and a longer antebrachium (length of radius is nearly equal to

humerus length).

Amongst described platanistoids, NMV P48861 is most

similar to Waipatia in having heterodont dentition including:

tusk-like anterior teeth; and double-rooted posterior upper

cheek teeth with small (<12 mm length) triangular crowns

bearing two or three posterior denticles. NMV P48861 differs

from Waipatia maerewhenua in its posterior upper cheek teeth

having finer and more diffuse ridges on the labial surface of the

crown. NMV P48861 differs from W. hectori (Benham, 1935)

by having larger and less labiolingually inflated cheek teeth

with shorter and more shallowly notched denticles. Neither

described species of Waipatia are known from appendicular

elements (Fordyce, 1994; Tanaka and Fordyce, 2015b), so it is

unclear whether Waipatia possessed forelimb morphology

similar to that of Otekaikea and NMV P48861. However, the

holotype of Sulakocetus dagestanicus, which is probably a

waipatiid (Fordyce, 1994, 2003; Fordyce and Muizon, 2001),

includes much of the forelimb skeleton (Mchedlidze, 1984;

Pilleri, 1986). NMV P48861 shares with Sulakocetus: small

heterodont cheek teeth; coracoid process of the scapula present

and apparently long and rod-like (Mchedlidze, 1984:43, Plate

XVI); elongated humerus (minimum anteroposterior width of

shaft is <30% humerus length); dorsal edge of the head of the

humerus approximately level with the dorsal edge of the lesser

tubercle; distinct intertubercular sulcus on humerus

(Mchedlidze, 1984:43, Plate XII); strongly salient deltoid

tuberosity with adjacent crest developed distally; a distal end of

the humeral shaft with an anteroposterior width less than that

of the proximal end of the shaft; and a radius with a transversely

narrow crest on its anterior edge. NMV P48861 differs from

Sulakocetus by having: somewhat larger humerus, radius and

ulna; a head of the humerus subequal in size to the lesser

tubercle; and a relatively longer and narrower radius. Because

NMV P48861 possesses a combination of dental and forelimb

characters only recorded in Waipatia or Sulakocetus, and lacks

any synapomorphies that link this specimen with other

odontocete clades, it is referred to an indeterminate species in

the family Waipatiidae. A modern redescription and

phylogenetic analysis of Sulakocetus (to test its relationship

with Waipatia ), plus discovery of forelimb bones referable to

Waipatia, are required to test the relationships of NMV P48861

hypothesized here.

134

E.M.G. Fitzgerald

Biogeography. NMV P48861 represents the first evidence of

Waipatiidae from Australia. Previously reported records of

waipatiids include Waipatia maerewhenua and W. hectori

from the late Chattian of New Zealand (Fordyce, 1994;

Tanaka and Fordyce, 2015b), plus the potential waipatiids

Sulakocetus dagestanicus from the late Chattian of Caucasus

(Mchedlidze, 1976, 1984) and Sachalinocetus cholmicus

Dubrovo in Siryk and Dubrovo, 1970 from the early Miocene

of Sakhalin. In addition, rostral and mandibular fragments

with teeth, as well as isolated periotics, referred to Waipatiidae

were described from the early Miocene of Malta (Bianucci et

al., 2011). Given this geographic and stratigraphic distribution,

the occurrence of Waipatiidae in late Oligocene strata of

southeast Australia is not surprising and indeed was

anticipated by Fordyce (2006: 766).

Nevertheless, the waipatiid from the Jan Juc Marl is only

the second odontocete taxon recognized from the Oligocene

of Australia, the first, and hitherto only, recorded odontocete

being Prosqualodon (represented by isolated teeth: Hall, 1911;

Fordyce, 1982; Fitzgerald, 2004). Other cetaceans in this

assemblage include a probable kekenodontid archaeocete

{‘‘Squalodori’ gambierensis : Fordyce, 2004; Fitzgerald, 2004),

and several small-bodied toothed mysticetes in the family

Mammalodontidae (Fitzgerald, 2006, 2010, 2012). Each of

these families also occurs in the late Oligocene of New

Zealand (Fordyce, 1984, 1991, 2003; Fordyce and Marx, this

volume), suggesting a generally similar cetacean fauna

throughout the southwest Pacific that lacks confirmed records

of taxa typical of Oligocene assemblages elsewhere, e.g.,

Aetiocetidae (North Pacific) and Xenorophidae (North

Atlantic) (Fordyce, 2003). Despite the family-level taxonomic

similarities between the late Oligocene cetacean assemblages

of Australia and New Zealand, a notable disparity lies in the

numerical dominance (and taxonomic richness) of toothed

mysticete fossils in Australia versus the rarity of their remains

in New Zealand (Fordyce and Marx, this volume). Furthermore,

whereas fossils of Eomysticetidae and other Chaeomysticeti

are relatively abundant and diverse in the late Oligocene of

New Zealand (Boessenecker and Fordyce, 2015a-c; Tsai and

Fordyce, 2015), they have not yet been recognized from

southeast Australia. However, with continuing research, the

absence in Australia of cetacean families recorded in the New

Zealand Oligocene will likely become more apparent than

real—as exemplified by the waipatiid described here.

Acknowledgements

T. Park, D. Pickering, A. Werner, and T. Ziegler are thanked

for finishing preparation of NMV P48861. R. E. Fordyce and

A. Grebneff (University of Otago) provided casts of OU

22095, the type specimen of Waipatia maerewhenua. L.

Barnes and S. McLeod (Natural History Museum of Los

Angeles County), D. Bohaska (National Museum of Natural

History, Smithsonian Institution), I. von Lichtan (University of

Tasmania), and M. Reguero (Museo de La Plata) are thanked

for providing access to specimens in their care. Part of this

research was carried out during a Smithsonian Postdoctoral

Fellowship at the National Museum of Natural History. R. E.

Fordyce and O. Lambert carefully and constructively reviewed

the manuscript. Tom Rich encouraged the author’s interest in

Australian fossil Cetacea, for which he is thanked.

References

Abele, C. 1979. Geology of the Anglesea area, central coastal Victoria.

Memoir of the Geological Survey of Victoria 31: 1-71.

Barnes, L.G. 2006. A phylogenetic analysis of the superfamily

Platanistoidea (Mammalia, Cetacea, Odontoceti). Beitrage zur

Palaontologie 30: 25-42.

Barnes, L.G., and Reynolds, R.E. 2009. A new species of early

Miocene allodelphinid dolphin (Cetacea, Odontoceti,

Platanistoidea) from Cajon Pass, southern California, U.S.A.

Museum of Northern Arizona Bulletin 65: 483-507.

Barnes, L.G., Kimura, M., Furusawa, H., and Sawamura, H. 1995.

Classification and distribution of Oligocene Aetiocetidae

(Mammalia; Cetacea; Mysticeti) from western North America

and Japan. The Island Arc 3: 392-431. [For 1994.]

Benham, W.B. 1935. The teeth of an extinct whale, Microcetus hectori

n. sp. Transactions of the Royal Society of New Zealand 65:

239-243.

Benke, H. 1993. Investigations on the osteology and the functional

morphology of the flipper of whales and dolphins (Cetacea).

Investigations on Cetacea 24: 9-252.

Berzin, A.A. 1972. The Sperm Whale (Kashalot). Israel Program for

Scientific Translations, Jerusalem. 394 pp. [Translated from

Russian.]

Bianucci, G., Gatt, M., Catanzariti, R., Sorbi, S., Bonavia, C.G.,

Curmi, R., and Varola, A. 2011. Systematics, biostratigraphy and

evolutionary pattern of the Oligo-Miocene marine mammals from

the Maltese Islands. Geobios 44: 549-585.

Boessenecker, R.W., and Fordyce, R.E. 2015a. A new genus and

species of eomysticetid (Cetacea: Mysticeti) and a reinterpretation

of Mauicetus ’ lophocephalus Marples, 1956: Transitional baleen

whales from the upper Oligocene of New Zealand. Zoological

Journal of the Linnean Society 175: 607-660.

Boessenecker, R.W., and Fordyce, R.E. 2015b. A new eomysticetid

(Mammalia: Cetacea) from the late Oligocene of New Zealand

and a re-evaluation of Mauicetus waitakiensis'. Papers in

Palaeontology 1: 107-140.

Boessenecker, R.W., and Fordyce, R.E. 2015c. Anatomy, feeding

ecology, and ontogeny of a transitional baleen whale: a new genus

and species of Eomysticetidae (Mammalia: Cetacea) from the

Oligocene of New Zealand. PeerJ 3:ell29. DOI 10.7717/peerj.ll29

Cabrera, A. 1926. Cetaceos fosiles del Museo de La Plata. Revista

Museo de La Plata 29: 363-411.

Cozzuol, M.A. 1996. The record of the aquatic mammals in southern

South America. Pp. 321-342 in: Arratia, G. (ed.). Contributions of

Southern South America to Vertebrate Paleontology. Miinchner

Geowissenschaftliche Abhandlungen, Reihe A, Geologie und

Palaontologie, 30.

Cozzuol, M.A., and Humbert-Lan, G. 1989. On the systematic position

of the genus Prosqualodon Lydekker, 1893, and some comments

on the odontocete family Squalodontidae. Abstracts of Papers and

Posters, Fifth International Theriological Congress, Rome, 22-29

August 1989, 1; 483-484.

Eshelman, R.E., and Ward, L.M. 1994. Tribute to Frank Clifford

Whitmore, Jr. Proceedings of the San Diego Society of Natural

History 29: 3-10.

Fitzgerald, E.M.G. 2004. A review of the Tertiary fossil Cetacea

(Mammalia) localities in Australia. Memoirs of Museum Victoria

61: 183-208.

A late Oligocene waipatiid dolphin (Odontoceti: Waipatiidae) from Victoria, Australia

135

Fitzgerald, E.M.G. 2006. A bizarre new toothed mysticete (Cetacea)

from Australia and the early evolution of baleen whales.

Proceedings of the Royal Society B: Biological Sciences 273:

2955-2963.

Fitzgerald, E.M.G. 2010. The morphology and systematics of

Mammalodon colliveri (Cetacea: Mysticeti), a toothed mysticete

from the Oligocene of Australia. Zoological Journal of the

Linnean Society 158: 367-476.

Fitzgerald, E.M.G. 2012. Archaeocete-like jaws in a baleen whale.

Biology Letters 8: 94-96.

Flower, W.H. 1885. An Introduction to the Osteology of the Mammalia.

Third Edition. Macmillan and Co., London [reprinted by A. Asher

and Co., Amsterdam (1966)]. 382 pp.

Fordyce, R.E. 1982. A review of Australian fossil Cetacea. Memoirs of

the National Museum of Victoria 43: 43-58.

Fordyce, R.E. 1983. Rhabdosteid dolphins (Mammalia: Cetacea) from

the Middle Miocene, Lake Frome area. South Australia.

Alcheringa 7: 27-40.

Fordyce, R.E. 1984. Evolution and zoogeography of cetaceans in

Australia. Pp. 929-948 in: Archer, M., and Clayton, G. (eds).

Vertebrate Zoogeography and Evolution in Australasia. Hesperian

Press: Perth. 1203 pp.

Fordyce, R.E. 1991. A new look at the fossil vertebrate record of New

Zealand. Pp. 1191-1316 in: Vickers-Rich, P., Monaghan, J.M.,

Baird, R.F., and Rich, T.H. (eds). Vertebrate Palaeontology of

Australasia. Pioneer Design Studio in cooperation with the

Monash University Publications Committee: Melbourne. 1437.

Fordyce, R.E. 1994. Waipatia maerewhenua, new genus and new

species (Waipatiidae, new family), an archaic Late Oligocene

dolphin (Cetacea: Odontoceti: Platanistoidea) from New Zealand.

Proceedings of the San Diego Society of Natural History 29: 147-

176.

Fordyce, R.E. 2003. Cetacean evolution and Eocene-Oligocene

oceans revisited. Pp. 154-170 in: Prothero, D.R., Ivany, L.C., and

Nesbitt, E.A. (eds). From Greenhouse to Icehouse: The Marine

Eocene-Oligocene Transition. Columbia University Press: New

York. 541 pp.

Fordyce, R.E. 2004. The transition from Archaeoceti to Neoceti:

Oligocene archaeocetes in the southwest Pacific. Journal of

Vertebrate Paleontology 24 (Supplement to 3): 59A.

Fordyce, R.E. 2006. A southern perspective on cetacean evolution and

zoogeography. Pp. 755-778 in: Merrick, J.R., Archer, M., Hickey,

G.M., and Lee, M.S.Y. (eds). Evolution and Biogeography of

Australasian Vertebrates. Auscipub: Oatlands. 942 pp.

Fordyce, R.E., and Marx, F.G. 2016. Mysticetes baring their teeth: a

new fossil whale, Mammalodon hakataramea, from the southwest

Pacific. Memoirs of Museum Victoria 74: in press.

Fordyce, R.E., and Muizon, C. de. 2001. Evolutionary history of

cetaceans: a review. Pp. 169-233 in: Mazin, J.-M. and Buffrenil,

V. de (eds). Secondary Adaptation of Tetrapods to Life in Water.

Verlag Dr. Friedrich Pfeil: Milnchen. 367 pp.

Geisler, J.H., McGowen, M.R., Yang, G., and Gatesy, J. 2011. A

supermatrix analysis of genomic, morphological, and

paleontological data from crown Cetacea. BMC Evolutionary

Biology 11: 112.

Geisler, J.H., Colbert, M.W., and Carew, J.L. 2014. A new fossil

species supports an early origin for toothed whale echolocation.

Nature 508: 383-386.

Glaessner, M.F. 1955. Pelagic fossils {Aturia, penguins, whales) from

the Tertiary of South Australia. Records of the South Australian

Museum 11: 353-372.

Gradstein, F.M., Ogg, J.G., Schmitz, M., and Ogg, G. 2012. The

Geologic Time Scale 2012. Elsevier, Oxford. 1144 pp.

Hall, T.S. 1911. On the systematic position of the species of Squalodon

and Zeuglodon described from Australia and New Zealand.

Proceedings of the Royal Society of Victoria 23: 257-265.

Howell, A.B. 1927. Contribution to the anatomy of the Chinese Unless

porpoise, Neomeris phocaenoides. Proceedings of the United

States National Museum 70: 1-43.

Howell, A.B. 1930. Myology of the narwhal ( Monodon monoceros).

The American Journal of Anatomy 46: 187-215.

Kellogg, A.R. 1925. On the occurrence of remains of fossil porpoises

of the genus Eurhinodelphis in North America. Proceedings of

the United States National Museum 66: 1-40.

Kellogg, A.R. 1965. Fossil marine mammals from the Miocene

Calvert Formation of Maryland and Virginia: Part 1. A new

whalebone whale from the Miocene Calvert Formation. United

States National Museum Bulletin 247: 1-45.

Klima, M., Oelschlager, H.A., and Wiinsch, D. 1980. Morphology of

the pectoral girdle in the Amazon dolphin Inia geoffrensis with

special reference to the shoulder joint and the movement of the

flippers. Zeitschrift fur Saugetierkunde 45: 288-309.

Lambert, O., Bianucci, G., and Urbina, M. 2014. Huaridelphis

raimondii, a new early Miocene Squalodelphinidae (Cetacea,

Odontoceti) from the Chilcatay Formation, Peru. Journal of

Vertebrate Paleontology 34: 987-1004.

Lambert, O., Muizon, C. de, and Bianucci, G. 2015. A new archaic

homodont toothed cetacean (Mammalia, Cetacea, Odontoceti)

from the Early Miocene of Peru. Geodiversitas 37: 79-108.

Li, Q., Davies, P.J., and McGowran, B. 1999. Foraminiferal sequence

biostratigraphy of the Oligo-Miocene Janjukian strata from

Torquay, southeastern Australia. Australian Journal of Earth

Sciences 46: 261-273.

Marx, F.G., Tsai, C.-H., and Fordyce, R.E. 2015. A new Early

Oligocene toothed ‘baleen’ whale (Mysticeti: Aetiocetidae) from

western North America: one of the oldest and the smallest. Royal

Society Open Science 2: 150476. http://dx.doi.org/10.1098/

rsos.150476

McGowran, B., Holdgate, G.R., Li, Q., and Gallagher, S.J. 2004.

Cenozoic stratigraphic succession in southeastern Australia.

Australian Journal of Earth Sciences 51: 459-496.

Mchedlidze, G.A. 1970. Nekotorye Obshchie Cherty lstorii

Kitoobraznykh. Chast’ I. Akademia Nauk Gruzinskoi S.S.R.,

Institut Pale obi ologii, Metsniereba, Tbilisi. 112 pp. [In Russian.]

Mchedlidze, G.A. 1976. Osnoynye Cherty Paleobiologicheskoi lstorii

Kitoobraznykh. Akademia Nauk Gruzinoskoi S.S.R., Institut

Paleobiologii, Metsniereba, Tbilisi. 136 pp. [In Russian.]

Mchedlidze, G.A. 1984. General Features of the Paleobiological

Evolution of Cetacea. Amerind Publishing Co. Pvt. Ltd., New

Delhi. 139 pp. [English translation from Russian.]

McLaren, S., Wallace, M.W., Gallagher, S.J., Dickinson, J.A., and

McAllister, A. 2009. Age constraints on Oligocene sedimentation

in the Torquay Basin, southeastern Australia. Australian Journal

of Earth Sciences 56: 595-604.

Muizon, C. de. 1987. The affinities of Notocetus vanbenedeni, an

Early Miocene platanistoid (Cetacea, Mammalia) from Patagonia,

southern Argentina. American Museum Novitates 2904: 1-27.

Muizon, C. de. 1988. Le polyphyletisme des Acrodelphidae,

odontocetes longirostres du Miocene europeen. Bulletin du

Museum Nationale d’Histoire Naturelle (Paris) (4)1, Sect. C 10:

31-88.

Muizon, C. de. 1991. A new Ziphiidae (Cetacea) from the Early

Miocene of Washington State (USA) and phylogenetic analysis of

the major groups of odontocetes. Bulletin du Museum Nationale

d’Histoire Naturelle (Paris) (4)3-4, Sect. C 12: 279-326. [For

1990.]

136

E.M.G. Fitzgerald

Muizon, C. de. 1994. Are the squalodonts related to the platanistoids?

Proceedings of the San Diego Society of Natural History 29: 135-

146.

Murakami, M., Shimada, C., Hikida, Y., and Hirano, H. 2012. A new

basal porpoise, Pterophocaena nishinoi (Cetacea, Odontoceti,

Delphinoidea), from the upper Miocene of Japan and its

phylogenetic relationships. Journal of Vertebrate Paleontology

32: 1157-1171.

Nickel, R., Schummer, A., Seiferle, E., Frewein, J., Wilkens, H., and

Wille, K.-H. 1986. The Anatomy of the Domestic Animals.

Volume 1. The Locomotor System of the Domestic Mammals.

Verlag Paul Parey, Berlin. 499 pp.

Perrin, W.F. 1975. Variation of spotted and spinner porpoise (genus

Stenella ) in the eastern Pacific and Hawaii. Bulletin of the Scripps

Institution of Oceanography of the University of California 21:

1-206.

Pilleri, G. 1986. Beobachtungen an den Fossilen Cetaceen des

Kaukasus. Hirnanatomisches Institut, Ostermundigen,

Switzerland. 40 pp.

Pritchard, G.B. 1939. On the discovery of a fossil whale in the older

Tertiaries of Torquay, Victoria. The Victorian Naturalist 55: 151—

159.

Rich, T.H.V. 1975. Potential pre-Pleistocene fossil tetrapod sites in

New Zealand. Mauri Ora 3: 45-54.

Rich, T.H. 1976. Recent fossil discoveries in Victoria. Victorian

Naturalist 93: 198-206.

Rich, T.H. 1999. Australia: vertebrate paleontology. Pp. 140-149 in:

Singer, R. (ed). Encyclopedia of Paleontology. Volume 1: A-L.

Fitzroy Dearborn Publishers: Chicago. 687 pp.

Rich, T.H., and Rich, P.V. 1982. Search for fossils in New Zealand and

Australia. National Geographic Society Research Reports 14:

557-568.

Sanders, A.E., and Geisler, J.H. 2015. A new basal odontocete from

the upper Rupelian of South Carolina, U.S.A., with contributions

to the systematics of Xenorophus and Mirocetus (Mammalia,

Cetacea). Journal of Vertebrate Paleontology 35:1, e890107.

Schaller, O. 2007. Illustrated Veterinary Anatomical Nomenclature.

Second Edition. Enke Verlag, Stuttgart. 614 pp.

Schulte, H. von W., and Smith, M. de Forest. 1918. The external

characters, skeletal muscles, and peripheral nerves of Kogia

breviceps (Blainville). Bulletin of the American Museum of

Natural History 38: 7-72.

Siesser, W.G. 1979. Oligocene-Miocene calcareous nannofossils from

the Torquay Basin, Victoria, Australia. Alcheringa 3: 159-170.

Siryk, I.M., and Dubrovo, I.A. 1970. Iskopayemyy Zubatyy Kit V

Miotsenovykh Otlozheniyakh Yuzhnogo Sakhalina [Fossil

toothed whale from the Miocene deposits of the south Sakhalin

Island.]. Geologija i geofizika, Novosibirsk 1970 (9): 123-129. [In

Russian.]

Smith, G.J.D., Browne, K.W., and Gaskin, D.E. 1976. Functional

myology of the harbour porpoise, Phocoena phocoena (L.).

Canadian Journal of Zoology 54: 716-729.

Strickler, T.L. 1978. Myology of the shoulder of Pontoporia blainvillei,

including a review of the literature on shoulder morphology in the

Cetacea. American Journal of Anatomy 152: 419-431.

Tanaka, Y., and Fordyce, R.E. 2014. Fossil dolphin Otekaikea marplesi

(latest Oligocene, New Zealand) expands the morphological and

taxonomic diversity of Oligocene cetaceans. PLoS ONE 9(9):

el07972.

Tanaka, Y., and Fordyce, R.E. 2015a. A new Oligo-Miocene dolphin

from New Zealand: Otekaikea huata expands diversity of the

early Platanistoidea. Palaeontologia Electronica 18.2.23A: 1-71.

Tanaka, Y., and Fordyce, R.E. 2015b. Historically significant late

Oligocene dolphin Microcetus hectori Benham 1935: a new

species of Waipatia (Platanistoidea). Journal of the Royal Society

of New Zealand DOI: 10.1080/03036758.2015.1016046.

Tsai, C.-H., and Fordyce, R.E. 2015. The earliest gulp-feeding

mysticete (Cetacea: Mysticeti) from the Oligocene of New

Zealand. Journal of Mammalian Evolution 22: 535-560.

Uhen, M.D. 2004. Form, function, and anatomy of Dorudon atrox

(Mammalia, Cetacea): an archaeocete from the Middle to Late

Eocene of Egypt. University of Michigan Papers on Paleontology

34: 1-222.

Vickers-Rich, P., and Rich, T.H. 1993. Wildlife of Gondwana. Reed,

Chatswood. 276 pp.

Warne, M.T., Archbold, N.W., Bock, P.E., Darragh, T.A., Detmann,

M.E., Douglas, J.G., Gratsianova, R.T., Grover, M., Holloway,

D.J., Holmes, F.C., Irwin, R.P., Jell, P.A., Long, J.A., Mawson, R.,

Partridge, A.D., Pickett, J.W., Rich, T.H., Richardson, J.R.,

Simpson, A.J., Talent, J.A., and VandenBerg, A.H.M. 2003.

Palaeontology: the biogeohistory of Victoria. Pp. 605-652 in:

Birch, W.D. (ed). Geology of Victoria. Geological Society of

Australia Special Publication 23, Geological Society of Australia

(Victoria Division): Melbourne. 842 pp.

Memoirs of Museum Victoria 74:137-150 (2016) Published 2016

ISSN 1447-2546 (Print) 1447-2554 (On-line)

http://museumvictoria.com.au/about/books-and-journals/journals/memoirs-of-museum-victoria/

Earliest known record of a hypercarnivorous dasyurid (Marsupialia), from

newly discovered carbonates beyond the Riversleigh World Heritage Area,

north Queensland

MICHAEL Archer 1 ’* (http://zoobank.org/urn:lsid:zoobank.org:author:F834D7C6-E2E5-47B3-9209-AF1820AF766B),

OLIVIA Christmas 1 , Suzanne J. Hand 1 (http://zoobank.org/urn:lsid:zoobank.org:author:632F42CF-93FF-48E2-A0BA-D2FA65639D98),

KAREN H. Black 1 (http://zoobank.org/urn:lsid:zoobank.org:author: 3DCED980-309C-44EE-8A79-7C2B7766FF19),

Phil Creaser 1 , Henk Godthelp 1 , Ian Graham 1 , David Cohen 1 , Derrick A. Arena 1 ’ 2 , Caitlin Anderson 1 ,

Georgia Soares 1 , Naomi Machin 1 , Robin M. D. Beck 1 ’ 3 , Laura A. B. Wilson 1 , Troy J. Myers 1 ,

Anna K. Gillespie 1 , Bok Khoo 1 , and Kenny J. Travouillon 4 ’ 5

1 PANGEA Research Centre, School of Biological, Earth & Environmental Sciences, UNSW, Sydney, NSW 2052,

Australia

2 Associated Scientific Ltd, Australia

3 School of Environment & Life Sciences, Peel Building, University of Salford, Salford M5 4WT, UK

4 School of Earth Sciences, University of Queensland, St Lucia, Queensland 4072, Australia

5 Western Australian Museum, Locked Bag 49, Welshpool DC, WA 6986, Australia

* To whom correspondence should be addressed. E-mail: m.archer@unsw.edu.au

http://zoobank.Org/urn:lsid:zoobank.org:pub:61B65B3D-6DC0-4763-83A7-7EA10DB105E7

Abstract Archer, M., Christmas, O., Hand, S.J., Black, K.H., Creaser, P., Godthelp, H., Graham, I., Cohen, D., Arena, D.A.,

Anderson, C., Soares, G., Machin, N., Beck, R.M.D., Wilson, L.A.B., Myers, T.J., Gillespie, A.K., Khoo, B., and

Travouillon, K.J. 2016. Earliest known record of a hypercarnivorous dasyurid (Marsupialia), from newly discovered

carbonates beyond the Riversleigh World Heritage Area, north Queensland. Memoirs of Museum Victoria 74: 137-150.

Whollydooleya tomnpatrichorum gen. et sp. nov. is anew, highly specialised hypercarnivorous dasyuromorphian

from a new mid-Cenozoic limestone deposit southwest of the Riversleigh World Heritage Area in northwestern

Queensland. Dental dimensions suggest it may have weighed at least twice as much as the living Tasmanian devil

(Sarcophilus harrisii). Although known only from a lower molar, it exhibits a plethora of carnivorous adaptations

including a hypertrophied protoconid, tiny metaconid and a battery of vertical camassial blades between most of the

major cusps, most of which incorporate camassial notches to immobilise materials being sheared. It is unique among

dasyuromorphians in having a massive entoconid that closes the entire lingual side of the talonid. Comparison with

previously known thylacinid and dasyurid hypercamivores suggests its relationships are closer to dasyurids than

thylacinids in the main because of the very large entoconid, a cusp that is relatively small to absent in all known

thylacinids but commonly small to large in dasyurids. However, the extent of enlargement of the entoconid suggests

that it is not closely related to previously known Cenozoic hypercarnivorous dasyurids in the genera Dasyurus,

Glaucodon, Sarcophilus or any of the other previously described Cenozoic dasyurids. Although the early late Miocene

Ganbulanyi djadjinguli is only known from an upper molar, the reduced area of its protocone suggests a correspondingly

reduced rather than hypertrophied entoconid in its as-yet-unknown lower molars. Reconsideration of the structure of

the talonid in species of Sarcophilus even suggests that within that Quaternary lineage, the entoconid may have been

entirely lost, with the posteriorly displaced metaconid taking its functional place as an occlusal counterpart for the

blades of the protocone. The large size of the new species signals the earliest indication within the dasyurid radiation

of a late Cenozoic trend towards gigantism that became evident in many lineages of Australian marsupials. While the

age is uncertain, on the basis of associated taxa such as species of Ekaltadeta, it is probably either mid or late Miocene

in age. Geological features of the deposit suggest it was formed within a pool in a cave environment that periodically

underwent desiccation. Some grains suggest an aeolian as well as an alluvial and pluvial origin for the deposit. This

may relate to current understanding about environmental change that took place in the region following the mid

Miocene climate oscillation.

Keywords Dasyuridae; Dasyurinae; Thylacinidae; Riversleigh; Miocene; Whollydooleya; Dasyurus; Sarcophilus; Glaucodon;

Ganbulanyi; Thylacinus

138

M. Archer, 0. Christmas, S.J. Hand, K.H. Black, P. Creaser, H. Godthelp, I. Graham, D. Cohen, D.A. Arena,

C. Anderson, G. Soares, N. Machin, R.M.D. Beck, L.A.B. Wilson, T.J. Myers, A.K. Gillespie, B. Khoo & K.J. Travouillon

Introduction

In 2012, M. Archer, P. Creaser and H. Godthelp, with the

assistance of the Queensland Parks and Wildlife Service, had the

opportunity to explore a Cenozoic terrain approximately 1.2 km

southeast of the southwestern border of the Riversleigh World

Heritage Area but within the boundaries of Boodjamulla (Lawn

Hill) National Park in northwestern Queensland. The purpose

was to test an unpublished hypothesis developed by geologist

Ned Stephenson. He had concluded (and speculatively mapped)

on the basis of satellite data that Cenozoic freshwater limestone

deposits similar to those contained within the World Heritage

Area might occur within an equally vast region southwest of the

known fossil deposits. A similar inference about the fossiliferous

potential of the rocks in this same general area had previously

been suggested by Rick Arena in 2009 on the basis of Google

Earth imagery. One of the first ground-based discoveries that

supported this hypothesis, made by P. Creaser, was a fossil

deposit named Wholly Dooley Site on an isolated limestone

knoll named Wholly Dooley Hill. Small samples of Wholly

Dooley matrix were treated with acetic acid in the UNSW

laboratory to obtain samples of the vertebrate fauna.

Among the first teeth recovered was a highly distinctive

lower molar of a carnivorous marsupial, which is the subject of

this paper. In 2013, the National Geographic Society provided

research funding to M. Archer and colleagues to continue

exploration in this remote region of northwestern Queensland.

Additional new sites and faunal assemblages were discovered

in 2013 and 2014 in the new area, now named New Riversleigh,

but as yet none have produced additional specimens of this

new carnivore.

Although several fossil Oligo-Miocene putative dasyurids

or dasyurid-like taxa have already been described from the

Riversleigh region (i.e., species of Ganbulanyi, Barinya,

Mayigriphus, Malleodectes and Joculusiunr, Wroe, 1997a,

1998, 1999, 2001; Arena et al., 2011), these are significantly

different from Whollydooleya tomnpatrichorum gen. et sp. nov.

in terms of key structural features (see below in Comparisons).

Although the only known specimen of Ganbulanyi djadjinguli

Wroe, 1998 is a fractured, isolated upper molar with some

features suggesting it was a hypercarnivore, in other features it

too is clearly distinct as well as significantly smaller than the

taxon described herein. Other distinctive Australian carnivorous

marsupials previously described as late Oligocene dasyurids

(e.g., Ankotarinja, Keeuna, Wakamatha and Dasyulurinja

[Archer, 1976a, 1982; Archer and Rich, 1979] relegated by Wroe

[1996, 1997b] to Dasyuromorphia incertae sedis and Godthelp

et al. [1999] to Marsupialia incertae sedis), are tiny to small

insectivores all of which lack the hypercarnivorous

specialisations evident in W. tomnpatrichorum.

Abbreviations used in this paper include the following:

QM F, Queensland Museum palaeontological collections;

NMV P, Museum Victoria Palaeontology Collection; prd,

protoconid; med, metaconid; hyd, hypoconid; end, entoconid;

hyld, hypoconulid; co, cristid obliqua; mcd, metacristid; STB,

stylar cusp B; STD, stylar cusp D. Molar morphology follows

that used by Archer (1976b) or is self-explanatory or in

common use. Thegotic terminology (e.g., alpha-scissorial)

follows that used by Every (1970). Molar serial homology

follows that used by Thomas (1888).

Systematics

Dasyuridae Goldfuss, 1820

?Dasyurinae Goldfuss, 1820

Whollydooleya Archer et al., gen. nov.

Generic diagnosis. Species of Whollydooleya differ from all

other dasyuromorphians in having a massive (rather than

blade-like, conical or reduced) entoconid that completely

closes the lingual flank of the talonid.

Type species. Whollydooleya tomnpatrichorum Archer et al.,

2015, sp. nov., by monotypy.

Whollydooleya tomnpatrichorum Archer et al., 2015, sp. nov.

Zoobank LSID. http://z 00 bank. 0 rg/urn:lsid:z 00 bank. 0 rg:act:

9BAE184C-8E93-4E74-B10C-DB66EF882E82

Specific diagnosis. That of the genus until additional species

are known.

Holotype. QM F57892, partial right lower molar, interpreted to

be either M 2 or M 3 .

Type locality. Wholly Dooley Site, Wholly Dooley Hill, the New

Riversleigh area, southwest and adjacent to the Riversleigh World

Heritage Area, northwestern Queensland. Wholly Dooley Site is

one of several fossiliferous localities discovered by P. Creaser et

al. on Wholly Dooley Hill in 2012. GPS coordinates for this site

have been recorded with the Queensland Museum, Brisbane.

Etymology. The generic name refers to Wholly Dooley Site,

which was discovered and named in 2012 by P. Creaser

following preliminary analyses of satellite data by Ned

Stephenson and Google Earth imagery by Rick Arena. The

generic name is hereby given masculine gender. The species

name honours Tom and Pat Rich for their years of research that

included joint work on the mid-Cenozoic deposits of Riversleigh.

Geological context

The Wholly Dooley Site deposit (Fig. 1) shares a number of

characteristics with other deposits at Riversleigh that indicate it

represents an accumulation formed within a cave whose walls

and ceiling have subsequently eroded away (Arena et al., 2014).

The host micrite is dominated by calcite, with moderately

common broken mollusc shell fragments and detrital quartz

grains (Fig. 2A) and uncommon calcite pelloids and calcite rafts

(Fig. 2B). The occurrence of calcite cave rafts is indicative of a

quiescent pool of carbonate-rich water and suggests that these

formed within a karst environment, possibly a small cave pool.

There is a clearly defined erosional contact between the host

micrite and an overlying layer (layer 2; Fig. 2C). This layer

comprises Fe-oxide (various mixtures of hematite and goethite),

pelloids (Fig. 2D), detrital quartz grains and abundant late

calcite-fill. Quartz in this later fill occurs as equant to prismatic,

New hypercarnivorous marsupial from the Riversleigh World Heritage Area

139

Figure 1. Wholly Dooley Site, Wholly Dooley Hill, northwestern Queensland. The grey limestone massifs to the left are unfossiliferous Cambrian

limestone and probably represent part of the western and southern walls of the original cave chamber. Fossiliferous Wholly Dooley matrix has

been obtained from the excavated depression in the centre. About 7.2 km NE from this point is the approximate centre of the Riversleigh World

Heritage Area; about 1.2 km NWN from this point is the southwestern boundary of the RWHA. Photo M. Archer.

subangular to well-rounded grains. Some of the quartz grains are

highly fractured and coated in hematite/goethite, suggesting an

aeolian origin. The abundance of goethite within this layer as

fine colloidal aggregates and growths suggests a change in the

environment, with drying-out of the cave system and evaporation

of oxygen-rich cave waters, leading to in situ Fe-oxide

precipitation. This drying-out of the environment is also indicated

by the aeolian-derived detrital quartz grains. However, the lack

of conspicuous desiccation cracks suggests that the sediment

remained damp as it accumulated within the cave environment.

Layer 2 is unconformably overlain (erosional contact) by

layer 3. Layer 3 comprises relatively abundant detrital quartz

grains, and distinctly rhythmically banded goethite/calcite

(Fig. 3A), with calcite also occurring as a void-fill phase (Fig.

3B). The layer also contains rare well-rounded subequant

detrital tourmaline grains and cryptocrystalline silica as

overgrowths around detrital quartz grains (Fig. 3C). Also

within this layer is possible microbial-mediated goethite (Fig.

3D). The occurrence of rhythmically banded goethite/calcite

suggests constantly changing water chemistry, from carbonate-

rich waters to oxygen-rich waters, throughout crystallisation

of this layer. The microbial-mediated goethite suggests that

the system was exposed to at least periodic sunlight, and may

indicate unroofing of the cave system.

The proposed paragenesis of Wholly Dooley Site is detrital

quartz and tourmaline and fossil bone fragments, followed by

calcite (CaC0 3 ), then rhythmic deposition of goethite (FeO(OH))

and calcite (CaC0 3 ). However the overall composition is highly

variable, ranging from 10 to 30 modal% quartz, 20 to 40

modal% colloidal goethite and 30 to 70 modal% calcite cement.

The size and shape of quartz grains are highly variable, ranging

from subangular to well-rounded and equant to subprismatic in

shape and the quartz grains are poorly sorted. There is also

evidence of well-developed undulose extinction within many of

140

M. Archer, 0. Christmas, S.J. Hand, K.H. Black, P. Creaser, H. Godthelp, I. Graham, D. Cohen, D.A. Arena,

C. Anderson, G. Soares, N. Machin, R.M.D. Beck, L.A.B. Wilson, T.J. Myers, A.K. Gillespie, B. Khoo & K.J. Travouillon

Figure 2. Photomicrographs of geological thin sections from Wholly Dooley Site. A, Detrital quartz grains within the host micrite UXP; B,

calcite raft fragment within host micrite UXP; C, erosional contact between host micrite (left) and overlying layer 1 (right) showing abundant

aeolian-derived quartz within layer 1 UXP; D, Fe-oxide pelloids and detrital quartz grains within layer 1 PPL. WD2 numbers refer to thin section

images produced in the School of Biological, Earth & Environmental Sciences.

the quartz grains, along with uncommon recrystallised quartz

grains. These features suggest that the detrital quartz grains are

derived from at least three sources, including: 1, a proximal

source via eluvial deposition (angular prismatic grains); 2, a

more distal source via aeolian deposition (hematite-coated

highly fractured equant-shaped well-rounded grains); and 3, a

more distal source (subrounded to well-rounded equant to

subprismatic grains) introduced via alluvial processes.

On balance, these sedimentological attributes suggest this

deposit accumulated under somewhat different circumstances

than those involved in accumulation of the previously known

Oligo-Miocene fossil deposits from the Riversleigh World

Heritage Area. The suggestion of intermittent desiccation may

indicate accumulation during a younger, climatically drier

period than those involved in triggering Depositional Phases

1-3 in the Riversleigh World Heritage Area (Creaser, 1997;

Arena, 2004; Woodhead et al., 2014).

Description of the holotype

The holotype (Figs 4A-C, 5, 6, 9A-A’), which is comparable in

size to the M 3 of the extant Tasmanian devil (Sarcophilus

harrisii) 1 , retains the entire talonid and broken posterior half

of the trigonid. Partially repaired damage to the protoconid,

from accidental breakage during preparation, has resulted in

retention of a small displaced fracture that extends across the

posterior face of the trigonid from the metaconid to the base of

the protoconid. Structures missing from this tooth because of

natural, taphonomic loss include the protocristid, paraconid

and anterior cingulid.

1 There is uncertainty in the current literature about the nomenclatural

relationship of laniarius Owen, 1838 and harrisii Boitard, 1841. Here

we follow Dawson (1982) in using harrisii as the appropriate name for

the living species on the interpretation that the larger extinct and

smaller living Tasmanian Devils are neither clearly synonymous nor

chronospecies.

New hypercarnivorous marsupial from the Riversleigh World Heritage Area

141

Figure 3. Photomicrographs of geological thin sections from Wholly Dooley Site later infills. A, Rhythmically banded calcite and goethite UXP; B,

calcite filling void in layer 2 UXP; C, late cryptocrystalline silica (indicated by red arrow) filling fractures UXP; D, possible microbially mediated

goethite precipitation (dendrites) in layer 2 PPL. WD2 numbers represent images of thin sections produced in the University of New South Wales.

The trigonid has a massive, well-developed protoconid and

very small metaconid on the posterolingual flank of the

protoconid. There is a rudimentary but distinct carnassial

notch between the metaconid and metacristid at the point

where the metaconid diverges from the posterolingual flank of

the trigonid. There is a nearly vertical postmetacristid that

descends from the metaconid to form the anterior half of a

carnassial notch in a small, very steeply inclined lingual blade

composed of the preentocristid and postmetacristid, as occurs

in some dasyurids (e.g., species of Dasyurus).

The extremely steeply inclined metacristid, which is close

to vertical in orientation, has sustained wear resulting from

mastication that has breached the enamel on the lingual side of

the trigonid with consequent production of a very sharp cutting

edge along the shearing edge of this blade. The enamel

exposed at the margins of this blade, and elsewhere over the

entire crown, is very thin.

The posterior flank of the metacristid has been secondarily

(during life) planed off by the preparacrista and preprotocrista

of the corresponding upper molar producing a thegotic facette

that covers most of the posterior flank of the trigonid trailing

the dorsal leading edge of the metacristid blade, as occurs in

all other dasyuromorphians. The wear striae on this facette

are uniformly shallow, unidirectional and extend the length of

the facette, features that are normal attributes of thegotic

facettes (Every, 1970). This extensive facette also extends

onto the posterobuccal apex of the metaconid after bridging

rather than faceting the notch between the metaconid and

metacristid. The tip of the protoconid, insofar as it is

preserved, exhibits a very small apical wear facette, suggesting

that this animal was relatively young at the time of death and

that the extremely well-developed facette on the posterior

flank of the metacristid is in fact primarily the result of tooth/

tooth thegosis rather than mastication involving food.

Nevertheless it is possible that the thegotic striations have

142

M. Archer, 0. Christmas, S.J. Hand, K.H. Black, P. Creaser, H. Godthelp, I. Graham, D. Cohen, D.A. Arena,

C. Anderson, G. Soares, N. Machin, R.M.D. Beck, L.A.B. Wilson, T.J. Myers, A.K. Gillespie, B. Khoo & K.J. Travouillon

Figure 4. Comparison of the holotype of Whollydooleya tomnpatrichorum n. gen. et sp. QM F57892 with other hypercarnivorous dasyuromorphians

(not to scale). A-C, W. tomnpatrichorum , QM F57892, m2 or m3. A, posterobuccal view; B-B’, occlusal stereoview; C, oblique posterobuccal

view. D-F, Dasyurus maculatus, AR21693, M r D, buccal view; E-E’, occlusal stereoview; F, oblique posterobuccal view. G-I, Sarcophilus

harrisii , AR21694, M r G, buccal oblique view; H-H’. occlusal stereoview; I, oblique posterobuccal view. J-L, Thylacinus cynocephalus,

AR21695, M 3 talonid (reversed from the LM 3 ). J, posterobuccal oblique view; K-K\ occlusal stereoview; L, oblique posterobuccal view.

New hypercarnivorous marsupial from the Riversleigh World Heritage Area

143

Figure 5. Stereoviews of the holotype of Whollydooleya tomnpatrichorum n. gen. et sp. QM F57892 from a buccal oblique view. The margin at

the right represents the missing flank of the trigonid.

been overprinted on what may have been more irregular

scorings resulting from mastication.

There is no buccal cingulum evident at the base of the

protoconid, which is also characteristic of most

dasyuromorphians (but not universal; e.g. species of

Dasyuroides can exhibit better-developed buccal cingulids;

Archer 1976b). However, the trigonid is broken, so structures

that may have been present anterior to the widest point of the

trigonid are unknown.

The talonid is wider than the trigonid and supports three

cusps. The hypoconid is robust with a well-developed cristid

obliqua and posthypocristid. The cristid obliqua appears to

have extended towards the midpoint of the posterior flank of

the trigonid, meeting the latter about midway up the height of

the trigonid, although it is worn from the lowest point along

this blade to the position of its attachment on the trigonid.

There is a kink in the cristid obliqua at the low point, suggesting

that there may have been a carnassial notch between the

posterior descending and anterior ascending sections of the

cristid obliqua, as occurs in most dasyuromorphians. On the

anterior side of the junction between these two parts of the

cristid obliqua, at the low point on the buccal side of the

interface between the trigonid and talonid, there also appears

to be a remnant of a small neomorphic cuspid. Wear of the

cristid obliqua has produced a small carnassial notch that

operates obliquely between this cuspid and the prehypocristid.

On the posterior flank of the hypoconid, subtending the

posthypocristid blade, there is a distinctive thegotic facette that

was produced by the premetacrista of the corresponding upper

molar that parallels the far larger facette produced by the

preparacrista of the upper molar on the posterior flank of the

trigonid, hence both are the result of the same movement of the

mandible. This facette similarly reveals fine, unidirectional

oblique striations cut into the enamel. This posthypocristid

facette extends ventrally across a small gap to include the

buccal half of the ascending posterior cingulid, resulting in

breached enamel and a dentine trough within the postcingulid

(= the posterior cingulid). This part of the facette developed on

the postcingulid was almost certainly caused by thegosis

involving the apex of the metacone of the corresponding upper

molar. There is also a wear facette on the anterior flank of the

hypoconid subtending the prehypocristid that was produced by

the postparacrista of the corresponding upper molar, a facette

also commonly seen in dasyuromorphians, even in species of

Sarcophilus that have undergone hypotrophy of the talonid.

The posthypocristid leading edge is notched along its

length suggesting that it has been used to segment reasonably

hard materials. While these notches could be the result of

postmortem damage, given their restriction to the leading

edge of this blade, it is more likely that they reflect use during

the life of the animal. The posthypocristid terminates lingually

before it includes or contacts the hypoconulid.

The hypoconulid is developed as a relatively (compared

with the hypoconid and entoconid) small talonid cusp in the

form of a short, imprecise, buccolingual vertical blade that is

orientated in line with the posthypocristid and in effect extends

that blade across a low gap between the two blades emanating

from these cusps. Holistically, the hypoconulidcristid and the

144

M. Archer, 0. Christmas, S.J. Hand, K.H. Black, P. Creaser, H. Godthelp, I. Graham, D. Cohen, D.A. Arena,

C. Anderson, G. Soares, N. Machin, R.M.D. Beck, L.A.B. Wilson, T.J. Myers, A.K. Gillespie, B. Khoo & K.J. Travouillon

posthypocristid form a combined transverse blade with a

carnassial notch in effect developed by the low gap between the

two blade segments. The buccal end of the hypoconulidcristid

terminates at the junction of the lingual end of the posterior

cingulid and the lingual end of the posthypocristid. Unlike the

entoconid and hypoconid, there is no distinct high point on the

hypoconulid; rather it is a small transverse ridge. It does not

extend sufficiently posteriorly to have acted as an effective

interlocking device inserted into the anterior cingulid of the

succeeding molar. On its posterior flank there is a coarse wear

facette indicating interdental abrasion with the anterior edge of

the succeeding molar. The overall relationship between the

hypoconid and hypoconulid is similar to that seen in some of

the largest dasyurine dasyurids (species of Glaucodon and

Sarcophilus ) but differs from the condition seen in other

Figure 6. Measurements of the holotype of Whollydooleya

tomnpatrichorum n. gen. et sp. QM F57892 in occlusal stereoview. In

the upper figure, the sides of the triangle link: 1, protoconid to

entoconid; 2, protoconid to hypoconid; and 3, entoconid to

hypoconid. The longitudinal measure is the distance between the

protoconid and the apex of the hypoconulid. In the lower figure, the

measures are: talonid width through the entoconid and hypoconid;

talonid length from the ascending extremity of the cristid obliqua

against the posterior flank of the trigonid to the posterior extremity

of the crown through the hypoconulid; and maximum width of the

damaged trigonid.

dasyurids and in thylacinids where the hypoconulid of the M 2

is as large as or larger than the entoconid and projects posteriad

to interlock within the anterior cingulid of M 3 .

The entoconid has a short preentocristid developed in

relation to the postmetacristid, as noted above. Although the

posterior as well as anterior flanks of the entoconid are

relatively massive, there is no distinct postentocristid as such

(although there is a poorly defined ridge descending its

posterobuccal flank that is probably a vestigial homologue of

this otherwise missing blade); as a result, there was a small

open trough between the bases of the entoconid and hypoconulid

such as occurs in most dasyuromorphians that have not reduced

their talonids. Hence, there was no shearing activity provided

by this specific region of the talonid.

The upper part of the buccal face of the entoconid exhibits

a wear facette produced by the lingual face of the protocone of

the corresponding upper molar. The intra-talonid bases of all

talonid cusps facing each other across the talonid basin as well

as the central part of the talonid basin itself, appear to be

missing enamel. This may have been the result of attrition

produced by occlusion with the corresponding protocone of the

upper molar, but this seems improbable given the very limited

masticatory wear on all cusps of this tooth. It is perhaps more

likely that the enamel is missing because of chemical erosion

resulting from some taphonomic process such as root-acid

dissolution. Hence when cleaning the tooth, we stopped when

the surface appeared to represent dentine.

The postcingulid descends from a distinct starting point

about three-quarters of the way from the buccal side of the

tooth, down around the base of the hypoconid and anteriorly as

far as the posterobuccal base of the protoconid. On the buccal

side, this cingulid is disrupted by small cuspids along its length.

Seemingly similar cuspids, unattached to the basal cingulid,

occur on the lowest anterobuccal flank of the prehypocristid

portion of the cristid obliqua.

Comparisons

Based on overall morphology, there can be little doubt that W.

tomnpatrichorum represents a dasyuromorphian. While

definitive familial identification is uncertain given the lack of

basicranial information (Wroe, 1999), the highly derived molar

morphology including the plethora of vertical blades with

carnassial notches (e.g., metacristid and preentocristid),

hypotrophied metaconid and hypertrophied protoconid, strongly

suggests that it represents this group. This overall conclusion is

also supported by the argument of Voss and Jansa (2009) that

while loss of the posterior cingulid (which is conspicuously

present in W. tomnpatrichorum ) is a synapomorphy of

Marsupialia, secondary presence of this structure is a

synapomorphy of dasyuromorphians.

Of the many different kinds of dasyuromorphians already

known (e.g., Archer, 1982; Wroe, 2003), in its very reduced

metaconid and hypertrophied protoconid it shares most derived

features with dasyurine dasyurids and, to a lesser extent,

thylacinids. Comparisons here have been made specifically with

all of the largest and most specialised of the dasyurine dasyurids

including the living Tiger Quoll {Dasyurus maculatus; Figs

New hypercarnivorous marsupial from the Riversleigh World Heritage Area

145

Figure 7. NMV P207018, cast of a right dentary referred to Glaucodon ballaratensis by Gerdtz and Archbold (2003) that preserves the RC 1? P | 2

and M 14 . Left upper, buccal view. Left lower, lingual view. Right, occlusal view of the dentition. Scale bar = 1 cm. Photographs by

Darren Bellingham.

4D-F, 9B-B’), the early late Miocene Gadbulanyi djadjinguli

(Wroe, 1998), the Pliocene Glaucodon ballaratensis (Figs 7,

9C-C’), the living Tasmanian Devil (S. harrisii; Figs 4G-I,

9E-E’), the early Pleistocene devil ( Sarcophilus moornaensis ;

Figs 8, 9D-D’), and the Thylacine ( Thylacinus cynocephalus;

Figs 4J-L, 9F-F’). Juxtaposed M 3 s of these taxa are presented in

Figure 9. No lower molar is known for G. djadjinguli. Other mid

to late Cenozoic dasyurid-like taxa known on the basis of molars

from Riversleigh include species of Barinya, Mayigriphus and

Joculusium but none of these have hypercarnivorous

specialisations of the kind exhibited by W. tomnpatrichorum.

Comparisons here primarily involve M 2 and/or M 3 (Fig. 9).

While position homology of the holotype of W. tomnpatrichorum

is uncertain, the large talonid of the Wholly Dooley specimen

means that it is highly likely that it is not M 4 given that the talonid

of the posterior molar in all dasyuromorphians is much narrower

than the trigonid. While it could be an M p this seems improbable

because the hypoconid is very much lower in height than the

protoconid. In the M, of larger dasyuromorphians, particularly

the hypercarnivores, the hypoconid on M l is commonly almost

half to three-quarters the height of the protoconid, while in the

M 3 it is rarely more than a quarter the height of the protoconid—

whichis the condition seeninthe holotype of W. tomnpatrichorum.

A buccal cingulid surrounding the base of the talonid,

which is well-developed in W. tomnpatrichorum and all of the

larger dasyurines including species of Sarcophilus (despite the

latter having significantly hypotrophied talonids), is not present

in the modern Thylacine (T. cynocephalus ). It is present,

however, in some Miocene thylacinids (e.g., Badjcinus

turnbulli ) and hence its absence in T. cynocephalus is almost

certainly an autapomorphy.

The most striking feature of the talonid of W.

tomnpatrichorum is the very large entoconid with well-

developed, longitudinal pre-entocristids and poorly-developed

postentocristid. These blades virtually enclose the whole of

the lingual side of the talonid. This is in distinct contrast to all

146

M. Archer, 0. Christmas, S.J. Hand, K.H. Black, P. Creaser, H. Godthelp, I. Graham, D. Cohen, D.A. Arena,

C. Anderson, G. Soares, N. Machin, R.M.D. Beck, L.A.B. Wilson, T.J. Myers, A.K. Gillespie, B. Khoo & K.J. Travouillon

Figure 8. NMV P28684, right dentary of the holotype of Sarcophilus moornaensis preserving damaged M 14 . A, Buccal view; B, buccal oblique

view; C, lingual view; D, occlusal view of M 1 4 . Photographs by Erich Fitzgerald.

known thylacinids, which have relatively small to almost

absent entoconids that are also conical in shape without

hypertrophy of any subtended longitudinal blades.

Considering entoconid development in dasyurines, none

exhibit hypertrophy of this cusp remotely to the extent shown

in W. tomnpatrichorum, although less specialised dasyurines

such as species of Dasyurus have large (albeit much smaller

than those in W. tomnpatrichorum ) entoconids with distinct

pre- andpostentocristids. InD. maculatus (Figs 4D-F, 9B-B’),

there is also a very well-developed carnassial notch, as in W.

tomnpatrichorum, developed between the preentocristid and

postmetacristid. In D. maculatus, however, the small, low

postentocristid barely contacts the hypoconulid, which results

in a failure of the entoconid and associated blades to effectively

enclose the lingual side of the talonid; this contrasts with the

condition in W. tomnpatrichorum, where the entire lingual

side of the talonid is enclosed by these structures.

The largest of the dasyurines, the species of Sarcophilus

and Glaucodon, have evolved in a different direction with

extreme reduction of the entoconid and hypotrophy of the

talonid as a whole. In the case of S. harrisii (Figs 4G-I, 9E-E’),

this correlates with significant reduction of the protocone and

other occlusal upper molar counterparts for talonid structures.

In G. ballaratensis (structure of the M 3 in this species being

based on the specimen described by Gerdtz and Archbold,

2003; Figs 7, 9C-C’), the entoconid is clearly present on the

foreshortened talonid, but very small. In S. moornaensis (Figs

8, 9D-D’), it is even smaller to minuscule, although the talonid

is less foreshortened than it is in G. ballaratensis. In S. harrisii

something quite different has occurred (Figs 4G-I, 9E-E’): in

conjunction with the far greater hypertrophy of the protoconid

and coordinate posterior displacement of the metaconid, the

talonid retains only two cusps and has become an ‘appendage’

at the base of the posterobuccal flank of the metacristid. The

buccal-most of these cusps is the hypoconid but the homology

New hypercarnivorous marsupial from the Riversleigh World Heritage Area

147

Figure 9. Comparison of the distribution of major cusps (in occlusal view) preserved in the Whollydooleya tomnpatrichorum holotype (QM

F57892) with the RM 3 of representative extinct (f) and recent hypercarnivorous dasyuromorphians. A-A\ Whollydooleya tomnpatrichorum (f);

B-B', Dasyurus maculatus , AR21693; C-C’, Glaucodon bailoratensis, P207018 (f); D-D’, Sarcophilus moornaensis, NMV P28684 (f); E-E’,

Sarcophilus harrisii, AR21694; F-F’, Thylacinus cynocephalus, AR21695 (image reversed from the LM 3 ) (f). Abbreviations: end, entoconid

(orange dot); hyd, hypoconid (green dot); hyld, hypoconulid (purple dot); med, metaconid (blue dot); prd, protoconid (red dot). Not to scale.

of the other cusp is unclear. If the minuscule size of the

entoconid in S. moornaensis presages changes leading to the

condition seen in S. harrisii, then it would seem possible that

the entoconid has been lost in S. harrisii and the cusp that

remains is the hypoconulid. If this is the case, the most

interesting consequence is that the posteriorly displaced

metaconid may have replaced the function of the entoconid as

the occlusal counterpart of the protocone, given that it now

occurs directly lingual to the hypoconid and in the topographic

position of the entoconid of other dasyurids. An arguably less

plausible alternative, given the medial posterior position of the

more lingual of the two talonid cusps in S. harrisii, would be

that this is the entoconid and that the hypoconulid has been

lost. Without annectant taxa bridging the transition between

Sarcophilus moornaensis and S. harrisii, or perhaps a detailed

analysis of occlusal relationships, the evolutionary fate of the

entoconid in these large hypercarnivorous dasyurines is

unclear. It is interesting to note that a similar argument has

been made (Forasiepi et al., 2014) to explain what may have

occurred on the other side of the world in relation to evolutionary

reduction of the talonid in borhyaenids (Marsupialia,

Sparassodonta) leaving uncertainty about the homology of the

vestigial cusps in forms that retain talonid structures. There is

even some uncertainty about the fate of the metaconid in more

derived hypercarnivorous sparassodontans.

Although the modern Thylacine lacks a metaconid,

presence in this Riversleigh taxon of a distinct (albeit reduced)

metaconid does not exclude a more distant relationship to

thylacinids given that that cusp is present in almost all of the

Oligo-Miocene thylacinids. Reduction to loss of the metaconid

is likely to be a convergent hypercarnivorous feature relating

to hypertrophy of the protoconid and metacristid and

longitudinal orientation and hypertrophy of the paracristid as

the primary alpha-scissorial carnassial blade (Every, 1970) in

derived thylacinids and dasyurids.

The hypoconulid in the M 13 of dasyurids and thylacinids

projects further posteriad than any other structure on the

talonid and locks into the hypoconulid notch of the anterior

cingulid of the following molar, thereby restricting differential

transverse movement between adjacent teeth. Given the

primarily alpha-scissorial function of the molars in these

insectivorous/carnivorous marsupials, stability of posture for

these molars during vertical shearing is aided by this

interlocking system. For this reason, it is curious that the

hypoconulid in W. tomnpatrichorum extends only just beyond

the posterior flank of the talonid. This is also the case, however,

148

M. Archer, 0. Christmas, S.J. Hand, K.H. Black, P. Creaser, H. Godthelp, I. Graham, D. Cohen, D.A. Arena,

C. Anderson, G. Soares, N. Machin, R.M.D. Beck, L.A.B. Wilson, T.J. Myers, A.K. Gillespie, B. Khoo & K.J. Travouillon

in species of Glaucodon (very small hypoconulid notch) and

Sarcopliilus (very small in M 2 , absent in posterior molars), and

may be a correlate of the larger, wider, potentially more

laterally stable molar structure in species of these genera. It is

uniquely developed in W. tomnpatrichorum more as an oblique

transverse ridge (confluent with the posthypocristid, somewhat

after the fashion seen in some peramelemorphians) than a

discrete, posteriorly directed cuspid.

Of all the taxa noted above by way of comparison, the one

that is the most enigmatic is the possibly early late Miocene

Ganbulanyi djadjinguli from Encore Site at Riversleigh. The

holotype, an isolated, incomplete, worn upper molar, was

originally described by Wroe (1998) as probably conspecific

with an isolated massive ovate premolar from the same deposit.

Arena et al. (2011), in the course of describing Malleodectes

mirabilis, suggested that the Encore molar and premolar

represented different taxa and named the premolar Malleodectes

moenia. They argued that both species of Malleodectes may

have used their massive premolars to crush snail shells. Wroe’s

(1988) description of the holotype of G. djadjinguli suggests

that this animal, based on the upper molar, was a hypercarnivore

and hence it could possibly be related to W. tomnpatrichorum.

Direct comparisons between the two teeth are, however, difficult

because they are very different in size, exhibit markedly

different degrees of wear and represent different tooth positions,

the holotype of W. tomnpatrichorum being a lower molar while

that of G. djadjinguli is an upper molar. Nevertheless, one of the

key distinctive features of D. djadjinguli noted by Wroe (1998)

is the relatively small occlusal area represented by the protocone.

The very wide talonid and hypertrophied entoconid of W.

tomnpatrichorum suggests an upper molar morphology that

would have been significantly different with an

uncharacteristically (for a hypercarnivorous dasyuromorphian)

large protocone and associated blades. While tooth size alone

clearly indicates that these taxa are not conspecific (the lower

molar of W. tomnpatrichorum being wider than the upper molar

of G. djadjinguli ), the differences in protocone/talonid

morphologies also suggest that the two taxa represent two quite

distinct lineages of hypercarnivorous dasyurids. The nature of

wear on the two specimens also suggests a significantly different

life-style. In the case of the upper molar of G. djadjinguli, the

wear is heavily apical resulting in conjoined paracone/STB and

metacone/STD, wear patterns of a kind commonly seen in the

molars of durophagous Tasmanian devils. In the lower molar of

W. tomnpatrichorum, wear is clearly evident along the leading

blade of the metacristid but there is very little apical wear

evident on the protoconid and almost none on the hypoconid or

entoconid, suggesting that this larger animal was nevertheless

not a durophagous hypercarnivore.

Discussion

Whollydooleya tomnpatrichorum exhibits classic features of a

marsupial hypercarnivore in being large (evidently larger than a

modern Tasmanian Devil), in having a robust molar morphology

with a hypertrophied protoconid, highly reduced metaconid and

a plethora of vertical shearing blades with carnassial notches on

both the trigonid and talonid. The uniformly fine, unidirectional

striations on the posterior flank of the metacristid and

posthypocristid indicate thegotic maintenance of the cutting

edge of those blades. These features are shared by the larger

dasyuromorphians such as species of Dasyurus, Glaucodon,

Sarcopliilus and some of the more plesiomorphic members of

the Thylacinidae. This new species differs, however, from all

thylacinids in having a very large entoconid and from the larger,

later thylacinids in retaining a distinct metaconid. The massive

entoconid distinguishes it additionally from all other dasyurids,

including the largest dasyurine dasyurids to which it is otherwise

most similar in terms of hypertrophy of the protoconid and

reduction of the metaconid.

Until now, the oldest known hypercarnivorous dasyurids

have been the dasyurine species of Dasyurus (earliest record

being middle Pliocene; Dasyurus dunmalli Bartholomai,

1971; Wroe and Mackness, 1998; Archer, 1982), Glaucodon

(only known from the Pliocene; Glaucodon ballaratensis

Stirton, 1957; Gerdtz and Archbold, 2003) and Sarcopliilus

(first known from deposits interpreted to be Early Pleistocene

in age; Sarcopliilus moornaensis Crabb, 1982). Although the

exact age of W. tomnpatrichorum is uncertain, it occurs with a

suite of taxa that, at least at the generic level (e.g., Ekaltadeta,

Hypsiprymnodon, cf. Rhizophascolonus), are broadly

comparable but not identical to Miocene assemblages known

from the Riversleigh World Heritage Area (Archer et al.,

2006). Differences include, for example, a new species of

Hypsiprymnodon that is abundantly present in the Wholly

Dooley Local Fauna. We have therefore tentatively concluded

that this faunal assemblage probably correlates with those

from either Faunal Zone C or D in the Riversleigh World

Heritage Area, which span approximately 16 to 13 Ma

(Woodhead et al., 2014). However, it could be younger than

this, possibly early late Miocene in age given the uncertain age

of Encore Site, which has been estimated on the basis of

biocorrelation to be early late Miocene (Black, 1997a, 1997b;

Myers et al., 2001; Brewer et al., 2007; Black et al., 2012;

Arena et al., 2014; Arena et al., 2015).

Wroe (2003) and Black et al. (2012) make the point, on the

basis of the rich Riversleigh record in particular, that there

appears to have been a gradual replacement of cat- and fox¬

sized thylacinids as the most abundant of the larger mammalian

terrestrial carnivores in the late Oligocene to middle Miocene

by comparable-sized dasyurids in the later Cenozoic.

Further, until now, few of the albeit rare dasyurids known

from the Oligo-Miocene were large enough to qualify as

hypercarnivores, Ganbulanyi djadjinguli being potentially the

only other hypercarnivorous dasyurid. In contrast,

Whollydooleya tomnpatrichorum was much larger than any of

the other dasyurids and most of the thylacinids known to have

been present during the Oligo-Miocene. Only late Miocene

thylacinids (e.g., Thylacinus potens from the Alcoota Local

Fauna in the Northern Territory) would have been larger.

Using the regression equations for M 3 and M 2 widths published

by Myers (2001; table 4; dasyuromorphian dataset), we

estimate that a minimum mass for W. tomnpatrichorum may

have been 20.3 kg if the holotype is an M 3 or 25.5 kg if it is an

M 2 . The average mass of living Tasmanian Devil males is

10-11 kg and of females is 7-8kg (Parks and Wildlife Service

New hypercarnivorous marsupial from the Riversleigh World Heritage Area

149

Tasmania). Hence W. tomnpatrichorum may well have been at

least twice the mass of living Devils, which are, currently,

Australia’s largest marsupial carnivore. Given Paddle’s (2000)

mass estimates for average-sized modern Thylacines

(Thylacinus cynoceplialus ) of 29.5 kg, it is possible that W.

tomnpatrichorum may have been as large as some of the

smaller individuals of this recently extinct hypercarnivore.

Contemporaneity of other large hypercarnivorous dasyurids

(e.g., species of Dasyurus and Sarcophilus ) with a large

thylacinid (71 cynocephalus ) has been the situation in Australia

since at least the Pliocene and persisted until Dingoes arrived

on the Australian mainland in the mid Holocene and Europeans

exterminated the Thylacine from Tasmania.

Whollydooleya tomnpatrichorum, however, seems

unlikely, on the basis of the hypertrophied entoconid, to have

been on the direct line leading to the species of Dasyurus,

Glaucodon or Sarcophilus. It is more likely to represent an

independent, probably mid to late Miocene lineage of

hypercarnivorous dasyurines that filled the niche occupied

later, in the Pliocene to Holocene, by large hypercarnivorous

dasyurines such as the living Tiger Quoll and Tasmanian Devil.

Appearance of this large hypercarnivorous dasyurid is the

first indication of a trend towards gigantism within this family,

which ultimately resulted in the largest known dasyurids of

the late Cenozoic. Given that it does not appear to have been a

member of the dasyurine lineage that includes Glaucodon and

Sarcophilus, it had to represent a second lineage of dasyurids

that underwent gigantism. If this was the case, it is possible

that competition between the two could have led to loss of the

lineage represented by W. tomnpatrichorum. That said, in

terms of larger dasyurines, the only known devil remains are in

the detail of the Pliocene and Quaternary record.

The presence in the deposit of what appear to be albeit rare

aeolian-transported quartz grains and the indications of at

least partial desiccation of the accumulating deposit may be a

reflection of changing palaeoenvironments within the region.

Following the late Oligocene and prior to the mid-Miocene

climate oscillation, the faunal assemblages (Faunal Zones B

and C) from the Riversleigh World Heritage area appear to

suggest closed, biologically-rich forests (Archer et al., 1994,

1997; Travouillon et al., 2009; Black et al., 2012). While very

few deposits from the Riversleigh World Heritage Area appear

to represent post-Miocene oscillation assemblages, the few

that do (e.g., Encore Site) suggest more open, drier forests

(Myers et al., 2001, Travouillon et al. 2009). Wholly Dooley

Site may have derived from a time when the region’s

palaeoenvironments were even drier, with potentially some

wind-blown components becoming parts of the accumulating

fossil deposit. Testing this possibility will involve ongoing

research into other components of the Wholly Dooley Local

Fauna as well as trying to radiometrically date speleothems

that have been obtained from Wholly Dooley Hill.

Acknowledgements

Support for research at Riversleigh has come from Australian

Research Council grants (LP100200486, DP1094569,

DP130100197, DE130100467 and DE120100957 to M. Archer,

S.J. Hand, K. H. Black, I. Graham and R.M.D. Beck at UNSW,

Australia); XSTRATA Community Partnership Program

(North Queensland); UNSW Australia; the National

Geographic Society; P. Creaser and the CREATE Fund; the

Queensland Parks and Wildlife Service; Environment

Australia; the Waanyi Nation; the Queensland Museum; the

Riversleigh Society Inc.; Outback at Isa; Mount Isa City

Council; private supporters including Ken and Margaret Pettit,

Elaine Clark, Margaret Beavis, and Martin Dickson; and the

Waanyi people of northwestern Queensland. Assistance in the

field has come from many hundreds of volunteers including

those from the Geological Society of the Hunter Valley, the

Waterhouse Club of South Australia, and the Riversleigh

Society as well as staff and postgraduate students of the

University of New South Wales. We thank Erich Fitzgerald of

Museum Victoria for access to the holotype of Sarcophilus

moornaensis and holotype as well as casts of the referred

specimen of Glaucodon ballaratensis. Wayne Gerdtz provided

advice about the origin of the referred specimen of G.

ballaratensis. Ned Stephenson is thanked for his early

recognition of the potential of the area now known as New

Riversleigh to produce new fossil deposits. Jon Woodhead has

developed the radiometric framework for dating speleothems

from the Riversleigh World Heritage Area.

References

Archer, M. 1976a. Miocene marsupicarnivores (Marsupialia) from

central South Australia, Ankotarinja tirarensis gen. et sp. nov.,

Keeuna woodburnei gen. et sp. nov., and their significance in

terms of early marsupial radiations. Transactions of the Royal

Society of South Australia 100: 53-73.

Archer, M. 1976b. The dasyurid dentition and its relationships to that

of didelphids, thylacinids, borhyaenids (Marsupicarnivora) and

peramelids (Peramelina: Marsupialia). Australian Journal of

Zoology Supplementary Series 39: 1-34.

Archer, M. 1982. Review of the dasyurid (Marsupialia) fossil record,

integration of data bearing on phylogenetic interpretation, and

suprageneric classification. Pp. 397-443 in: Archer, M. (ed.).

Carnivorous Marsupials. Surrey Beatty and Sons: Chipping

Norton.

Archer, M. and Rich, T.H. 1979. Wakamatha tasselli gen. et sp. nov., a

fossil dasyurid (Marsupialia) from South Australia convergent on

modern Sminthopsis. Memoirs of the Queensland Museum 19:

309-317.

Archer, M., Hand, S.J., and Godthelp, H. 1994. Riversleigh. Reed

Books Pty Ltd: Sydney. 264 pp.

Archer, M., Hand, S.J., Godthelp, H., and Creaser, P. 1997. Correlation

of the Cainozoic sediments of the Riversleigh World Heritage

Fossil Property, Queensland, Australia. Pp. 131-152 in: Aguilar,

J-P, Legendre, S., and Michaux, J. (eds), Actes du Congres

BiocroM’97. Memoirs et Traveaux Ecole Pratique des Hautes

Etudes. Institut de Montpellier: Montpellier.

Archer, M., Arena, D.A., Bassarova, M., Beck, R.M.D., Black, K.,

Boles, W.E., Brewer, P., Cooke, B.N., Crosby, K., Gillespie, A.,

Godthelp, H., Hand, S.J., Kear, B.P., Louys, J., Morrell, A.,

Muirhead, J., Roberts, K.K., Scanlon, J.D., Travouillon, K.J., and

Wroe, S. 2006. Current status of species-level representation in

faunas from selected fossil localities in the Riversleigh World

Heritage Area, northwestern Queensland. Alcheringa: An

Australasian Journal of Palaeontology 30:1-17.

150

M. Archer, 0. Christmas, S.J. Hand, K.H. Black, P. Creaser, H. Godthelp, I. Graham, D. Cohen, D.A. Arena,

C. Anderson, G. Soares, N. Machin, R.M.D. Beck, L.A.B. Wilson, T.J. Myers, A.K. Gillespie, B. Khoo & K.J. Travouillon

Arena, D.A. 2004. The geological history and development of the

terrain at the Riversleigh World Heritage Area during the middle

Tertiary. Unpublished Ph. D. Thesis, University of New South

Wales, Sydney. 275 pp.

Arena, D.A., Archer, M., Godthelp, H., Hand, S.J., and Hocknull, S.

2011. Hammer-toothed ‘marsupial skinks’ from the Australian

Cenozoic. Proceedings of the Royal Society B 278: 3529-3533.

Arena, D.A., Black, K.H., Archer, M., Hand, S.J., Godthelp, H., and

Creaser, P. 2014. Reconstructing a Miocene pitfall trap: recognition

and interpretation of fossiliferous Cenozoic palaeokarst.

Sedimentary Geology doi:10.1016/j.sedgeo.2014.01.005.

Arena, D.A., Travouillon, K.J., Beck, R.M.D., Black, K.H., Gillespie,

A.K., Myers, T.J., Archer, M., Hand, S.J. (2015). Mammalian

lineages and the biostratigraphy and biochronology of Cenozoic

faunas from the Riversleigh World Heritage Area, Australia.

Lethaia, doi:10.1111/let.l2131.

Bartholomai, A. 1971. Dasyurus dunmalli, a new species of fossil

marsupial (Dasyuridae) in the upper Cainozoic deposits of

Queensland. Memoirs of the Queensland Museum 16: 19-26.

Black, K. 1997a. A new species of Palorchestidae (Marsupialia) from

the late middle to early late Miocene Encore Local Fauna,

Riversleigh, northwestern Queensland. Memoirs of the

Queensland Museum 41: 181-186.

Black, K. 1997b. Diversity and biostratigraphy of the Diprotodontoidea

of Riversleigh, northwestern Queensland. Memoirs of the

Queensland Museum 41: 187-192.

Black, K.H., Archer, M., Hand, S. J., and Godthelp, H. 2012. The rise of

Australian marsupials: a synopsis of biostratigraphic, phylogenetic,

palaeoecological and palaeobiogeographic understanding. Pp.

983-1078 in: Talent, J.A. (ed.). Earth and Life: Global Biodiversity,

Extinction Intervals and Biogeographic Perturbations through

Time. Springer Verlag: Dordrecht.

Brewer, P, Archer, M., Hand, S., Godthelp, H. 2007. A new species of

the wombat Warendja from late Miocene deposits at Riversleigh,

north-west Queensland, Australia. Palaeontology 50: 811-828.

Crabb, P.L. 1982. Pleistocene dasyurids (Marsupialia) from southwestern

New South Wales. Pp. 511-16 in: Archer, M. (ed.) Carnivorous

Marsupials. Surrey Beatty and Sons: Chipping Norton.

Creaser, P. 1997. Oligocene-Miocene sediments of Riversleigh: the

potential significance of topography. Memoirs of the Queensland

Museum 41: 303-314.

Dawson, L. 1982. Taxonomic status of fossil devils ( Sarcophilus ,

Dasyuridae, Marsupialia) from late Quaternary eastern Australian

localities. Pp. 517-25 in: Archer, M. (ed.) Carnivorous Marsupials.

Surrey Beatty and Sons: Chipping Norton.

Every, R.G. 1970. Sharpness of teeth in man and other primates.

Postilla 143: 1-30.

Gerdtz, W. and Archbold, N. 2003. Glaucodon ballaratensis

(Marsupialia, Dasyuridae), a Late Pliocene ‘Devil’ from Batesford,

Victoria. Proceedings of the Royal Society of Victoria 115: 35-44.

Forasiepi, A.M., Babot, M.J. and Zimicz, N. 2014. Australohyaena

antiqua (Mammalia, Metatheria, Sparassodonta), a large predator

from the late Oligocene of Patagonia. Journal of Systematic

Palaeontology. doi:10.1080/14772019.2014.926403.

Godthelp, H., Wroe, S. and Archer, M. 1999. A new marsupial from

the early Eocene Tingamarra Local Fauna of Murgon, southeastern

Queensland: a prototypical Australian marsupial? Journal of

Mammalian Evolution 6: 289-313.

Goldfuss, G.A. 1820. Handbuch der Zoologie. Nuremberg, 196 pp.

Myers, T.J. 2001. Prediction of marsupial body mass. Australian

Journal of Zoology 49: 99-118.

Myers, T.J., Crosby, K., Archer, M. and Tyler, M. 2001. The Encore

Local Fauna, a late Miocene assemblage from Riversleigh,

northwestern Queensland. Memoirs of the Association of

Australasian Palaeontologists 25: 147-154.

Owen, R. 1838. Fossil remains from Wellington Valley, Australia.

Marsupialia. Pp 359-369 in: Appendix to Mitchell, T.L. (ed.). Three

expeditions into the interior of eastern Australia, with descriptions

of the recently explored region of Australia felix, and of the present

colony of New South Wales. T. and W. Boone: London.

Parks and Wildlife Service Tasmania. No date. Tasmanian Devil—

frequently asked questions. http://www.parks.tas.gov.

au/?base=4756 accessed 25 June, 2015.

Paddle, R. 2000. The Last Tasmanian Tiger: the History and

Extinction of the Thylacine. Cambridge University Press:

Melbourne. 284 pp.

Stirton, R.A. 1957. Tertiary marsupials from Victoria, Australia.

Memoirs of the National Museum of Victoria 21: 121-134.

Thomas, O. 1888. Catalogue of the Marsupialia and Monotremata in

the collection of the British Museum (Natural History). British

Museum (Natural History): London. 401 pp.

Travouillon, K.J., Legendre, S., Archer, M., and Hand, S.J. 2009.

Palaeoecological analyses of Riversleigh’s Oligo-Miocene sites:

implications for Oligo-Miocene climate change in Australia.

Palaeogeography, Palaeoclimatology, Palaeoecology 276: 24-37.

Voss, R.S. and Jansa, S.A. 2009. Phylogenetic relationships and

classification of didelphid marsupials of New World metatherian

mammals. Bulletin of the American Museum of Natural History

322: 1-177.

Wroe, S. 1996. Muribacinus gadiyuli (Thylacinidae; Marsupialia), a

very plesiomorphic thylacinid from the Miocene of Riversleigh,

northwestern Queensland, and the problem of paraphyly for the

Dasyuridae (Marsupialia). Journal of Paleontology 70: 1032-1044.

Wroe, S. 1997a. Mayigriphus orbus, anew species of dasyuromorphian

(Marsupialia) from the Miocene of Riversleigh, north-western

Queensland. Memoirs of the Queensland Museum 41: 439-448.

Wroe, S. 1997b. A reexamination of proposed morphology-based

synapomorphies for the families of Dasyuromorphia (Marsupialia).

I. Dasyuridae. Journal of Mammalian Evolution 4: 19-52.

Wroe, S. 1998. A new ‘bone-cracking’ dasyurid (Marsupialia), from

the Miocene of Riversleigh, north-western Queensland.

Alcheringa 22: 277-284.

Wroe, S. 1999. The geologically oldest dasyurid (Marsupialia) from

the Miocene of Riversleigh, north-western Queensland.

Palaeontology 42: 501-527.

Wroe, S. 2001. A new genus and species of dasyuromorphian from the

Miocene of Riversleigh, northern Australia. Memoirs of the

Association of Australasian Palaeontologists 25: 53-59.

Wroe, S., 2003. Australian marsupial carnivores: recent advances in

palaeontology. Pp. 102-123 in: Jones, M., Dickman, C. and

Archer, M. (eds), Predators with Pouches: the Biology of

Marsupial Carnivores. CSIRO Publishing: Collingwood.

Wroe, S., and Mackness, B.S. 1998. Revision of the Pliocene dasyurid,

Dasyurus dunmalli (Dasyuridae: Marsupialia). Memoirs of the

Queensland Museum 42: 605-612.

Woodhead, J., Hand, S.J., Archer, M., Graham, I., Sniderman, K.,

Arena, D.A., Black, K.H., Godthelp, H., Creaser, P., and Price, L.

2014. Developing a radiometrically dated chronologic sequence

for Neogene biotic change in Australia, from the Riversleigh

World Heritage Area of Queensland. Gondwana Research.

doi:10.1016/j.gr.2014.10.004.

Memoirs of Museum Victoria 74:151-171 (2016) Published 2016

ISSN 1447-2546 (Print) 1447-2554 (On-line)

http://museumvictoria.com.au/about/books-and-journals/journals/memoirs-of-museum-victoria/

Going underground: postcranial morphology of the early Miocene marsupial mole

Naraboryctes philcreaseri and the evolution of fossoriality in notoryctemorphians

Robin M. D. Beck 1 *, Natalie M. Warburton 2 , Michael Archer 3 , Suzanne J. Hand 4 , and Kenneth R Aplin 5

1 School of Environment & Life Sciences, Peel Building, University of Salford, Salford M5 4WT, UK and School of

Biological, Earth and Environmental Sciences, University of New South Wales, Sydney, NSW 2052, Australia

(R.M.D.Beck@salford.ac.uk)

2 School of Veterinary and Life Sciences, Murdoch University, 90 South Street, Murdoch, WA 6150, Australia

(N.Warburton@murdoch.edu. au)

3 School of Biological, Earth and Environmental Sciences, University of New South Wales, Sydney, NSW 2052,

Australia (m.archer@unsw.edu.au)

4 School of Biological, Earth and Environmental Sciences, University of New South Wales, Sydney, NSW 2052,

Australia (s.hand@unsw.edu.au)

5 National Museum of Natural History, Division of Mammals, Smithsonian Institution, Washington, DC 20013-7012,

USA (aplink@si.edu)

* To whom correspondence should be addressed. E-mail: R.M.D.Beck@salford.ac.uk

Abstract Beck, R.M.D., Warburton, N.M., Archer, M., Hand, S. J. and Aplin, K.P. 2016. Going underground: postcranial morphology

of the early Miocene marsupial mole Naraboryctes philcreaseri and the evolution of fossoriality in notoryctemorphians.

Memoirs of Museum Victoria 74: 151-171.

We present the first detailed descriptions of postcranial elements of the fossil marsupial mole Naraboryctes

philcreaseri (Marsupialia: Notoryctemorphia), from early Miocene freshwater limestone deposits in the Riversleigh

World Heritage Area, northwestern Queensland. Qualitative functional analysis of these elements suggest that Na.

philcreaseri was very well-adapted for burrowing, albeit somewhat less so than the living marsupial moles Notoryctes

typhlops and N. caurinus. Quadratic discriminant analysis of limb measurements suggests that Na. philcreaseri was

subterranean, and its Index of Fossorial Ability is almost identical to that of Notoryctes species, being among the

highest known for any mammal. These results suggest that notoryctemorphians evolved their specialised, “mole-like”

subterranean lifestyle prior to the early Miocene. Given that forested environments predominated in Australia until the

middle-late Miocene, this transition to subterranean behaviour may have occurred via burrowing in forest floors, in

which case fossorial mammals that live in tropical rainforests today (such as the placental golden moles Chrysospalax

trevelyani and Huetia leucorhina) may represent reasonable living analogues for early notoryctemorphians. However,

alternative scenarios, such as a cave-dwelling or semi-aquatic ancestry, should be considered. Phylogenetic analysis

using a Bayesian total evidence dating approach places Naraboryctes as sister to Notoryctes with strong support

(Bayesian posterior probability = 0.91), and indicates that Naraboryctes and Notoryctes diverged 30.3 MYA (95%

HPD: 17.7-46.3 MYA). The age and known morphology of Na. philcreaseri does not preclude its being ancestral to

Notoryctes. Using estimates of divergence times and ratios of nonsynonymous to synonymous substitutions per site,

we infer that the nuclear gene “Retinol-binding protein 3, interstitial” ( RBP3 ), which plays a key role in vision,

became inactive in the Notoryctes lineage ~5.4 MYA (95% HPD: 4.5-6.3 MYA). This is much younger than previous

published estimates, and postdates considerably the age of Na. philcreaseri, implying that RBP3 was active in this

fossil taxon; hence, Na. philcreaseri may have retained a functional visual system. Our estimate for the inactivation of

RBP3 in the Notoryctes lineage coincides with palaeobotanical evidence for a major increase in the abundance of

grasses in Australia, which may indicate the appearance of more open environments, and hence selection pressure on

notoryctemorphians to spend less time on the surface, leading to relaxed selection on RBP3. Ultimately, however, a

fuller understanding of the origin and evolution of notoryctemorphians - including when and why they became “mole¬

like” - will require improvements in the Palaeogene fossil record of mammals in Australia.

Keywords marsupial moles; Notoryctes-, Naraboryctes-, Notoryctemorphia; fossorial; subterranean; postcranial; functional morphology;

marsupial phylogeny; Riversleigh; Miocene; RBP3; IRBP

152

“Some people might say my life is in a rut,

But I’m quite happy with what I got”

Going Underground - The Jam

Introduction

The two living species of marsupial mole - Notoryctes

typhlops (the Southern Marsupial Mole, or Itjaritjari) and

Notoryctes caurinus (the Northern or Northwestern Marsupial

Mole, or Kakarratul) - are remarkably specialised,

subterranean mammals that live in the western deserts of

continental Australia (Johnson and Walton, 1989; Benshemesh

and Johnson, 2003; Benshemesh, 2008; Benshemesh and

Aplin, 2008). The only extant representatives of the order

Notoryctemorphia (Kirsch, 1977; Aplin and Archer, 1987),

both species show extreme anatomical adaptations for

burrowing, including a heavily fused and conical skull, loss of

external ears and functional eyes, and a postcranial skeleton

that is highly modified for parasagittal scratch-digging

(Stirling, 1891; Carlsson, 1904; Sweet, 1906; Johnson and

Walton, 1989; Warburton, 2006). One striking molecular

feature of Notoryctes is also probably related to its subterranean

lifestyle: the gene “Retinol-binding protein 3, interstitial”

(. RBP3 , often referred to as “interphotoreceptor retinoid

binding protein”, or IRBP ) of N. typhlops is non-functional,

exhibiting both frameshift mutations and premature stop

codons (Springer et al., 1997; Emerling and Springer, 2014).

RBP3 plays a key role in visual pigment regeneration

(Pepperberg et al., 1993), and loss of function of this gene in N.

typhlops is presumably related to its degenerate visual system:

its eyes are tiny, lack lenses and are covered by skin, and the

optic nerve is absent (Stirling, 1891; Sweet, 1906). The visual

system of N. caurinus is similarly degenerate (Benshemesh

and Aplin, 2008), but to date RBP3 has not been sequenced for

this species. RBP3 has also been shown to be pseudogenic in

some subterranean rodents (Kim et al., 2011: table 2; Emerling

and Springer, 2014) and in several echolocating bat species

(Shen et al., 2013); as in Notoryctes , vision is probably of

limited importance in these species, and hence selection to

maintain a functional copy of RBP3 has presumably been

relaxed (Kim et al., 2011; Shen et al., 2013; Emerling and

Springer, 2014).

Both Notoryctes species spend the vast majority of their

time below ground, surfacing only rarely (Johnson and Walton,

1989; Benshemesh and Johnson, 2003; Dennis, 2004;

Benshemesh, 2008; Benshemesh and Aplin, 2008); because of

this, many basic aspects of their ecology and life history are

unknown. Fundamental questions regarding the evolutionary

history of notoryctemorphians also remain unanswered.

Resolution of their phylogenetic relationships has proven

difficult due to their numerous morphological autapomorphies,

lack of obvious close living relatives and, until recently,

complete absence of a fossil record. Indeed, some studies have

even questioned whether they are marsupials (Stirling, 1888;

Cope, 1892; Turnbull, 1971). However, there is now

overwhelming evidence that Notoryctes is indeed a marsupial

(e.g. Horovitz and Sanchez-Villagra, 2003; Asher et al., 2004;

Nilsson et al., 2004, 2010; Beck, 2008; Beck et al., 2008;

R.M.D. Beck, N.M. Warburton, M. Archer, S.J. Hand, and K.P. Aplin

Meredith et al., 2008, 2009b, 2011; Mitchell et al., 2014).

Recent phylogenetic analyses, particularly those based on

molecular data (e.g. Nilsson et al., 2004, 2010; Beck, 2008;

Meredith et al., 2008, 2009b, 2011; Mitchell et al., 2014),

typically place Notoryctes in a clade with the Australian

marsupial orders Dasyuromorphia (predominantly carnivorous

forms such as quolls, thylacines and the numbat) and

Peramelemorphia (bandicoots and bilbies); Beck et al. (2014)

named this clade Agreodontia. However, the precise branching

relationship between Notoryctemorphia, Dasyuromorphia and

Peramelemorphia is uncertain. The previous lack of fossil

evidence has also meant that the underlying factors driving the

evolution of the extreme fossorial adaptations of Notoryctes,

and the timing of when they arose, remain enigmatic.

Crucial information regarding the evolutionary history of

Notoryctemorphia is now being provided with the discovery

of cranial and postcranial remains of the first known fossil

member of the order, Naraboryctes philcreaseri, from early

Miocene freshwater limestone deposits at Riversleigh (Gott,

1988; Warburton, 2003; Archer et al., 2011). Na. philcreaseri

provides the first hard evidence showing how the zalambdodont

molar dentition of Notoryctes evolved, namely via suppression

of the paracone (Archer et al., 2011). In addition, postcranial

elements tentatively referred to Na. philcreaseri show apparent

fossorial specialisations, albeit not as extreme as in Notoryctes

species (Warburton, 2003; Archer et al., 2011). The presence

of these features in Na. philcreaseri is perhaps surprising

because available evidence suggests that Riversleigh was a

rainforest environment during the early Miocene (Travouillon

et al., 2009; Archer et al., 2011; Bates et al., 2014). It had

previously been suggested that notoryctemorphians acquired

their fossorial adaptations in a desert milieu (Archer, 1984),

but there is no evidence of deserts anywhere in Australia prior

to the Pleistocene (McLaren and Wallace, 2010), nor extensive

grasslands prior to the middle Pliocene (Martin and McMinn,

1994; Martin, 2006; Stromberg, 2011; Black et al., 2012). The

discovery of Na. philcreaseri has led to an alternative

hypothesis, namely that these adaptations evolved for

burrowing through soft rainforest floors (Archer et al., 1994;

Archer et al., 2011). In support of this, analogies have been

drawn with the morphologically similar but unrelated placental

chrysochlorids (golden moles) of sub-Saharan Africa (Archer

et al., 1994), as some extant chrysochlorid species (e.g.

Chrysospalax trevelyani, Huetia leucorhina) occur in wet

forest environments (Bronner, 2013).

In this paper, we present the first detailed descriptions of

the postcranial skeleton of Naraboryctes philcreaseri,

including bones not discussed by Archer et al. (2011). We

present a qualitative functional interpretation of its postcranial

morphology, focusing on evidence for fossoriality, and test this

within a quantitative framework using multivariate discriminant

analysis (see Hopkins and Davis, 2009). We also calculate the

widely-used Index of Fossorial Ability (Vizcaino et al., 1999;

Vizcaino and Milne, 2002) for Na. philcreaseri and compare it

to values for Notoryctes typhlops and for other burrowing

mammals. We discuss the changes required to derive the

postcranial morphology seen in Notoryctes species from that

seen in Na. philcreaseri. We present the first phylogenetic

Fossil marsupial mole postcrania

153

analyses to include Naraboryctes, using a dated Bayesian total

evidence dating approach (Ronquist et al., 2012a), the first time

this method has been applied to marsupials. Finally, we use the

divergence dates from our phylogenetic analysis and estimates

of non-synonymous to synonymous substitution ratios to

estimate the timing of inactivation of the RBP3 gene in the

notoryctemorphian lineage (Chou et al., 2002; Zhang et ah,

2008), and discuss its implications for our understanding of the

ecology of Na. philcreaseri.

In 1971, Tom and Pat Rich undertook their first fieldwork

in Australia, working with the American Museum of Natural

History’s Dick Tedford, and using “conventional” fossil

collecting methods (Archer and Hand, 1984). In 1984, however,

Tom, Pat and a team of brave volunteers turned to tunnelling

as a way to obtain fossils from Dinosaur Cove, on the south

coast of Victoria. This was apparently the first time that a

tunnel had ever been dug solely to find fossils, and over the

following ten years it proved a remarkable success, albeit one

requiring immense efforts in terms of time and manpower.

Tom has gone on to use a similar tunnelling approach in Late

Cretaceous deposits in Alaska. We think, therefore, that a

paper on the evolution of burrowing in marsupial moles is

particularly appropriate for this volume celebrating Tom’s

long and productive career and his extraordinary contribution

to Australian palaeontology.

Materials and methods

As with most other vertebrate specimens collected from fossil

deposits in the Riversleigh World Heritage Area, the material

of Naraboryctes philcreaseri described here was obtained by

processing limestone blocks with 5-10% acetic acid, and

subsequent microscope-assisted sorting of the concentrate.

None of the postcranial elements were found in direct

association with dental specimens of Na. philcreaseri ; referral

is based on compatible size and presence of apparent fossorial

adaptations, particularly features that closely resemble the

morphology seen in Notoryctes. It is possible that some or all

of these elements in fact belong to a different taxon, but we

consider this unlikely: no mammalian group present at

Riversleigh besides Notoryctemorphia is known to have either

living or fossil members that show extreme specialisations for

digging. Nevertheless, as with all specimens not found in

direct association, these referrals should be considered

tentative. All known specimens of Na. philcreaseri are from

Riversleigh Faunal Zone B sites, which have been interpreted

to be early Miocene in age based on biostratigraphy (Archer et

al., 1997; Archer et al., 2006; Travouillon et al., 2006), an age

assignment that is now being validated by radiometric dating

(Woodhead et al., 2014). A full list of Na. philcreaseri

postcranial specimens is given in supplementary information;

they are registered in the fossil vertebrate collection of the

Queensland Museum (prefix QM F).

Description and functional interpretation of the postcranial

morphology of Na. philcreaseri follows previous studies of

fossorial and subterranean mammals, both living and fossil

(e.g. Stirling, 1891; Thompson and Hillier, 1905; Chapman,

1919; Campbell, 1939; Orcutt, 1940; Reed, 1951; Lehmann,

1963; Puttick and Jarvis, 1977; Rose and Emry, 1983; Gasc et

al., 1986; Lessa, 1990; Stein, 1993; Warburton, 2006), as well

as more general studies of mammalian morphology (e.g.

Coues, 1872; Barbour, 1963; Davis, 1964; Schaller, 1992;

Evans, 1993). We consider mammals to be “fossorial” if they

engage in regular burrowing activities below ground but

nevertheless spend considerable periods of time above ground,

and “subterranean” if they spend the clear majority of their

time below ground (Nevo, 1999: character 2; Lessa et al.,

2008); under this definition, both living Notoryctes species are

subterranean. Measurements for Na. philcreaseri were taken

from complete adult specimens, where available, while

measurements for N. caurinus and N. typhlops were taken

from Warburton (2003: Appendix 1) and an additional N.

typhlops specimen (SAM M637).

To quantitatively test whether the known morphology of

Na. philcreaseri supports its interpretation as a fossorial or

subterranean mammal, we used a quadratic discriminant

analysis based on the “limbs only” dataset of Hopkins and

Davis (2009). This dataset comprises four measurements from

the humerus, two from the ulna, and two from the femur, for

115 extant subterranean, fossorial and non-burrowing mammal

species (including placentals, marsupials and monotremes; see

Hopkins and Davis, 2009 for full details). Hopkins and Davis

(2009) found this dataset to be 86.1% accurate at distinguishing

burrowing (= subterranean + fossorial) and non-burrowing

taxa, and 85.2% accurate for distinguishing subterranean,

fossorial and non-burrowing taxa. Hopkins and Davis (2009)

estimated measurements for Notoryctes typhlops from

published images; we have replaced these with measurements

taken directly from N. typhlops specimen SAM M637. We

added a further three modern marsupial taxa to the Hopkins

and Davis (2009) dataset, namely the non-burrowing dasyurid

Dasyurus viverrinus, the non-burrowing peramelemorphians

Isoodon macrourus, I. obesulus and Perameles gunnii, and

the fossorial peramelemorphian Macrotis lagotis. Na.

philcreaseri was then added, with measurements taken from

the most complete specimens and with its fossorial ability

treated as unknown. Measurements for these additional taxa

are given in supplementary information. The quadratic

discriminant analysis was implemented in JMP, assuming

equal prior probabilities for the different locomotor modes,

and a plot of the first two canonical axes was produced,

following Hopkins and Davis (2009: fig. 3).

To further quantify the likely fossorial ability of Na.

philcreaseri, we compared its Index of Fossorial Ability (IFA;

sometimes referred to as Olecranon Length Index) - a

commonly used metric that is strongly correlated with the

degree of fossoriality in various groups of mammals (Vassallo,

1998; Vizcaino et al., 1999; Elissamburu and Vizcaino, 2004;

Lagaria and Youlatos, 2006; Warburton et al., 2013; Woodman

and Gaffney, 2014) - with that of Notoryctes and other

fossorial and subterranean mammals (table 1). We calculated

IFA as (olecranon length/(total ulnar length-olecranon

length))* 100, following Vizcaino and Milne (2002). Olecranon

length was measured to the distal margin of the trochlear

notch, as in Hopkins and Davis (2009; see also comments by

Woodman and Gaffney, 2014: table 1). For Na. philcreaseri

154

R.M.D. Beck, N.M. Warburton, M. Archer, S.J. Hand, and K.P. Aplin

Table 1. Selected Recent and fossil mammals with an Index of Fossorial Ability (IFA) greater than 60. Identification of Recent taxa as

subterranean, fossorial or non-burrowing follows Hopkins and Davis (2009). Fossil taxa are indicated by f, and their burrowing behaviour is

treated as unknown. All measurements are taken from Hopkins and Davis (2009) except those for Naraboryctes philcreaseri (which were taken

from QM F57706; see Figs. 4a, c, e) and Notoryctes typhlops (which were taken from SAM M637; see Figs. 4b, d, f). IFA is calculated as

(olecranon length/(total ulnar length-olecranon length))* 100, following Vizcaino and Milne (2002). IFA for Na. philcreaseri is likely a slight

overestimate, as QM F57706 is missing the distal epiphysis (see Figs. 4a, c, e), and hence total ulnar length is slightly too short; the true value is

probably the same as or slightly less than that for N. typhlops.

taxon

order

family

type

total ulnar

length

(mm)

Olecranon

length

(mm)

Index of

fossorial

ability

(IFA)

f Xenocranium pileorivale

fPalaeanodonta

fEpoicotheriidae

unknown

38.46

22.08

134.7

Amblysomus hottentotus

Afrosoricida

Chrysochloridae

subterranean

17.56

8.96

104.2

Priodontes maxirnus

Cingulata

Dasypodidae

fossorial

131.96

67.07

103.4

Cabassous centralis

Cingulata

Dasypodidae

fossorial

58.15

27.88

92.1

f Pentapassalus pearcei

fPalaeanodonta

fEpoicotheriidae

unknown

89.7

t Naraboryctes philcreaseri

Notoryctemorphia

Notoryctidae

unknown

17.9

8.4

88.4

Notoryctes typhlops

Notoryctemorphia

Notoryctidae

subterranean

16.1

7.5

87.2

Scaptochirus moschatus

Eulipotyphla

Talpidae

subterranean

15.93

7.36

85.9

Scapanus orarius

Eulipotyphla

Talpidae

subterranean

16.95

7.8

85.2

Euphractus sexcinctus

Cingulata

Dasypodidae

fossorial

70.83

30.72

76.6

Spalax giganteus

Rodentia

Spalacidae

subterranean

49.81

21.44

75.6

Chaetophractus villosus

Cingulata

Dasypodidae

fossorial

60.44

25.84

74.7

Dasypus novemcinctus

Cingulata

Dasypodidae

fossorial

76.63

32.58

74.0

Nannospalax leucodon

Rodentia

Spalacidae

subterranean

33.3

14.04

72.9

Talpa europaea

Eulipotyphla

Talpidae

subterranean

18.27

7.57

70.7

Scalopus aquaticus

Eulipotyphla

Talpidae

subterranean

17.97

7.44

70.6

Eremitalpa grand

Afrosoricida

Chrysochloridae

subterranean

12.16

4.98

69.4

f Mesoscalops montanensis

?Eulipotyphla

fProscalopidae

unknown

24.83

9.6

63.0

Manis pentadactyla

Pholidota

Manidae

non-burrowing

60.7

23.46

63.0

Parascalops breweri

Eulipotyphla

Talpidae

subterranean

15.66

6.04

62.8

f Palaeanodon ignavus

fPalaeanodonta

fMetacheiromyidae

unknown

61.4

Tolypeutes matacus

Cingulata

Dasypodidae

non-burrowing

42.71

16.1

60.5

Fossil marsupial mole postcrania

155

and N. typhlops, we calculated IFA based on our own

measurements, whereas values for other taxa were calculated

from measurements for total ulnar length and olecranon length

given in Hopkins and Davis (2009).

Our phylogenetic analysis is based on the total evidence

matrix used by Beck et al. (2014), specifically the version that

includes character scores from isolated tarsals tentatively

referred to Yalkaparidon. This matrix comprises 258

morphological characters and 9012 bp from five nuclear genes,

namely APOB (exon 26), BRCA1 (exon 11), RBP3 (exon 1),

RAG1, and VWF (exon 28), and includes a range of fossil and

extant metatherians and non-metatherian outgroups (see Beck

et al., 2014, and supplementary information for full details). We

subdivided character 192 of Beck et al. (2014), such that

presence/absence of the transverse canal foramen and its

position (if present) are now treated as separate characters

(characters 192-193), and hence the morphological matrix used

here comprises a total of 259 characters. We also deleted the

indels and retroposons included by Beck et al. (2014) because

we used a total evidence dating approach (see below) and it is

not obvious how rare genomic changes should be treated in this

framework. We added Naraboryctes to this matrix, with

character scores based on the craniodental and postcranial

material described by Archer et al. (2011) and here. As in Beck

et al. (2014), an appropriate partitioning scheme for the nuclear

sequence data was determined using PartitionFinder (Lanfear

et al., 2012), with initial partitioning by gene and codon

position, and assuming linked branch lengths. Only models

implemented by MrBayes were tested, using the “greedy”

heuristic search algorithm, with the Bayesian Information

Criterion used for model selection. The morphological partition

was assigned the Mk model (Lewis, 2001), assuming that only

parsimony-informative characters were scored, and with a

gamma distribution to model rate heterogeneity between

characters. As in Beck et al. (2014), multistate morphological

characters representing putative morphoclines were ordered.

We employed a Bayesian total evidence dating approach

(Ronquist et al., 2012a; Beck and Lee, 2014), which

simultaneously estimates phylogeny and divergence times for

both extant and fossil taxa, as implemented in MrBayes 3.2.2

(Ronquist et al., 2012b). Each terminal taxon was assigned an

age: Recent taxa were assigned a point estimate of 0 Ma,

whilst each fossil taxon was assigned an age range as a hard-

bounded uniform prior, based on current age estimates (see

supplementary information for full justification). The age of

the root was also constrained as an offset exponential prior,

with a minimum of 176.15 Ma and a mean (expectation) of

201.3 Ma. The minimum is based on the minimum radiometric

age of the Queso Rellado locality of the Canadon Asfalto

Formation (Ciineo et al., 2013), which contains the oldest

known putative crown-group mammals, Asfaltomylos and

Henosferus, which are usually recovered as stem-monotremes

in published phylogenetic analyses (e.g. Rougier et al., 2007;

Bi et al., 2014). The mean corresponds to the Triassic-Jurassic

boundary, and is broadly congruent with recent molecular

estimates for the age of Mammalia (Meredith et al., 2011; dos

Reis et al., 2012; 2014). Analysis of eutherian mammals

suggests that total evidence dating in which only the age of the

root is constrained can result in implausibly ancient divergence

dates for at least some nodes (Beck and Lee, 2014); we

therefore specified 13 additional topological and temporal

constraints on internal nodes, with ages specified as offset

exponential priors, namely: Theria, Marsupialia, Didelphidae,

Didelphinae, crown-group Australidelphia, Dasyuridae,

Dasyurinae, Peramelidae, Diprotodontia, Phalangerida,

Petauroidea, Macropodidae and Vombatiformes (see

supplementary information for full details). To assist

convergence of the Bayesian analyses, monophyly of Eutheria

and Metatheria was also enforced a priori, but the ages of

these nodes were not calibrated.

The independent gamma rates (IGR) model was used

(Ronquist et al., 2012a), with a single clock model applied to

the entire nuclear sequence partition and a separate clock

model applied to the morphological partition. The MrBayes

analysis comprised four runs of four chains, each run for 15

million generations and with the temperature of the heated

chains increased to 0.2. A burn-in fraction of 25% (i.e. 3.75

million generations) was specified; examination of plots of log

likelihood against number of generations revealed that

stationarity and convergence between chains had been achieved

prior to this. The post-burnin trees were then summarised

using majority-rule consensus, with all compatible partitions

retained. Bayesian posterior probabilities (BPPs) were used as

measures of clade support. The MrBayes file used in our

analysis and the post-burnin majority-rule consensus that

resulted are included in supplementary information.

To calculate the time of inactivation of RBP3 in the

notoryctemorphian lineage we used the general approach of

Chou et al. (2002; see also Zhang et al., 2008). We first pruned

our RBP3 exon 1 alignment and the final consensus tree from

our dated total evidence analysis so that only extant marsupials

were retained (i.e. all fossil taxa and the monotreme outgroups

Ornithorhynchus and Tachyglossus were deleted). We then

deleted all codon positions in the pruned alignment that

contained insertions. The modified RBP3 alignment and

pruned tree were then used to calculate the ratio of non-

synonymous substitutions per non-synonymous site to

synonymous substitutions per synonymous site (co, also

referred to as d N /d s , or K/K s ). Using the codeml package of

PAML 4.7 (Yang, 2007), we firstly calculated to for all

branches, and then calculated co separately for: 1) the branch

leading to Notoryctes (the “foreground” branch; co f ), and 2) all

other branches (the “background” branches; co b ). We used a

likelihood ratio test to compare the relative fit of the two

models (i.e. single co, versus co f for Notoryctes and co b for all

other branches), with critical values taken from a x 2

distribution with a single degree of freedom (see Jansa and

Voss, 2011). Values for co f and co b were then entered into the

equation of Zhang et al. (2008: supplementary materials), to

calculate the time of gene inactivation (T ): T N = T((co f - co b )/

(1- ro b )), where T is the estimated time of divergence of the

Notoryctes lineage (which was taken from the results of our

total evidence dating analysis). This method assumes that,

after a gene becomes inactive, co for that gene becomes 1 (i.e.,

that non-synonymous substitutions and synonymous

substitutions occur with the same frequency).

156

R.M.D. Beck, N.M. Warburton, M. Archer, S.J. Hand, and K.P. Aplin

Anatomical description

General. Most of the limb bones of Na. philcreaseri discovered

to date are missing the epiphyses, indicating that the epiphyseal

sutures were unfused. Those that do preserve the epiphyses

retain open sutures. In part, this may reflect the general

marsupial pattern of most epiphyseal sutures remaining unfused

throughout adult life (Geiger et al., 2014). However, based on

their small size and very porous metaphyseal surfaces, at least

some of the Na. philcreaseri specimens appear to represent

juveniles (e.g., QM F57696, a right femur; QM F57703, a partial

left ulna). By contrast, limb epiphyseal sutures are consistently

fused in the skeletal specimens of Notoryctes typhlops and N.

caurinus that we have examined, but these all appear to be

adults based on the presence of a fully erupted permanent

dentition (juvenile specimens of Notoryctes are exceptionally

rare in scientific collections). Consistent closure of the epiphyseal

sutures in the limbs of Notoryctes adults (which contrasts with

the condition observed in most other marsupial species; Geiger

et al. 2014) may reflect the high mechanical loadings their limbs

are subjected to during digging; however, we note that Geiger et

al. (2014) examined both fossorial and non-fossorial mammals

and found no clear relationship between locomotor mode and

the sequence of epiphyseal closure (see also Meier et al., 2013).

Scapula. Three partial scapulae are referable to Na. philcreaseri

(see supplementary information), of which the most complete

(QM F57716, which lacks only the acromion) is illustrated here

(fig. la). In Notoryctes (and also some other burrowing

mammals, such as talpids and chrysochlorids; Edwards, 1937;

Reed, 1951; Puttick and Jarvis, 1977: fig. 3; Gasc et al., 1986),

the scapular spine is oriented roughly parallel to the anterior

thoracic vertebral column; hence the so-called “dorsal” (or

“vertebral”) border of the scapula is actually positioned

caudally (rather than dorsomedially, towards the vertebral

column), the “cranial” border is positioned dorsomedially, and

the “caudal” (or “axillary”) border is positioned ventrolaterally

(Warburton, 2006). Based on its morphology, it seems likely

that the scapula was similarly oriented in Na. philcreaseri.

Nevertheless, to avoid confusion and to maintain consistency

with anatomical descriptions of other mammals, we will refer

to “cranial”, “dorsal” and “caudal” borders in the scapula of

Na. philcreaseri and Notoryctes following the terminology of,

inter alia, Schaller (1992) and Evans (1993).

In dorsal view, the scapula appears narrow and elongate

compared to those of most other marsupials. The supraspinous

fossa is subtriangular in shape, and much larger in surface

area than the infraspinous fossa. The infraspinous fossa is

elongate and uniform in breadth along its length; its length-to-

width ratio is approximately 10:1. Although slightly damaged

in all three specimens, the scapular spine is clearly well-

developed along its entire length, and its glenoid third is

markedly broadened, slightly overhanging the infraspinous

fossa. The angle between the “cranial” and “dorsal” borders of

the scapula is approximately 80°, but is smoothly rounded.

The “dorsal” border is short, rounded, and thickened for

muscle attachment (principally the m. rhomboideus).

The glenoid third of the “caudal” border is markedly

raised, forming a prominent secondary scapular spine that is

separated from the scapular spine by the intervening

infraspinous fossa. Although slightly damaged in QM F57716,

the “caudal” angle appears rounded, thickened and slightly

extended “caudally” (parallel to the “dorsal” border), for

attachment of the m. teres major. The subscapular fossa is

smoothly concave. The glenoid is large relative to the size of

the supra- and infraspinous fossae, and the glenoid cavity is

elliptical and relatively long. A small raised area from the

“cranial” side of the glenoid cavity marks the coracoid process.

The acromion is not preserved in any of the three specimens.

The elongate scapula of Na. philcreaseri bears a strong

resemblance to that of Notoryctes (fig. lb) and is broadly

similar in size: maximum length of QM F57716 is 15.7 mm,

compared to 13-13.5 mm in N. caurinus and 14.3-18.4 mm in

N. typhlops. However, the modifications of the muscular

attachment sites are markedly less extreme in the fossil species.

Most notably, the “cranial” and “caudal” angles of the scapula

(for attachment of the m. subscapularis and m. teres major

respectively) appear to be gently rounded in Na. philcreaseri

(although the caudal angle is slightly damaged in QM F57716,

and hence its exact shape when intact is unclear), whereas in

Notoryctes they are elongated and modified into recurved,

hook-shaped processes, particularly the “caudal” angle, giving

the bone an overall fan shape. Notoryctes also has a prominent

postscapular fossa “caudal” to the secondary scapular spine,

Figure 1. Comparison of scapulae of Naraboryctes philcreaseri and

Notoryctes typhlops in lateral view: a, left scapula of Naraboryctes

philcreaseri (QM F57716); b, right scapula (reversed) of Notoryctes

typhlops (SAM M637). Abbreviations: acr, acromion process; cau,

“caudal” angle; cor, coracoid process; era, “cranial” angle; inf,

infraspinous fossa; psf, postscapular fossa; ssf, supraspinous fossa;

ssp, scapular spine; sssp, secondary scapular spine.

Fossil marsupial mole postcrania

157

reflecting enlargement of the scapular (long) head of the m.

triceps brachii. Enlargement of the triceps is common in

fossorial mammals, particularly scratch diggers, and is also

associated with elongation of the olecranon on the ulna

(resulting in a larger IFA; Windle and Parsons, 1899; Taylor,

1978; Gasc et al., 1986; Stein, 1986; Lagaria and Youlatos,

2006; Warburton, 2006; Warburton et al., 2013). In Na.

philcreaseri, the development of the secondary scapular spine

reflects enlargement of the triceps compared to non-fossorial

mammals. However, the absence of a postscapular fossa

suggests that this muscle was probably not as well developed in

Na. philcreaseri as it is in species of Notoryctes. The scapular

spine is markedly broadened over two-thirds of its length in

Notoryctes , towards the glenoid end, whereas only a third of

the scapular spine is broadened in Na. philcreaseri.

Furthermore, in Notoryctes, the scapular spine curves towards

the secondary scapular spine, forming a tube-like structure that

almost completely encloses the infraspinous fossa and

infraspinatus muscle (this feature is better developed in N.

typhlops than in N. caurinus; Warburton, 2006); no such tube

is present in Na. philcreaseri.

Humerus. 31 humeri referable to Na. philcreaseri are known

(see supplementary information). Their morphology is clearly

indicative of fossoriality, and they are very similar in size and

overall shape to those of Notoryctes species (see brief

description by Archer et al., 2010: appendix 1). Most have lost

their proximal and distal epiphyses. The specimen illustrated

here, QM F57719, has the epiphyses in place, but the epiphyseal

sutures remain unfused (figs. 2a, c).

The humeral head is large relative to the length of the

bone, roughly hemispherical in shape, and protrudes

posteromedially. The proximal tubercles are relatively low and

broad; the greater tubercle extends slightly higher than the

humeral head, while the lesser tubercle is slightly lower. The

greater tubercle bears a short proximal crest, and is more than

twice the breadth of the much smaller, ovoid lesser tubercle.

The bicipital (intertubercular) groove is moderately broad and

deep, and somewhat displaced medially due to the

disproportionate sizes of the tubercles. The humeral neck is

deeply constricted, particularly on its posterior aspect. The

proximal half of the humeral shaft is relatively robust, and is

marked by broad rugosities cranially and a broad sulcus

caudally. The deltopectoral crest is massive and protrudes

cranially and laterally from roughly the midpoint of the

humeral shaft. This crest is thick and robust, with numerous

rugosities, and its proximal end is distinctly concave in cranial

view. Distal to this, a crest that appears continuous with the

deltopectoral crest sweeps distally and medially, toward the

medial epicondyle and over the supratrochlear foramen (which

is proximal to the medial edge of the trochlea), giving the

impression that the distal and proximal halves of the humerus

are twisted relative to each other. Distally, the humerus

broadens rapidly, primarily due to massive enlargement of the

medial epicondyle; the maximum distal width of the humerus

is slightly less than half its length. The medial epicondyle

projects medially and is long and robust, with a bluntly

rounded terminus. The trochlea and capitulum are smoothly

contiguous, with only a slight medial constriction, and the

combined articular surface is broad and convex. The coronoid

gtu

d 8 j“

f i F hh

wTi

}

■

mep

tro cap

tro cap m mm

Figure 2. Comparison of humeri of Naraboryctes philcreaseri and Notoryctes typhlops : a, left humerus of Naraboryctes philcreaseri (QM

F57719) in cranial view; b, right humerus (reversed) of Notoryctes typhlops (SAM M637) in cranial view; c, QM F57719 in caudal view; d, SAM

M637 (reversed) in caudal view. Abbreviations: bg, bicipital groove; cap, capitulum; dpc, deltopectoral crest; gtu, greater tuberosity; hh, humeral

head; lsr, lateral supracondylar ridge; ltu, lesser tuberosity; mep, medial epicondyle; stf, supratrochlear foramen; tro, trochlea.

158

R.M.D. Beck, N.M. Warburton, M. Archer, S.J. Hand, and K.P. Aplin

fossa, proximal to the trochlea, is broad and shallow. The

lateral epicondyle extends only slightly beyond the lateral

border of the capitulum. The lateral supracondylar ridge is

strongly developed, and extends proximally for more than

one-third of the length of the humeral diaphysis, before

merging with the mid-posterior surface of the shaft.

Although very robust compared to the humeri of non-

fossorial mammals, the humerus of Na. philcreaseri is

proportionately somewhat more elongate and gracile than that

of Notoryctes (figs. 2b, d): maximum length and distal width

of the most complete adult humeri of Na. philcreaseri (QM

F23710, F57652 and F57719) are 15.1-17.2 mm and 7.9-83 mm

respectively, whereas in N. caurinus they are 11.9-12.7 mm

and 7.5-8.2 mm, and in A. typhlops they are 13.4-15.6 mm and

7.6-10.4 mm. As noted by Archer et al. (2011: appendix 1), the

humeral shaft is oval in cross-section in Na. philcreaseri,

rather than triangular as it is in Notoryctes, the proximal

surface of the deltopectoral crest is concave rather than

convex, and a supratrochlear foramen is present whereas this

foramen is absent in Notoryctes. In addition, the bicipital

groove is distinctly shallower than it is in Notoryctes, and the

deltopectoral crest, medial epicondyle and lateral

supracondylar ridge are all smaller and less strongly developed

in the fossil species. A further key difference is that in

Notoryctes the capitulum is relatively broader and more

cylindrical, and there is a distinct separation of the capitulum

and trochlea, whereas these form a single, continuous articular

surface in Na. philcreaseri.

Radius. The 11 radii attributed here to Na. philcreaseri (see

supplementary information) are very robust for their overall

size. Both the proximal and distal ends of the bone are

expanded, and are approximately equal width, although most

are missing the proximal and distal epiphyses; QM F57679,

however, is complete and is illustrated here (figs. 3a, c). The

fovea is ovoid and smoothly concave, and demarcated by a

well-marked border. Along the posterolateral aspect of the

shaft is an obvious groove; this groove is also present in

Notoryctes, where it houses a tendinous sheet (the interosseous

membrane) that binds the shafts of the radius and ulna together

(Warburton, 2006). The distal end of the radius (for the

radiocarpal articulation) is very broad, and possesses a short

styloid process. In its general size, shape and robustness, the

radius of Na. philcreaseri is very similar to that of Notoryctes

(figs. 3b, d), although this bone is slightly longer in the fossil

species (10.7 mm in QM F57679; 10.3 mm in QM F57680) than

in either A. caurinus (6.9-7.2 mm) or A. typhlops (7.4-93 mm).

Ulna. 30 ulnae referable to Na. philcreaseri are known (see

supplementary information). Together with the humerus, the

ulna of Na. philcreaseri arguably exhibits the most obvious

evidence of fossorial adaptations, as briefly discussed by

Archer et al. (2011: appendix 1). The ulnar shaft is robust and

relatively deep in the cranial-caudal plane, but narrow

mediodistally. The olecranon is long, robust and somewhat

curved medially; measured from the distal margin of the

trochlear notch (following Hopkins and Davis, 2009), it forms

more than 40% of the total length of the ulna, although the

exact percentage is uncertain as none of the ulnae of Na.

philcreaseri are fully intact; most are missing either the

proximal or distal epiphysis or both. The most complete

specimen, QM F57706, is illustrated here (figs. 4a, c e) and was

used to calculate IFA for Na. philcreaseri (table 1); however,

QM F57706 lacks the distal epiphysis, and hence the calculated

IFA is undoubtedly a slight overestimate (see below).

The trochlear (semilunar) notch is moderately deep, and its

distal (coronoid) surface is very steep in comparison to the

proximal (anconeal) surface. The trochlear notch is broad and

lies in somewhat oblique alignment, from proximomedial to

distolateral. The proximal anconeal process is long and extends

laterally. A crest extends proximally from the medial margin of

the proximal anconeal process, along the cranial aspect of the

olecranon. A crest also extends distally along the cranial aspect

of the ulnar shaft. The broad radial notch is on the lateral side

of the anconeal process, slightly distal to the articular surface

for the trochlea of the humerus. Medial and lateral coronoid

processes surround the radial notch in a slightly oblique

orientation. In medial view, a very prominent flexor sulcus

extends from the proximal ulnar shaft onto the olecranon.

The ulna of Na. philcreaseri resembles that of Notoryctes

(figs. 4b, d, f), and is similar in size; total preserved length of

QM F57706 is 17.9 mm (when intact, this bone may have been

0.5-1.0 mm longer), whereas the length of this bone is 13.4-

14.2 mm in A. caurinus and 15.4-18.7 mm in A. typhlops.

However, the ulnar shaft is proportionately longer and

mediolaterally narrower in Na. philcreaseri, and therefore

appears more gracile. The morphology of the ulnar articular

surfaces are very similar between the fossil and extant species,

particularly in the distinct distal displacement of the radial

articulation; however, the trochlear notch is not as broad and

Figure 3. Comparison of radii of Naraboryctes philcreaseri and

Notoryctes typhlops : a, left radius of Naraboryctes philcreaseri (QM

F57679) in cranial view; b, right radius (reversed) of Notoryctes

typhlops (SAM M637) in cranial view; c, QM F57679 in lateral view;

d, SAM M637 (reversed) in lateral view. Abbreviations: gr, groove for

interosseous membrane (tendinous sheet binding shafts of radius and

ulna together); sp, styloid process.

Fossil marsupial mole postcrania

159

trn

cop

Figure 4. Comparison of ulnae of Naraboryctes philcreaseri and Notoryctes typhlops: a, right ulna of Naraboryctes philcreaseri (QM F57706)

in cranial view; b, right ulna of Notoryctes typhlops (SAM M637) in cranial view; c, QM F57706 in medial view; d, SAM M637 in medial view;

e, QMF 57706 in lateral view; f, SAM M637 in lateral view. Abbreviations: anp, anconeal process; cop, coronoid process; fls, flexor sulcus; ol,

olecranon; rn, radial notch; sp, styloid process; trn, trochlear notch.

the radial notch not as deep in Na. philcreaseri. Interestingly,

in QM F57706 at least, the flexor sulcus on the medial aspect

of the ulna is deeper in Na. philcreaseri than in Notoryctes.

The olecranon is also less curved medially in the fossil species;

as a result, although the total length of the olecranon, measured

following its curvature, is clearly proportionately greater in

Notoryctes , the anteroposterior length of the olecranon,

measured parallel to the ulnar shaft, is proportionately similar

between the fossil and extant species.

Femur. The 17 femora referred here to Na. philcreaseri (see

supplementary information) are highly distinctive (figs. 5a, c),

but are broadly similar in size to femora of Notoryctes

species; based on the three most complete Na. philcreaseri

specimens (QM F57678, F57718, F57708), total length is

13.6-17.3 mm, compared to 12.4-14.1 mm in N. caurinus and

14.3-16.8 mm in N. typhlops. The femoral head is essentially

spherical in shape. The greater trochanter is broad and

laterally extended. In specimens in which the proximal

femoral epiphyses are intact (e.g. QM F57678), the greater

trochanter is very slightly lower than the femoral head. The

lesser trochanter is broad and rounded in outline and does not

extend proximally as far as the femoral head. In some

specimens (e.g. QM F57708), the lesser trochanter is deeply

excavated such that its cranial face forms a distinct ‘pocket’,

whereas in others (e.g. QM F57678) a somewhat shallower,

less well-defined depression is present. In caudal view, a

small, shallow trochanteric fossa is visible in a quite lateral

position, on the caudal face of the greater trochanter. The

femoral shaft is robust for its length and oval in cross-section.

The distal shaft flares towards the enlarged condylar region.

The intercondylar groove is relatively broad. A large,

anteroposteriorly-compressed process, the third trochanter,

protrudes laterally from the mid-shaft. At the level of the

third trochanter, the maximum transverse width of the femur

is more than one-fifth of femoral length.

The femur of Na. philcreaseri is substantially modified

from those of more generalised marsupials in its being

particularly robust, the strong development (relative to

absolute size) of the greater and lesser trochanters, and the

presence of a very large third trochanter. In comparison to

Notoryctes (figs. 5b, d), the femoral shaft of Na. philcreaseri

is not as robust with respect to femoral length, nor is the

greater trochanter as robust or broad. The lesser trochanter is

located in a more proximocaudal position in Notoryctes , such

that it is proximodistally level with the femoral head

(Warburton, 2006: fig. 13). Na philcreaseri retains a small

trochanteric fossa, whereas no such fossa is found in

Notoryctes. A particularly striking difference between Na.

philcreaseri and Notoryctes is that the latter lacks a distinct

third trochanter. However, the greater trochanter of Notoryctes

is very large and extends distally as a wing-like process along

160

R.M.D. Beck, N.M. Warburton, M. Archer, S.J. Hand, and K.P. Aplin

icg icg

Figure 5. Comparison of femora of Naraboryctes philcreaseri and Notoryctes typhlops : a, right femur of Naraboryctes philcreaseri (QM

F57678) in cranial view; b, right femur of Notoryctes typhlops (SAM M637) in cranial view; c, QM F57678 in caudal view; d, SAM M637 in

caudal view. Abbreviations: fh, femoral head; gtr, greater trochanter; icg, intercondylar groove; ltr, lesser trochanter; trf, trochanteric fossa; ttr,

third trochanter.

the lateral margin of the proximal quarter of the femur; we

suspect that this process originated via ancestral fusion of the

greater and third trochanters, with the distal end of this

process homologous with the third trochanter of Na.

philcreaseri , as we discuss in more detail below

(see Discussion).

Tibia. The five tibiae described here are referred to Na.

philcreaseri (see supplementary information) on the basis of

their size and unusual morphology (figs. 6a, c), which somewhat

resembles that observed in Notoryctes (figs. 6b, d). All known

specimens are missing the proximal and distal epiphyses, with

preserved lengths ranging from 10.2 to 12.7 mm. By

comparison, intact tibial length is 10.1-10.5 mm in A. caurinus

and 12.6-13.6 mm in N. typhlops. The most distinctive feature

of the Na. philcreaseri tibiae is the massively developed tibial

crest on the cranial aspect, which extends along the proximal

two-thirds of the length of the tibia and more than doubles its

craniocaudal depth. The development of the massive tibial

crest is strongly reminiscent of the strong development of the

same feature in Notoryctes. Unlike the flat, blade-like tibial

crest of Notoryctes, however, the crest in Na. philcreaseri is

curved in cross-section, such that it is slightly convex on its

medial surface and slightly concave on its lateral surface. In

addition, whereas the crest is most extensive cranially at its

proximal end (adjacent to the articulation with the femur) in

Notoryctes, in Na. philcreaseri it is most extensive more

distally, reaching its maximum extent about two-thirds along

its length from the proximal end of the tibia. The tibial crest of

Notoryctes is mediolaterally broad and concave at its proximal

end, for articulation with the distinctive enlarged patella. By

contrast, the crest in Na. philcreaseri is mediolaterally narrow,

and there is no sulcus that might have housed a patella. It is

therefore unclear whether or not a patella was present in Na.

philcreaseri. Absence of the proximal and distal epiphyses also

means that the precise morphology of the femoral and tarsal

articular surfaces is unknown; in particular, it is unclear

whether or not the peculiar, rounded process present

posterolaterally at the proximal end of the tibia in Notoryctes

(which articulates with the lateral condyle of the femur and also

with the fibula; Stirling, 1891: 179) was present in Na.

philcreaseri. However, proximally, it appears that the medial

condyle was probably slightly larger than the lateral condyle in

Na. philcreaseri, as in Notoryctes. The distal shaft is also short

and very robust, although slightly less so than in Notoryctes.

Fossil marsupial mole postcrania

161

Figure 6. Comparison of tibiae of Naraboryctes philcreaseri and

Notoryctes typhlops : a, left femur of Naraboryctes philcreaseri (QM

F57686) in medial view; b, left femur of Notoryctes typhlops (SAM

M637) in medial view; c, QM F57686 in lateral view; d, SAM M637 in

lateral view. Abbreviations: mma, medial malleolus; pltp,

posterolateral tibial process for articulation with lateral femoral

condyle and fibula; sup, sulcus for patella; tc, tibial crest.

Functional interpretation of the postcranial skeleton of

Naraboryctes philcreaseri

Forelimb. The scapula of Na. philcreaseri (fig. la) is elongate

and narrow in comparison to those of more generalised

mammals. In Notoryctes (fig. lb) and many other fossorial

and subterranean mammals, the scapula is similarly elongate

and narrow (Reed, 1951; Gasc et al., 1986; Warburton, 2006;

Salton and Sargis, 2008). In these forms, this morphology

reflects the reduction of dorsally-placed muscles between the

thoracic vertebral column and scapula (e.g. the m.

rhomboideus), which allows posterior rotation of the scapula.

It is also correlated with the more cranio-caudal alignment of

the long axis of the scapula (Reed, 1951; Puttick and Jarvis,

1977; Gasc et al., 1986; Warburton, 2006), which results in

enhanced mobility of the pectoral girdle along the cranial-

caudal axis (as demonstrated in Eremitalpa; Gasc et al.,

1986). As a result, the pectoral girdle is highly mobile,

facilitating protraction and retraction of an extended forelimb,

as is required for scratch-digging (Hildebrand, 1985; Nevo,

1999:71).

The humerus (figs. 2a, c), radius (figs. 3a, c) and ulna (figs. 4a,

c, e) of Na. philcreaseri are all very robust for their length, and

the articular surfaces of these bones are relatively large in surface

area. Short, robust limb bones are characteristic of semi-fossorial

and fossorial mammals (Casinos et al., 1993; Farina and

Vizcaino, 1997; Vizcaino et al., 1999; Campione and Evans,

2012), being better able to withstand the high muscular and

reaction forces acting through the bones during digging

(Hildebrand and Goslow, 2001; Elissamburu and Vizcaino,

2004). The relatively large surface area of humeral head,

capitulum and radial fovea of Na. philcreaseri would have acted

to improve the stability and structure of the joint by spreading

large mechanical forces over a greater area. Similarly enlarged

articular surfaces are found in the forelimb of many other

fossorially-adapted mammals, including chrysochlorids (Gasc et

al., 1986) and anteaters (Taylor, 1978).

The enlarged and strengthened muscle attachment sites of

the forelimb of Na. philcreaseri, notably the development of a

secondary spine on the scapula (fig. la), the enlarged deltopectoral

crest and medial epicondyle of the humerus (figs. 2a, c), and the

proportionately robust and elongate olecranon of the ulna (fig.

4a), are also all characteristic of a fossorial or subterranean

lifestyle (Reed, 1951; Lehmann, 1963; Taylor, 1978; Thorington

et al., 1997; Warburton, 2006; Warburton et al., 2013).

Collectively, they reflect enlargement of the m. triceps longus,

mm. pectorales, m. flexor carpi radialis, and m. flexor carpi

ulnaris and m. flexor digiti profundus, each of which contribute

during the power stroke of digging and thus require increased

power in order to act against the resistance of the substrate.

Besides muscle enlargement, power can also be increased

by migration of muscle attachment sites to improve mechanical

advantage: typically, fossorial mammals enhance mechanical

advantage by increasing the length of the in-lever relative to

the out-lever length (Lehmann, 1963; Nevo, 1979; Nevo, 1999;

Hildebrand and Goslow, 2001). The distally placed

deltopectoral crest of the humerus (figs. 2a, c) and greatly

enlarged olecranon of the ulna (fig. 4a) of Na. philcreaseri

would have acted to functionally increase the in-lever length

of their respective muscles by moving the insertions further

from the joint across which they act, while the relatively short

limb bones would have reduced the length of the out-lever. In

comparison to other fossorial mammals, the deltopectoral

insertion is more distally placed in Na. philcreaseri than in

digging armadillos (MacAlister, 1875a; Macalister, 1875b;

Burne, 1901; Hildebrand and Goslow, 2001) and fossorial

rodents (Lehmann, 1963), but less distally placed than the

homologous structure in some chrysochlorids (Gasc et al.,

1986). Similarly, the elongate medial epicondyle of Na.

philcreaseri (figs. 2a, c) reflects improved mechanical

advantage by providing a long lever arm of the forearm flexor

musculature (as well an enlarged area for muscle attachment),

for strong flexion of the wrist against the substrate during

digging. This is convergent with other scratch diggers,

particularly where the elbow is somewhat extended during

limb retraction in the power-stroke of digging (Gasc et al.,

1986; Warburton, 2006). It is also characteristic of fossorial

mammals that utilise a partially abducted (half-sprawling) or

sprawling posture of the carpus during digging (Gambaryan

and Kielan-Jaworowska, 1997).

Turning now to the ulna, IFA for Na. philcreaseri is 88.4,

which is slightly higher than that for Notoryctes typhlops

(87.2; see table 1). However, the specimen used to calculate

IFA in Na. philcreaseri, QM F57706, is missing the distal

epiphysis (see fig. 4a, c, e) and so this value is undoubtedly a

slight overestimate; the true value for Na. philcreaseri is

probably the same as, or slightly less than, that for A. typhlops.

It should also be noted that IFA is based on the anteroposterior

length of the olecranon; it does not take into account the

degree of medial curvature of the olecranon, which is greater

in Notoryctes species (fig. 4b; Warburton, 2006: fig. 9) than in

Na. philcreaseri (fig. 4a). The IFA values for Na philcreaseri

and N. typhlops are some of the highest seen in mammals

162

R.M.D. Beck, N.M. Warburton, M. Archer, S.J. Hand, and K.P. Aplin

(table 1); of the taxa measured by Hopkins and Davis (2009),

only the chrysochlorid Amblysomus hottentotus (which is

subterranean), the dasypodids Priodontes maximus and

Cabassous centralis (which are both fossorial), and the fossil

epoicotheriid palaeanodonts Xenocranium pileorivale and

Pentapassalus pearcei exhibit higher values. Other

subterranean mammals, such as talpids and spalacid rodents

are characterised by lower IFA values (table 1).

along the proximal third of the length of the bone (figs. 5b, d).

We suggest that this proximally-extended lateral crest in

Notoryctes is the result of ancestral fusion of the third

trochanter seen in Na. philcreaseri with the greater trochanter,

and that the distal end of this crest in Notoryctes is homologous

with the third trochanter of Na. philcreaseri.

Discriminant analysis

Hind limb. The femur of Na. philcreaseri is robust, and the

greater and lesser trochanters are very large for a bone belonging

to a mammal of this size (figs. 5a, c). The lateral expansion of the

proximal femur compared to more generalised marsupials seen

in Na. philcreaseri is largely due to enlargement of the greater

trochanter. Lateral extension of the proximal femur is even more

extreme in Notoryctes, due to its very wide greater trochanter

(figs. 5b, d). Although not as large as in Notoryctes, the

transversely wide greater trochanter of Na. philcreaseri would

have improved the mechanical advantage of the gluteal muscles

for limb abduction by increasing the length of the in-lever; this

suggests a strongly abducted posture of the hind limbs, most

likely to brace the hind part of the body during digging, but

potentially also during the propulsive phase of locomotion.

The prominent third trochanter protruding laterally from

the midshaft of the femur of Na. philcreaseri is highly

distinctive (figs. 5a, c). A much smaller, more proximally-

positioned third trochanter occurs in caenolestids, vombatids

and some stem-metatherians (Szalay and Sargis, 2001;

Horovitz et al., 2008; Abello and Candela, 2010). However, no

such distinct process is observed in Notoryctes (figs. 5b, d) or

most other marsupials. Among placental mammals at least,

the third trochanter (where present) is for attachment of the

gluteal muscles. The gluteal musculature of Notoryctes is

modified compared with that of other marsupials, due to the

highly modified morphology of the pelvis and the expanded

greater trochanter (Warburton, 2006). While there is no

indication in the living species of an insertion by the gluteal

musculature on the mid-lateral femur, the morphology of the

femur of Naraboryctes may represent an adaptation for

increasing the leverage of the gluteal muscles for limb

abduction. If so, this may indicate that the hindlimbs were

strongly abducted in Na. philcreaseri (at least during

burrowing), which would be a plausible intermediate stage

between the parasagittal stance of generalised terrestrial

mammals and the highly specialised stance of Notoryctes,

where the legs appear to project laterally from the body and no

longer play a role in supporting the body off the substrate. In

addition, among extant and fossil xenarthrans, a prominent

third trochanter has been proposed as a mechanism to mitigate,

by muscular action, bending forces acting through the femur

(Milne and O’Higgins, 2012). The hypothesised shift to a

strongly abducted hind limb position in Na. philcreaseri,

potentially for bracing or active excavation of substrate, would

tend to increase bending moments through the femur; if so,

this may have favoured the evolution of enlarged gluteal

musculature to mitigate these forces, with concomitant

development of a large third trochanter. In Notoryctes, there is

a well-developed crest extending from the greater trochanter

Compared to Hopkins and Davis’ (2009) original “limbs only”

dataset, our quadratic discriminant analysis based on a slightly

expanded and modified dataset showed slightly better accuracy

for discriminating burrowing from non-burrowing mammals

(13.2% misidentified, versus 13.9% in Hopkins and Davis,

2009), but was somewhat less accurate at determining the

degree of fossoriality, i.e. distinguishing between non¬

burrowing, fossorial and subterranean mammals (18.2%

misidentified, versus 14.8% in Hopkins and Davis, 2009). In

the “burrowing” versus “non-burrowing” analysis, Na.

philcreaseri was identified as burrowing with very high

probability (p = 0.99), and in the “non-burrowing” versus

“fossorial” versus “subterranean” analysis it was identified as

subterranean, again with very high probability (p =1.00). A

plot of the first two canonical axes from the non-burrowing

versus fossorial versus subterranean analysis indicates that the

first axis encompasses 91.2% of the total variability, with

subterranean taxa exhibiting the lowest values and non¬

burrowing taxa the highest values (fig. 7). Na. philcreaseri

exhibits a very low value for the first canonical axis, even

lower than for A. typhlops, and plots closest to the subterranean

chrysochlorid Eremitalpa grand (fig. 7).

-6

-7-

§ -9

°- 10 -

-11

■remitalpa

grant! %

V v- - -

Iarabofyaes * *

phllcreastri ■

0 2

Canonical axis 1

Figure 7. Plot of first two canonical axes from quadratic discriminant

analysis of degree of fossoriality (non-burrowing versus fossorial

versus subterranean) in mammals based on an expanded version of the

“limbs only” dataset of Hopkins and Davis (2009). Non-burrowing

species are represented by green circles, fossorial species by pink

squares and subterranean species by blue asterisks. Inner ellipses

represent 95% confidence intervals for the means for each class, whilst

the outer ellipses represent the 50% prediction intervals. Naraboryctes

philcreaseri is represented by a black triangle and was treated as

unknown, but falls among subterranean species and is predicted to be

subterranean with very high probability (p = 1.0). Abbreviations: F,

fossorial; N, non-burrowing; S, subterranean.

Fossil marsupial mole postcrania

163

Phytogeny and divergence times

The dated phylogeny that results from Bayesian analysis of our

total matrix using the IGR clock model is given in Figure 8.

The topology is broadly congruent with that found in analyses

of a similar version of the matrix by Beck et al. (2014). As

expected, Na. philcreaseri is recovered as sister-taxon to

Notoryctes, with strong support (BPP=0.91). Notoryctemorphia

(i.e. Naraboryctes + Notoryctes) is sister to Peramelemorphia

(i.e. Echymipera + Perameles ), with Dasyuromorphia (i.e.

Dasyuroides + Dasyurus + Phascogale) outside this, but

support values for these two nodes are very low (BPP <0.5).

There is somewhat stronger support (BPP = 0.71) for the clade

comprising Notoryctemorphia, Peramelemorphia,

Dasyuromorphia and Yalkaparidon, which corresponds to

Agreodontia sensu Beck et al. (2014).

In terms of divergence times, the base of Agreodontia is

estimated to be 55.2 MYA (95% HPD = 48.0-63.9 MYA), with

Dasyuromorphia diverging at 54.4 MYA (95% HPD = 46.8-

63.9 MYA), and the Notoryctemorphia-Peramelemorphia split

estimated at 52.2 MYA (95% HPD = 43.6-60.7 MYA). The

point estimate for the split between Notoryctes and

Naraboryctes is 30.3 MYA, which is considerably older than

the age of Naraboryctes inferred here (19.4 Ma old; 95% HPD

= 16.0-22.6 Ma old). However, the 95% HPD for this split is

wide (17.7-46.3 MYA) and overlaps the inferred age of Na.

philcreaseri. Furthermore, the 95% HPD for the effective

branch length leading to Naraboryctes (i.e. the estimated

change relative to that inferred for the last common ancestor of

Naraboryctes and Notoryctes) is 0.0000-0.0733 substitutions/

site (= changes/character) for the morphological partition (see

supplementary information); this encompasses the possibility

that Na.philcreaseri exhibits no morphological apomorphies

relative to the Naraboryctes-Notoryctes common ancestor.

Thus, when 95% HPDs are taken into account, the age and

morphology of Na. philcreaseri are both compatible with its

being ancestral to Notoryctes (M.S.Y. Lee, pers. comm.).

RBP3 inactivation

Specifying separate values of co for the RBP3 alignment for

Notoryctes (the “foreground” branch, ( 0 f ) and for all other

branches (the “background” branches, co b ) did not result in a

significantly better fit than the null model of a single value of

to for all branches (p »0.05). However, entering the values

for G) f (0.244) and to b (0.15635) into the equation of Zhang et

al. (2008), and assuming that Notoryctes diverged from

(Echymipera + Perameles) at 52.2 MYA (see above) gives an

estimate of 5.4 MYA for the timing of inactivation of RBP3 in

the Notoryctes lineage, with a range of 4.5-6.3 MYA if the

Time before present (Ma)

Figure 8. Dated total evidence phylogeny based on 259 morphological characters and 9012 bp of sequence data from five nuclear genes ( APOB ,

BRCA1, RBP3, RAG1, and VWF) analysed using MrBayes 3.2.2 assuming the Independent Gamma Rates (IGR) clock model, with topological

and temporal constraints applied to selected internal nodes (see supplementary information). Notoryctes and Naraboryctes are indicated in bold.

Blue bars represent 95% highest posterior density intervals (HPDs) on the divergence times. Nodes without Bayesian posterior probability (BPP)

were constrained a priori.

164

R.M.D. Beck, N.M. Warburton, M. Archer, S.J. Hand, and K.P. Aplin

95% HPD for the time of divergence is taken into account (fig.

9; see above). This corresponds to the latest Miocene-earliest

Pliocene, and considerably postdates the age of Na.

philcreaseri (fig. 9), suggesting that RBP3 was functional in

this fossil taxon. Interestingly, it is similar to Mitchell et al.’s

(2014) relaxed molecular clock estimate of 4.5 MYA for the

divergence between the living species Notoryctes typhlops

and A. caurinus.

Discussion

Qualitative functional analysis of the postcranial anatomy of

Na. philcreaseri indicates that it was well-adapted for scratch¬

digging. Of particular significance are the narrow, elongate

scapula with secondary scapular spine, very robust humerus

with enormously enlarged deltopectoral crest and medial

epicondyle, and greatly enlarged and medially-inflected

olecranon of the ulna. All of these are characteristic of extant

scratch-diggers, and of fossil mammals that have been

interpreted to exhibit similar burrowing behaviour, such as the

early Miocene meridiolestidan Necrolestes and the late Eocene

epocoitheriid palaeanodonts Epoicotherium and Xenocranium

(Turnbull and Reed, 1967; Rose and Emry, 1983). This

interpretation is bolstered by the results of our quadratic

discriminant analysis, in which Na. philcreaseri is predicted

Time before present (Ma)

Figure 9. Part of the dated total evidence phylogeny shown in Figure

8, restricted to the clade Agreodontia (which includes

Notoryctemorphia), with divergence dates compared to global

temperatures and environmental change in Australia. The estimated

time of inactivation of the RBP3 gene in the Notoryctes lineage is

indicated: the black bar represents the point estimate (5.4 MYA),

whilst the grey bars represent 95% HPDs (4.5-6.3 MYA). The global

temperature curve is modified from Zachos et al. (2001). The date for

the major increase in grass pollen is taken from Martin and McMinn

(1994: fig. 2) whilst the date for the onset of major aridity in Australia

(~1.4-1.5 MYA) is taken from McLaren and Wallace (2010).

to be subterranean based on its limb proportions, and lies

closest to the subterranean chrysochlorid Eremitalpa granti in

our plot of the first two canonical axes. In addition, although

the IFA value calculated for Na. philcreaseri (88.4) is

undoubtedly a slight overestimate, it is almost identical to that

for N. typhlops (87.2), and is greater than that for many fully

subterranean mammals (table 1). In summary, all available

morphological evidence supports the conclusion that Na.

philcreaseri was a specialised, most probably subterranean,

scratch-digger.

These results indicate that specialised fossorial, probably

subterranean, behaviour originated in the notoryctemorphian

lineage some time before the early Miocene. The Faunal Zone

B deposits at Riversleigh from which all known specimens of

Na. philcreaseri have been recovered appear to represent

rainforest environments (Travouillon et al., 2009; Archer et

al., 2011; Bates et al., 2014), and forests likely predominated

throughout Australia until the middle-late Miocene (Martin,

2006). This may be an indication that the burrowing

specialisations of fossil notoryctemorphians were well-suited

to, and potentially evolved in, such closed forest environments.

If so, the golden moles Chrysospalax trevelyani and Huetia

leucorhina (which forage in or above leaf litter as well as

digging burrows in the rainforests of sub-Saharan Africa;

Bronner, 2013) and possibly other fossorially-adapted

mammals that live in rainforest environments (e.g. the tenrec

Oryzorictes hova of Madagascar, and the peramelemorphian

Microperoryctes murina of New Guinea) may represent

reasonable living analogues for early notoryctemorphians.

However, rainforest soils are typically thin and do not appear

favourable for the evolution of a truly subterranean lifestyle.

Furthermore, Riversleigh appears to have been a karst

environment during the early Miocene (Arena, 2004; Arena et

al., 2014), and karst (including karst in rainforest)

typically has exposed rocky surfaces only partially covered by

thin soils (although deep, dirt-filled fissures may be present;

D.A. Arena, pers. comm.). We therefore propose two further

scenarios for the origin of specialised fossorial behaviour

in notoryctemorphians.

It is interesting that all specimens of Na. philcreaseri

collected to date are from sites at Riversleigh that are

interpreted as representing cave deposits (Arena, 2004; Arena

et al., 2014). An alternative hypothesis is therefore that

notoryctemorphians acquired fossorial behaviours foraging

for invertebrates within cave environments, burrowing through

guano-rich cave floors. However, we note that no modern

mammal or other terrestrial vertebrate of which we are aware

occupies such a niche, nor is their compelling evidence for

occupation of this niche by any other fossil vertebrate species.

Furthermore, all Riversleigh sites representing the early

Miocene Faunal Zone B appear to be cave deposits (D. A.

Arena, pers. comm.). Thus, the presence of Na. philcreaseri

fossils at these sites may simply be because notoryctemorphians

only occurred in the Riversleigh area during the early Miocene,

and are present in cave deposits because those are the only

deposits preserved from this time period, not because they

were cave-dwelling. Several other Riversleigh mammal

species are known only from Riversleigh Faunal Zone B (and

Fossil marsupial mole postcrania

165

hence are found only in cave deposits), including the thylacinids

Wabulacinus ridei and Ngamalacinus timmulvaneyi (see

Muirhead, 1997) and species of the miralinid possum

Durudawiri (e.g. Crosby and Archer, 2000; Crosby, 2002),

none of which are likely to have been cave-dwellers. However,

remains of these species are admittedly much rarer than those

of Na. philcreaseri. On the other hand, Rzebik-Kowalska

(2013) reported very large numbers of individuals of the talpid

“true moles” Talpa europaea and T. minor in early Pleistocene

deposits at Zabia Cave in Poland, and suggested that the cave

acted as a natural trap for moles due to their burrowing

activities. This might explain the abundant remains of Na.

philcreaseri in Faunal Zone B cave deposits, without needing

to assume that it was a specialised cave dweller.

A third hypothesis is that early notoryctemorphians

were semi-aquatic, and that their semi-aquatic adaptations

subsequently became exapted for fossoriality. Several

species of talpid moles are semi-aquatic, and some authors

have proposed that modern talpids all descend from a semi-

aquatic ancestor (Campbell, 1939; Whidden, 1999; but see

Reed 1951; Hickman 1984). Similarly, the fossorial

adaptations of the living platypus (which is semi-aquatic)

and echidnas (which are terrestrial) may represent

exaptations from a semi-aquatic monotreme ancestor

(Phillips et al., 2009; Phillips et al., 2010; Mirceta et al.,

2013; but see Camens 2010; Ashwell 2013; Musser 2013).

However, there is only a single extant semi-aquatic

marsupial, the yapok or water-opposum Chironectes

minimus (see Marshall, 1975; Stein and Patton, 2008), and

only one questionably semi-aquatic stem-metatherian,

Didelphodon vorax (Szalay, 1994; Longrich, 2004; but see

Fox and Naylor, 2006). This suggests that marsupials are, in

general, poorly adapted to semi-aquatic or aquatic niches,

probably due to their reproductive mode (Lillegraven, 1975;

Phillips et al., 2009). We consider it unlikely, based on

available evidence, that Na. philcreaseri was an exception to

this general rule. However, this hypothesis can be tested:

Mirceta et al. (2013) showed that mammals with some

degree of aquatic ancestry show elevations in myoglobin

net surface charge, which is correlated with maximal

skeletal muscle concentration of myoglobin, which in turn

is linked with maximal active dive time underwater.

Mirceta et al.’s (2013) study suggests that scalopine talpids

(members of other talpid subfamilies were not sampled) and

modern monotremes derive from at least semi-aquatic

ancestors, but that chrysochlorids apparently do not. Using

this approach, sequencing of the myoglobin locus for one or

both living Notoryctes species should reveal whether or not

they have semi-aquatic ancestry. Histological analysis of

humeral microanatomy, specifically bone compactness,

may also indicate whether or not Na. philcreaseri was semi-

aquatic (Canoville and Laurin, 2010). However, a recent

study of bone compactness in the humeri of talpids found

no significant difference between fossorial and semi-

aquatic taxa (Meier et al., 2013).

The point estimates for the age of Na. philcreaseri (19.4

Ma) and the timing of the Naraboryctes-Notoryctes split (30.3

Ma) imply that Na. philcreaseri post-dates the divergence

between the Naraboryctes and Notoryctes lineages. However,

when 95% HPDs are taken into account, these estimates

overlap. In addition, the 95% HPD for the effective branch

length leading to Naraboryctes encompasses a zero length

branch. Collectively, this means that the age and known

morphology of Na. philcreaseri is compatible with its being

ancestral to Notoryctes. Furthermore, the postcranial

differences observed between Na. philcreaseri and the extant

Notoryctes species (N. typhlops and N. caurinus ) appear to

represent further developments of morphological trends that

are apparent when comparing Na. philcreaseri to more

generalised marsupials. They can plausibly be interpreted as

reflecting more extreme specialisation towards burrowing

in Notoryctes.

Specifically, derivation of the postcranial morphology

seen in the extant Notoryctes species, N. typhlops and N.

caurinus, from a Naraboryctes- like ancestor would require

the following major changes. In the scapula: lengthening of

the “vertebral” border and modification of the coracoid and

axillary angles into hook-like processes for attachment of the

subscapularis and teres major muscles; lengthening of the

secondary scapular spine towards the “caudal” angle; and

enlargement of the scapular spine, which curves medially

towards the secondary scapular spine to enclose the

infraspinatus muscle within a nearly complete tube;

development of a postscapular fossa, reflecting enlargement of

the triceps muscle group. In the humerus: further enlargement

of the deltopectoral crest; further expansion of the medial

epicondyle, increasing the surface area of origin for the

forearm flexor musculature; enlargement of the lateral

supracondylar ridge, increasing the surface area of origin for

the forearm extensors. In the ulna: further elongation of the

olecranon and development of a more marked medial

curvature; broadening of the trochlear notch; deepening of the

radial notch. In the radius: increased concavity of the articular

surfaces for the humerus and ulna at the proximal end;

broadening of the distal end. In the femur: further enlargement

of the greater trochanter and the head of the femur; loss of the

third trochanter, probably by proximal migration and fusion

with the greater trochanter; a slight proximocaudal shift in the

position of the lesser trochanter, such that it is proximodistally

level with and caudal to the femoral head; increased robusticity

of the femoral shaft. In the tibia: increased robusticity of the

tibial shaft and further enlargement of the tibial crest,

particularly at its proximal end, with development of a

proximal sulcus that houses an enlarged patella.

As discussed by Archer et al. (2011), the apomorphic

dental morphology of extant Notoryctes species can also be

plausibly derived from that seen in Na. philcreaseri, via loss of

a premolar locus (PI is tiny in Na. philcreaseri ), complete

suppression of the paracone on the upper molars (this cusp is

greatly reduced and shifted labially in Na. philcreaseri), and

loss of the talonid on the lower molars (the talonid is present

but narrow in Na. philcreaseri ). Thus, the known dental and

postcranial morphology of Na. philcreaseri renders it a

plausible model for a plesiomorphic ancestor of Notoryctes.

If our estimate for the timing of inactivation of RBP3 in

the Notoryctes lineage, 5.4 MYA (95% HPD = 4.5-6.3 MYA;

166

latest Miocene to earliest Pliocene), is correct, then RBP3 was

presumably functional in the early Miocene Na. philcreaseri.

In turn, this might be an indication that Na. philcreaseri still

retained some visual capabilities. If so, we might predict that

orbit will be more obviously identifiable on the cranium of Na.

philcreaseri than in Notoryctes, in which osteological evidence

for the orbit is minimal (Stirling, 1891; Carlsson, 1904;

Johnson and Walton, 1989); for example, the frontal process of

the jugal (for attachment of the postorbital ligament) might be

more strongly developed. However, testing of this hypothesis

will require the discovery of additional, more complete cranial

material of Na. philcreaseri. Archer et al. (2011) noted that the

jaw and dentition of Na. philcreaseri seem adapted for

somewhat harder food items than those of modern Notoryctes

species which may be an indication that the fossil taxon

foraged at least some of the time on the surface (particularly

soft-bodied invertebrates such as insect larvae and worms

being more common below ground). Based on this combined

genetic and morphological evidence (and pending the

discovery of more complete cranial remains that would reveal

the degree of development of its orbit), we suggest that Na.

philcreaseri probably spent a greater proportion of its time

above ground than does Notoryctes today, which surfaces only

rarely (Johnson and Walton, 1989; Benshemesh and Johnson,

2003; Dennis, 2004; Benshemesh, 2008; Benshemesh and

Aplin, 2008).

The late Miocene saw the replacement of rainforest with

more open woodland environments across much of Australia,

in response to a general global drying and cooling trend

(Martin, 2006; Black et al., 2012). This coincided with a

period of extensive faunal turnover among Australian

terrestrial vertebrates (Black et al., 2012), probably including

major radiations of dasyurids (Krajewski et al., 2000) and

macropodids (Prideaux and Warburton, 2010).This is followed

by palaeobotanical evidence for a sudden increase in the

abundance of grasses in the latest Miocene (-6 MYA), from

1-2% to -35% of the total pollen count (Martin and McMinn,

1994: fig. 2). However, true grasslands did not become

widespread in Australia until the middle Pliocene (Martin,

2006; Stromberg, 2011; Black et al., 2012). If our value for the

timing of RBP3 inactivation in Notoryctes of 5.4 MYA (95%

HPD = 4.5-6.3 MYA) is correct, it is tempting to postulate

that loss of RBP3 function is causally linked to the development

of more open environments - probably open sclerophyllous

woodland with a grassy understory (Stromberg, 2011), rather

than true grasslands - at this time. This reduction in vegetation

cover would have led to strong selection pressure on the

Notoryctes ancestor to spend minimal time on the surface

(where it would be particularly vulnerable to predators),

leading to relaxation of selection on RBP3.

Our estimate for the timing of RBP3 inactivation in the

Notoryctes lineage is markedly more recent than that of

Springer et al. (1998), which was 18.5-24.7 MYA, and that of

Emerling and Springer (2014), which was 12.23 MYA (range

of 6.87-17.84 MYA). There are a number of potential

explanations for this. Springer et al.’s (1998) estimate was

based on the presence of three or four indels in the RBP3

sequence of N. typhlops, and the assumption that indels occur

R.M.D. Beck, N.M. Warburton, M. Archer, S.J. Hand, and K.P. Aplin

in pseudogenes at a rate of 0.17/kb/Ma. This value is a weighted

average calculated from four nuclear pseudogenes (none of

them RBP3 ) for seven primates, five of which (Homo, Pan,

Gorilla, Pongo,and Hylobates ) were hominoids (Saitou and

Ueda, 1994). Rates varied between 0.14 and 0.24/kb/Ma across

the four genes examined by Saitou and Ueda (1994), but it is

unclear how much additional rate variation would be observed

if more genes were to be examined. Furthermore, molecular

rates of evolution vary considerably within mammals

(Bininda-Emonds, 2007; Welch et al., 2008), and rates are

particularly slow in hominoids relative to most other mammals,

including marsupials (Bininda-Emonds, 2007). Thus, the

indel rate assumed by Springer et al. (1998) may be an

underestimate, in which case their estimate for the timing of

RBP3 inactivation is too old.

The approach of Emerling and Springer (2014) is

conceptually similar to that used here, namely using estimated

values of to and divergence times to infer the timing of gene

inactivation (see Meredith et al., 2009a). However, their

estimate for co b (0.1007-0.1385) was lower than that calculated

here (0.15635), and their estimate for co f (0.2928-0.3355) was

higher than ours (0.244). These discrepancies may be due to

differences in taxon sampling: Emerling and Springer’s

(2014) RBP3 sequence alignment included 13 placentals but

only five marsupials, whereas our alignment comprises 23

marsupials. In addition, Emerling and Springer (2014) used

empirical estimates for oo after inactivation of RBP3 (0.9071-

1.1489) based on bat sequence data, whereas our method

assumes a priori that to after gene inactivation is 1, and they

assumed a somewhat older estimate for the divergence of

Notoryctes from other extant marsupials (61.28 MYA;

Meredith et al., 2011) versus that used here (52.2 MYA; 95%

HPD: 43.6-60.7 MYA).

Conclusion

Our qualitative and quantitative analysis of the postcranial

morphology of Na. philcreaseri indicate that it was most likely

subterranean, albeit somewhat less specialised in this regard

than the living species of Notoryctes ; the fact that RBP3 was

probably functional in Na. philcreaseri might be an indication

that vision was more important to the fossil taxon than to

Notoryctes, which might be evidence that it spent more time

on the surface than the two living marsupial mole species.

Nevertheless, it appears that notoryctemorphians had already

evolved specialised, probably fully subterranean, burrowing

behaviour prior to the early Miocene. A single upper molar,

(NTM P2815-6) from the late Oligocene Pwerte Marnte

Marnte Local Fauna in the Northern Territory, probably

represents a notoryctemorphian, albeit more plesiomorphic

than Na. philcreaseri based on its larger paracone and

metaconule (Murray and Megirian, 2006: 151; Beck et al.,

2014). This is the only other fossil occurrence of

Notoryctemorphia currently known. A fuller understanding of

the evolution of notoryctemorphians, including when and why

they evolved “mole-like” behaviour, will therefore require

marked improvements in the Palaeogene fossil record of

mammals in Australia.

Fossil marsupial mole postcrania

167

Acknowledgements

We thank Laura Wilson and Rick Arena for discussion,

Sharon Jansa for help with PAML, Hayley Bates for access

to JMP, and Mike Lee for assistance and advice with our

Bayesian phylogenetic analyses. Kenny Travouillon

provided helpful information regarding the taxonomic

composition of different Riversleigh sites. Financial support

for this research has been provided by the National Science

Foundation (via grant DEB-0743039, in collaboration with

Robert Voss at the American Museum of Natural Flistory)

and the Australian Research Council (via Discovery Early

Career Researcher Award DE120100957). Additional

support for research at Riversleigh has come from the

Australian Research Council (LP0989969, LP100200486,

and DP130100197 grants to MA and SJH), the XSTRATA

Community Partnership Program (North Queensland), the

University of New South Wales, Phil Creaser and the

CREATE Fund, the Queensland Parks and Wildlife Service,

Environment Australia, the Queensland Museum, the

Riversleigh Society Inc., Outback at Isa, Mount Isa City

Council, and private supporters including K. and M. Pettit,

E. Clark, M. Beavis, and M. Dickson. Many volunteers, staff

and students of the University of New South Wales have also

assisted in field work at Riversleigh. We thank Sandrine

Ladeveze and the editor Erich Fitzgerald for their helpful

and constructive comments that greatly improved the

final draft.

References

Abello, M. A., and Candela, A. M. (2010). Postcranial skeleton of the

Miocene marsupial Palaeothentes (Paucituberculata,

Palaeothentidae): paleobiology and phylogeny. Journal of

Vertebrate Paleontology 30: 1515-1527.

Aplin, K., and Archer, M. (1987). Recent advances in marsupial

systematics with a new syncretic classification. In M. Archer

(Ed.), Possums and opossums: studies in evolution, (pp. xv-

lxxii). Sydney: Surrey Beatty and Sons and the Royal Zoological

Society of New South Wales.

Archer, M. (1984). The Australian marsupial radiation. In M. Archer,

and G. Clayton (Eds.), Vertebrate zoogeography and evolution in

Australasia (pp. 633-808). Perth: Hesperian Press.

Archer, M., Arena, D. A., Bassarova, M., Beck, R. M. D., Black, K.,

Boles, W. E., et al. (2006). Current status of species-level

representation in faunas from selected fossil localities in the

Riversleigh World Heritage Area, northwestern Queensland.

Alcheringa 31: 1-17.

Archer, M., Beck, R. M. D., Gott, M., Hand, S., Godthelp, H., and

Black, K. (2011). Australia’s first fossil marsupial mole

(Notoryctemorphia) resolves controversies about their evolution

and palaeoenvironmental origins. Proceedings of the Royal

Society B: Biological Sciences 278: 1498-1506.

Archer, M., and Hand, S. (1984). Background to the search for

Australia’s oldest mammals. In M. Archer, and G. Clayton (Eds.),

Vertebrate zoogeography and evolution in Australasia. (Animals

in space and time) (pp. 517-565). Carlisle, Western Australia:

Hesperian Press.

Archer, M., Hand, S. J., and Godthelp, H. (1994). Riversleigh: the

story of animals in ancient rainforests of inland Australia.

Sydney: Reed Books.

Archer, M., Hand, S. J., Godthelp, H., and Creaser, P. (1997).

Correlation of the Cainozoic sediments of the Riversleigh World

Heritage fossil property, Queensland, Australia. In J.-P. Aguilar,

S. Legendre, and J. Michauz (Eds.), Actes du Congres

BiochroM’97 (pp. 131-152). Montpellier: Ecole Pratique des

Hautes Etudes, Institut de Montpellier.

Arena, D. (2004). The geological history and development of the

terrain at the Riversleigh World Heritage Area during the middle

Tertiary. Ph.D. dissertation. University of New South Wales,

School of Biological, Earth and Environmental Sciences, Sydney.

Arena, D. A., Black, K. H., Archer, M., Hand, S. J., Godthelp, H., and

Creaser, P. (2014). Reconstructing a Miocene pitfall trap:

recognition and interpretation of fossiliferous Cenozoic

palaeokarst. Sedimentary Geology 304: 28-43.

Asher, R. J., Horovitz, I., and Sanchez-Villagra, M. R. (2004). First

combined cladistic analysis of marsupial mammal

interrelationships. Molecular Phylogenetics and Evolution 33:

240-250.

Ashwell, K. W. S. (2013). Reflections: monotreme neurobiology in

context. In K. Ashwell (Ed.), Neurobiology ofmonotremes: brain

evolution in our distant mammalian cousins (pp. 285-298).

Collingwood, Australia: CSIRO Publishing.

Barbour, R. A. (1963). The musculature and limb plexuses of

Trichosurus vulpecula. Australian Journal of Zoology 11: 488-

610.

Bates, H., Travouillon, K. J., Cooke, B., Beck, R. M. D., Hand, S. J.,

and Archer, M. (2014). Three new Miocene species of musky rat-

kangaroos ( Hypsiprymnodontidae, Macropodoidea): description,

phylogenetics, and paleoecology. Journal of Vertebrate

Paleontology 34: 383-396.

Beck, R. M. D. (2008). A dated phylogeny of marsupials using a

molecular supermatrix and multiple fossil constraints. Journal of

Mammalogy 89: 175-189.

Beck, R. M. D., Godthelp, H., Weisbecker, V., Archer, M., and Hand,

S. J. (2008). Australia’s oldest marsupial fossils and their

biogeographical implications. PLoS ONE 3: el858.

Beck, R. M. D., and Lee, M. S. Y. (2014). Ancient dates or accelerated

rates? Morphological clocks and the antiquity of placental

mammals. Proceedings of the Royal Society B: Biological

Sciences 281: 20141278.

Beck, R. M. D., Travouillon, K. J., Aplin, K. P, Godthelp, H., and

Archer, M. (2014). The osteology and systematics of the enigmatic

Australian Oligo-Miocene metatherian Yalkaparidon

(Yalkaparidontidae; Yalkaparidontia; ?Australidelphia;

Marsupialia). Journal of Mammalian Evolution 21: 127-172.

Benshemesh, J. (2008). Itjaritjari, Notorytes typhlops. In S. Van Dyck,

and R. Strahan (Eds.), The Mammals of Australia, 3rd edition

(pp. 412-413). Sydney: Reed New Holland.

Benshemesh, J., and Aplin, K. P. (2008). Kakarratul, Notoryctes

caurinus. In S. Van Dyck, and R. Strahan (Eds.), The Mammals of

Australia, 3rd edition (pp. 410-411). Sydney: Reed New Holland.

Benshemesh, J., and Johnson, K. (2003). Biology and conservation of

marsupial moles ( Notoryctes ). In M. Jones, C. R. Dickman, and

M. Archer (Eds.), Predators with pouches: the biology of

carnivorous marsupials (pp. 464-474). Melbourne: CSIRO

Publshing.

Bi, S., Wang, Y., Guan, J., Sheng, X., and Meng, J. (2014). Three new

Jurassic euharamiyidan species reinforce early divergence of

mammals. Nature 514: 579-584.

Bininda-Emonds, O. R. P. (2007). Fast genes and slow clades:

comparative rates of molecular evolution in mammals.

Evolutionary Bioinformatics 3: 59-85.

168

Black, K. H., Archer, M„ Hand, S. J., and Godthelp, H. (2012). The

rise of Australian marsupials: a synopsis of biostratigraphic,

phylogenetic, palaeoecologic and palaeobiogeographic

understanding. In J. A. Talent (Ed.), Earth and Life: Global

Biodiversity, Extinction Intervals and Biogeographic

Perturbations Through Time (pp. 983-1078). Dordrecht: Springer

Verlag.

Bronner, G. N. (2013). Family Chrysochloridae: golden moles. In J.

Kingdon, D. C. D. Happold, T. M. Butynski, M. Hoffmann, M.

Happold, and J. Kalina (Eds.), Mammals of Africa. Volume I:

introductiory chapters and Afrotheria (Vol. Volume 1, pp. 223-

257). London: Bloomsbury Publishing.

Burne, R. H. (1901). A contribution to the myology and visceral

anatomy of Chlamydophorus truncatus. Proceedings of the

Zoological Society, London .: 104-121.

Camens, A. B. (2010). Were early Tertiary monotremes really all

aquatic? Inferring paleobiology and phylogeny from a depauperate

fossil record. Proceedings of the National Academy of Sciences of

the United States of America 107: E12-E12.

Campbell, B. (1939). The shoulder anatomy of the moles. A study in

phylogeny and adaptation. American Journal of Anatomy 64:

1-39.

Campione, N. E., and Evans, D. C. (2012). A universal scaling

relationship between body mass and proximal limb bone

dimensions in quadrupedal terrestrial tetrapods. BMC Biology 10:

60.

Canoville, A., and Laurin, M. (2010). Evolution of humeral

microanatomy and lifestyle in amniotes, and some comments on

palaeobiological inferences. Biological Journal of the Linnean

Society 100: 384-406.

Carlsson, A. (1904). Zur Anatomie des Notoryctes typhlops.

Zoologische Jahrbiicher. Abteilung fiir Anatomie und Ontogenie

der Tiere 20: 81-122.

Casinos, A., Quintana, C., and Viladiu, C. (1993). Allometry and

adaptation in the long bones of a digging group of rodents

(Ctenomyinae). Zoological Journal of the Linnean Society 107:

107-155.

Chapman, R. N. (1919). A study of the correlation of the pelvic

structure and the habits of certain burrowing mammals. American

Journal of Anatomy 25: 185-219.

Chou, H. H., Hayakawa, T., Diaz, S., Krings, M., Indriati, E., Leakey,

M., et al. (2002). Inactivation of CMP-N-acetylneuraminic acid

hydroxylase occurred prior to brain expansion during human

evolution. Proceedings of the National Academy of Sciences of

the United States of America 99: 11736-11741.

Cope, E. D. (1892). On the habits and affinities of the new Australian

mammal, Notoryctes typhlops. American Naturalist 26: 121-128.

Coues, E. (1872). The osteology and myology of Didelphys virginiana.

Memoirs of the Boston Society of Natural History 2: 41-154.

Crosby, K. (2002). A second species of the possum Durudawiri

(Marsupialia: Miralinidae) from the early Miocene of Riversleigh,

northwestern Queensland. Alcheringa 26: 333-340.

Crosby, K., and Archer, M. (2000). Durudawirines, a new group of

phalangeroid marsupials from the Miocene of Riversleigh,

northwestern Queensland. Journal of Paleontology 74: 327-335.

Cuneo, R., Ramezani, J., Scasso, R., Pol, D., Escapa, I., Zavattieri, A.

M., et al. (2013). High-precision U-Pb geochronology and a new

chronostratigraphy for the Canadon Asfalto Basin, Chubut,

central Patagonia: Implications for terrestrial faunal and floral

evolution in Jurassic. Gondwana Research 24: 1267-1275.

Davis, D. D. (1964). The giant panda: a morphological study of

evolutionary mechanisms. Fieldiana: Zoology Memoirs 3: 1-339.

Dennis, C. (2004). Zoology: a mole in hand. Nature 432: 142-143.

R.M.D. Beck, N.M. Warburton, M. Archer, S.J. Hand, and K.P. Aplin

dos Reis, M., Donoghue, P. C. J., and Yang, Z. (2014). Neither

phylogenomic nor palaeontological data support a Palaeogene

origin of placental mammals. Biology Letters 10: 20131003.

dos Reis, M., Inoue, J., Hasegawa, M., Asher, R. J., Donoghue, P. C.,

and Yang, Z. (2012). Phylogenomic datasets provide both precision

and accuracy in estimating the timescale of placental mammal

phylogeny. Proc Biol Sci 279: 3491-3500.

Edwards, L. F. (1937). Morphology of the forelimb of the mole

(Scalops aquaticus, L.) in relation to its fossorial habits. The Ohio

Journal of Science 37: 20-41.

Elissamburu, A., and Vizcaino, S. F. (2004). Limb proportions and

adaptations in caviomorph rodents (Rodentia: Caviomorpha).

Journal of Zoology 262: 145-159.

Emerling, C. A., and Springer, M. S. (2014). Eyes underground:

Regression of visual protein networks in subterranean mammals.

Molecular Phylogenetics and Evolution 78C: 260-270.

Evans, H. E. (1993). Miller’s anatomy of the dog. Philadelphia: W.B.

Saunders.

Farina, R. A., and Vizcaino, S. F. (1997). Allometry of the leg bones

of living and extinct armadillos. Mammalian Biology 62: 65-70.

Fox, R. C., and Naylor, B. G. (2006). Stagodontid marsupials from the

Late Cretaceous of Canada and their systematic and functional

implications. Acta Palaeontologica Polonica 51: 13-36.

Gambaryan, P. P, and Kielan-Jaworowska, S. (1997). Sprawling

verses parasagittal stance in multituberculate mammals. Acta

Palaeontologica Polonica 42: 13-44.

Gasc, J. P, Jouffroy, F. K., Renous, S., and Von Blottnitz, F. (1986).

Morphofunctional study of the digging system of the Namib

Desert golden mole (Eremitalpa granti nambiensis):

cinefluorographical and anatomical analysis. Journal of Zoology,

London 208: 9-35.

Geiger, M., Forasiepi, A. M., Koyabu, D., and Sanchez-Villagra, M. R.

(2014). Heterochrony and post-natal growth in mammals - an

examination of growth plates in limbs. Journal of Evolutionary

Biology 27: 98-115.

Gott, M. (1988). A Tertiary marsupial mole (Marsupialia:

Notoryctidae) from Riversleigh, northeastern Australia and its

bearing on notoryctemorphian phylogenetic systematics. Masters

Thesis, School of Biological Science, University of New South

Wales,

Hickman, G. C. (1984). Swimming ability of talpid moles, with

particular reference to the semi-aquatic Condylura cristata.

Mammalia 48: 505-513.

Hildebrand, M. (1985). Digging of quadrupeds. In M. Hildebrand, D.

M. Bramble, K. F. Liem, and D. B. Wake (Eds.), Functional

vertebrate morphology (pp. 89-109). Cambridge, Massachusetts:

Belknap Press.

Hildebrand, M., and Goslow, G. E. J. (2001). Analysis of vertebrate

structure. USA: John Wiley & Sons, Inc.

Hopkins, S. S. B., and Davis, E. B. (2009). Quantitative morphological

proxies for fossoriality in small mammals. Journal of Mammalogy

90: 1449-1460.

Horovitz, I., Ladeveze, S., Argot, C., Macrini, T. E., Martin, T.,

Hooker, J. J., et al. (2008). The anatomy of Herpetotherium cf.

fugax Cope, 1873, a metatherian from the Oligocene of North

America. Palaeontographica Abteilung A 284: 109-141.

Horovitz, I., and Sanchez-Villagra, M. R. (2003). A morphological

analysis of marsupial mammal higher-level phylogenetic

relationships. Cladistics 19: 181-212.

Jansa, S. A., and Voss, R. S. (2011). Adaptive evolution of the venom-

targeted vWF protein in opossums that eat pitvipers. PLoS ONE

6: e20997.

Fossil marsupial mole postcrania

169

Johnson, K. A., and Walton, D. W. (1989). 23. Notoryctidae. In D. W.

Walton, and B. J. Richardson (Eds.), Fauna of Australia: Volume

IB Mammalia (pp. 1-24). Canberra: AGPS.

Kim, E. B., Fang, X., Fushan, A. A., Huang, Z., Lobanov, A. V., Han,

L. , et al. (2011). Genome sequencing reveals insights into

physiology and longevity of the naked mole rat. Nature 479: 223-

227.

Kirsch, J. A. W. (1977). The comparative serology of Marsupialia, and

a classification of marsupials. Australian Journal of Zoology,

Supplementary Series 52: 1-152.

Krajewski, C., Wroe, S., and Westerman, M. (2000). Molecular

evidence for the pattern and timing of cladogenesis in dasyurid

marsupials. Zoological Journal of the Linnean Society 130: 375-

404.

Lagaria, A., and Youlatos, D. (2006). Anatomical correlates to scratch

digging in the forelimb of European ground squirrels

(,Spermophilus citellus). Journal of Mammalogy 87: 563-570.

Lanfear, R., Calcott, B., Ho, S. Y„ and Guindon, S. (2012).

Partitionfinder: combined selection of partitioning schemes and

substitution models for phylogenetic analyses. Molecular Biology

and Evolution 29: 1695-1701.

Lehmann, W. H. (1963). The forelimb architecture of some fossorial

rodents. Journal of Morphology 113: 59-76.

Lessa, E. P. (1990). Morphological evolution of subterranean

mammals: integrating structural, functional, and ecological

perspectives. Progress in Clinical and Biological Research 335:

211-230.

Lessa, E. P., Vassallo, A. I., Verzi, D. H., and Mora, M. S. (2008).

Evolution of morphological adaptations for digging in living and

extinct ctenomyid and octodontid rodents. Biological Journal of

the Linnean Society 95: 267-283.

Lewis, P. O. (2001). A likelihood approach to estimating phylogeny

from discrete morphological character data. Systematic Biology

50: 913-925.

Lillegraven, J. A. (1975). Biological considerations of the marsupial-

placental dichotomy. Evolution 29: 707-722.

Longrich, N. (2004). Aquatic specialization in mammals from the

Late Cretaceous of North America. Journal of Vertebrate

Paleontology 24: 84A.

MacAlister, A. (1875a). Report on the anatomy of the insectivorous

edentates. Transactions of the Royal Irish Academy 24: 491-508.

Macalister, M. B. (1875b). A monograph of the anatomy of

Chlamydophorus Truncatus (Harlan) with notes on the structure

of other species of Edentata. Transactions of the Royal Irish

Academy 25: 219-278.

Marshall, L. G. (1975). Chironectes minimus. Mammalian Species

109: 1-6.

Martin, H. A. (2006). Cenozoic climatic change and the development

of the arid vegetation in Australia. Journal of Arid Environments

66: 533-563.

Martin, H. A., and McMinn, A. (1994). Late Cainozoic vegetation

history of north-western Australia from the palynology of a deep

sea core (ODP site 765). Australian Journal of Botany 42: 95-102.

McLaren, S., and Wallace, M. W. (2010). Plio-Pleistocene climate

change and the onset of aridity in southeastern Australia. Global

and Planetary Change 71: 55-72.

Meier, P. S., Bickelmann, C., Scheyer, T. M., Koyabu, D., and Sanchez-

Villagra, M. R. (2013). Evolution of bone compactness in extant

and extinct moles (Talpidae): exploring humeral microstructure in

small fossorial mammals. BMC Evolutionary Biology 13: 55.

Meredith, R. W., Gatesy, J., Murphy, W. J., Ryder, O. A., and Springer,

M. S. (2009a). Molecular decay of the tooth gene enamelin

(ENAM) mirrors the loss of enamel in the fossil record of placental

mammals. PLoS Genetics 5: el000634.

Meredith, R. W., Janecka, J. E., Gatesy, J., Ryder, O. A., Fisher, C. A.,

Teeling, E. C., et al. (2011). Impacts of the Cretaceous Terrestrial

Revolution and KPg extinction on mammal diversification.

Science 334: 521-524.

Meredith, R. W., Krajewski, C., Westerman, M., and Springer, M. S.

(2009b). Relationships and divergence times among the orders

and families of Marsupialia. Museum of Northern Arizona

Bulletin 65: 383-406.

Meredith, R. W., Westerman, M., Case, J. A., and Springer, M. S.

(2008). A phylogeny and timescale for marsupial evolution based

on sequences for five nuclear genes. Journal of Mammalian

Evolution 15: 1-36.

Milne, N., and O’Higgins, P. (2012). Scaling of form and function in

the xenarthran femur: a 100-fold increase in body mass is

mitigated by repositioning of the third trochanter. Proceedings of

the Royal Society B: Biological Sciences 279: 3449-3456.

Mirceta, S., Signore, A. V., Burns, J. M., Cossins, A. R., Campbell, K.

L., and Berenbrink, M. (2013). Evolution of mammalian diving

capacity traced by myoglobin net surface charge. Science 340:

1234192.

Mitchell, K. J., Pratt, R. C., Watson, L. N., Gibb, G. C., Llamas, B.,

Kasper, M., et al. (2014). Molecular phylogeny, biogeography, and

habitat preference evolution of marsupials. Molecular Biology

and Evolution.

Muirhead, J. (1997). Two new early Miocene thylacines from

Riversleigh, northwestern Queensland. Memoirs of the

Queensland Museum 41: 367-377.

Murray, P. F., and Megirian, D. (2006). The Pwerte Marnte Marnte

Local Fauna: a new vertebrate assemblage of presumed Oligocene

age from the Northern Territory of Australia. Alcheringa Special

Issue 1: 211-228.

Musser, A. M. (2013). Classification and evolution of the monotremes.

In K. Ashwell (Ed.), Neurobiology of monotremes: brain evolution

in our distant mammalian cousins (pp. 1-16). Collingwood,

Australia: CSIRO Publishing.

Nevo, E. (1979). Adaptive convergence and divergence of subterranean

mammals. Annual Review of Ecology and Systematics 10: 269-

308.

Nevo, E. (1999). Mosaic evolution of subterranean mammals:

regression, progression and global convergence. Oxford: Oxford

University Press.

Nilsson, M. A., Arnason, U., Spencer, P. B. S., and Janke, A. (2004).

Marsupial relationships and a timeline for marsupial radiation in

South Gondwana. Gene 340: 189-196.

Nilsson, M. A., Churakov, G., Sommer, M., Tran, N. V., Zemann, A.,

Brosius, J., et al. (2010). Tracking marsupial evolution using

archaic genomic retroposon insertions. PLoS Biology 8: el000436.

Orcutt, E. E. (1940). Studies on the muscles of the head, neck, and

pectoral appendages of Geomys bursarius. Journal of Mammalogy

21:37-52.

Pepperberg, D. R., Okajima, T. L., Wiggert, B., Ripps, H., Crouch, R.

K., and Chader, G. J. (1993). Interphotoreceptor retinoid-binding

protein (IRBP). Molecular biology and physiological role in the

visual cycle of rhodopsin. Molecular Neurobiology 7: 61-85.

Phillips, M. J., Bennett, T. H., and Lee, M. S. Y. (2009). Molecules,

morphology, and ecology indicate a recent, amphibious ancestry

for echidnas. Proceedings of the National Academy of Sciences of

the United States of America 106: 17089-17094.

Phillips, M. J., Bennett, T. H., and Lee, M. S. Y. (2010). Reply to

Camens: How recently did modern monotremes diversify?

Proceedings of the National Academy of Sciences of the United

States of America 107: E13-E13.

170

Prideaux, G. J., and Warburton, N. M. (2010). An osteology-based

appraisal of the phylogeny and evolution of kangaroos and

wallabies (Macropodidae: Marsupialia). Zoological Journal of

the Linnean Society 159: 954-987.

Puttick, G. M., and Jarvis, J. U. M. (1977). The functional anatomy of

the neck and forelimbs of the Cape Golden Mole, Chrysochloris

asiatica (Lipotyphla: Chrysochloridae). Zoologica Africana 12:

445-458.

Reed, C. A. (1951). Locomotion and appendicular anatomy in three

soricoid insectivores. American Midland Naturalist 45: 513-665.

Ronquist, F., Klopfstein, S., Vilhelmsen, L., Schulmeister, S., Murray,

D. L., and Rasnitsyn, A. P. (2012a). A total-evidence approach to

dating with fossils, applied to the early radiation of the

Hymenoptera. Systematic Biology 61: 973-999.

Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D. L., Darling, A.,

Hohna, S., et al. (2012b). MrBayes 3.2: efficient Bayesian

phylogenetic inference and model choice across a large model

space. Systematic Biology 61: 539-542.

Rose, K. D., and Emry, R. J. (1983). Extraordinary fossorial adaptations

in the Oligocene palaeanodonts Epoicotherium and Xenocranium

(Mammalia). Journal of Morphology 175: 33-56.

Rougier, G. W., Martinelli, A. G., Forasiepi, A. M., and Novacek, M.

J. (2007). New Jurassic mammals from Patagonia, Argentina: a

reappraisal of australosphenidan morphology and

interrelationships. American Museum Novitates 3566: 1-54.

Rzebik-Kowalska, B. (2013). Sorex bifidus n. sp. and the rich

insectivore mammal fauna (Erinaceomorpha, Soricomorpha,

Mammalia) from the Early Pleistocene of Zabia Cave in Poland.

Palaeontologia Electronica 16: 12A.

Saitou, N., and Ueda, S. (1994). Evolutionary rates of insertion and

deletion in noncoding nucleotide sequences of primates.

Molecular Biology and Evolution 11: 504-512.

Salton, J. A., and Sargis, E. J. (2008). Evolutionary morphology of the

Tenrecoidea (Mammalia) forelimb skeleton. In E. J. Sargis, and

M. Dagosto (Eds.), Mammalian Evolutionary Morphology: A

Tribute to Frederick S. Szalay (pp. 51-72). Dordrecht, The

Netherlands: Springer.

Schaller, O. (Ed.). (1992). Illustrated veterinary anatomical

nomenclature. Stuttgart: Enke.

Shen, B., Fang, T., Dai, M., Jones, G., and Zhang, S. (2013).

Independent losses of visual perception genes GjalO and Rbp3 in

echolocating bats (Order: Chiroptera). PLoS ONE 8: e68867.

Springer, M. S., Burk, A., Kavanagh, J. R., Waddell, V. G., and

Stanhope, M. J. (1997). The interphotoreceptor retinoid binding

protein gene in therian mammals: implications for higher level

relationships and evidence for loss of function in the marsupial

mole. Proceedings of the National Academy of Sciences of the

United States of America 94: 13754-13759.

Springer, M. S., Westerman, M., Kavanagh, J. R., Burk, A.,

Woodburne, M. O., Kao, D. J., et al. (1998). The origin of the

Australasian marsupial fauna and the phylogenetic affinities of the

enigmatic monito del monte and marsupial mole. Proceedings of

the Royal Society B Biological Sciences 265: 2381-2386.

Stein, B. R. (1986). Comparative limb myology of four arvicolid

rodent genera (Mammalia; Rodentia). Journal of Morphology

187: 321-342.

Stein, B. R. (1993). Comparative hindlimb morphology in geomyine and

thomomyine pocket gophers. Journal of Mammalogy 74: 86-94.

Stein, B. R., and Patton, J. L. (2008). Genus Chironectes Illiger, 1811.

In A. L. Gardner (Ed.), Mammals of South America. Vol. 1.

Marsupials, xenarthrans, shrews, and bats (pp. 14-17). Chicago:

Chicago University Press.

Stirling, E. C. (1888). Preliminary notes on a new Australian mammal.

Transactions of the Royal Society of South Australia 11: 21-24.

R.M.D. Beck, N.M. Warburton, M. Archer, S.J. Hand, and K.P. Aplin

Stirling, E. C. (1891). Description of a new genus and species of

Marsupialia, Notoryctes typhlops. Transactions of the Royal

Society of South Australia 14: 154-187.

Stromberg, C. A. E. (2011). Evolution of grasses and grassland

ecosystems. Annual Review of Earth and Planetary Sciences 39:

517-544.

Sweet, G. (1906). Contribution to our knowledge of the anatomy of

Notoryctes typhlops , Stirling. Part III - the eye. Quarterly Journal

of Microscopical Science 50: 547-572.

Szalay, F. S. (1994). Evolutionary history of the marsupials and an

analysis of osteological characters. Cambridge: Cambridge

University Press.

Szalay, F. S., and Sargis, E. J. (2001). Model-based analysis of

postcranial osteology of marsupials from the Palaeocene of

Itaboraf (Brazil) and the phylogenetics and biogeography of

Metatheria. Geodiversitas 23: 139-302.

Taylor, B. K. (1978). The anatomy of the forelimb of the anteater

(Tamandua ) and its functional implications. Journal of

Morphology 157: 347-368.

Thompson, P., and Hillier, W. T. (1905). The myology of the hindlimb

of the marsupial mole ( Notoryctes typhlops ). Journal of Anatomy

and Physiology 34: 308-331.

Thorington, R. W. J., Darrow, K., and Betts, A. D. K. (1997).

Comparative myology of the forelimb of squirrels (Sciuridae).

Journal of Morphology 234: 155-182.

Travouillon, K. J., Archer, M., Hand, S. J., and Godthelp, H. (2006).

Multivariate analyses of Cenozoic mammalian faunas from

Riversleigh, northwestern Queensland. Alcheringa Special Issue

1: Proceedings of CAVEPS 2005: 323-349.

Travouillon, K. J., Legendre, S., Archer, M., and Hand, S. J. (2009).

Palaeoecological analyses of Riversleigh’s Oligo-Miocene sites:

implications for Oligo-Miocene climate change in Australia.

Palaeogeography, Palaeoclimatology, Palaeoecology 276: 24-

37.

Turnbull, W. D. (1971). The Trinity therians: their bearing on evolution

in marsupials and other therians. In A. A. Dahlberg (Ed.), Dental

morphology and evolution (pp. 151-179). Chicago: University of

Chicago Press.

Turnbull, W. D., and Reed, C. A. (1967). Pseudochrysochloris, a

specialized burrowing mammal from the early Oligocene of

Wyoming. Journal of Paleontology 41: 623-631.

Vassallo, A. I. (1998). Functional morphology, comparative behaviour,

and adaptation in two sympatric subterranean rodents genus

Ctenomys (Caviomorpha: Octodontidae). Journal of Zoology 244:

415-427.

Vizcaino, S. F., Farina, R. A., and Mazzetta, G. V. (1999). Ulnar

dimensions and fossoriality in armadillos. Acta Theriologica 44:

309-320.

Vizcaino, S. F., and Milne, N. (2002). Structure and function in

armadillo limbs (Mammalia : Xenarthra : Dasypodidae). Journal

of Zoology 257: 117-127.

Warburton, N. M. (2003). Functional morphology and evolution of

marsupial moles (Marsupialia; Notoryctemorphia). Unpublished

Ph.D. dissertation. University of Western Australia,

Warburton, N. M. (2006). Functional morphology of marsupial moles

(Marsupialia, Notoryctidae). Verhandlungen des

Naturwissenschaftlichen Vereins in Hamburg 42: 39-149.

Warburton, N. M., Gregoire, L., Jacques, S., and Flandrin, C. (2013).

Adaptations for digging in the forelimb muscle anatomy of the

southern brown bandicoot ( Isoodon obesulus) and bilby ( Macrotis

lagotis ). Australian Journal of Zoology 61: 402-419.

Fossil marsupial mole postcrania

171

Welch, J. J., Bininda-Emonds, O. R., and Bromham, L. (2008).

Correlates of substitution rate variation in mammalian protein¬

coding sequences. BMC Evolutionary Biology 8: 53.

Whidden, H. P. (1999). The evolution of locomotor specializations in

moles. American Zoologist 39: 135A.

Windle, B. C. A., and Parsons, F. G. (1899). On the myology of the

Edentata. Proceedings of the Zoological Society, London 67: 314-339.

Woodhead, J., Hand, S. J., Archer, M., Graham, I., Sniderman, K., Arena,

D. A., et al. (2014). Developing a radiometrically-dated chronologic

sequence for Neogene biotic change in Australia, from the Riversleigh

World Heritage Area of Queensland. Gondwana Research.

Woodman, N., and Gaffney, S. A. (2014). Can they dig it? Functional

morphology and semifossoriality among small-eared shrews,

genus Cryptotis (Mammalia, Soricidae). Journal of Morphology

275: 745-759.

Yang, Z. (2007). PAML 4: phylogenetic analysis by maximum

likelihood. Molecular Biology and Evolution 24: 1586-1591.

Zhang, Z. D., Cayting, P., Weinstock, G., and Gerstein, M. (2008).

Analysis of nuclear receptor pseudogenes in vertebrates: how the

silent tell their stories. Molecular Biology and Evolution 25:

131-143.

Memoirs of Museum Victoria 74:173-187 (2016) Published 2016

ISSN 1447-2546 (Print) 1447-2554 (On-line)

http://museumvictoria.com.au/about/books-and-journals/journals/memoirs-of-museum-victoria/

New specimens of ektopodontids (Marsupialia: Ektopodontidae) from South

Australia

Neville S. Pledge

South Australian Museum, Adelaide, South Australia 5000 (neville.pledge@samuseum.sa.gov.au)

Abstract Pledge, N.S. 2016. New specimens of ektopodontids (Marsupialia: Ektopodontidae) from South Australia. Memoirs of

Museum Victoria 74: 173-187.

Knowledge about the extinct phalangeroid family Ektopodontidae is increased by the discovery of new material

from several localities. Previously unknown teeth of Chunia illuminata and Ektopodon stirtoni are described respectively

from White Sands Basin and Mammalon Hill, Lake Palankarinna, Lake Eyre Basin, South Australia, with M 1 being

recorded for the first time for any species of Chunia., and a full maxilla of Ektopodon stirtoni showing the positional

relationship between P3 and Ml for the first time; this is even more extreme than the arrangement postulated previously.

Another species Ektopodon litolophus has been described on the basis of an M 1 found at the Leaf Locality, Lake

Ngapakaldi, Lake Eyre Basin. Material from Lake Tarkarooloo, referred to Ektopodon stirtoni, is redescribed as a new

species Ektopodon tommosi. Comparisons of M 1 of Chunia and Ektopodon species now allow evolutionary trends, such

as increasing number of cusps on the molar lophs, and simplification of the cusps, to be discerned.

Keywords Marsupialia, Ektopodontidae, Chunia, Ektopodon, Ditjimanka, Ngama, Kutjamarpu, Etadunna Formation, Wipajiri

Formation, Tertiary, Oligocene, Miocene, Lake Eyre Basin, Australia.

Introduction

Ektopodon is a genus of extinct possum-like marsupials

established by Stirton et al. (1961) for isolated teeth found at the

early Miocene Leaf Locality at Lake Ngapakaldi, northeastern

South Australia. Further specimens from this locality were

described and interpreted by Woodbume and Clemens (1986b),

together with new, slightly older late Oligocene species in the

plesiomorphic genus Chunia (C. illuminata, C. sp. cf. C.

illuminata and C. omega ) from sites in the Lake Eyre Basin (e.g.

Tedford Locality, Lake Palankarinna) and the Frame Embayment

(e.g. Tom O’s Quarry, Lake Tarkarooloo). A second, later

Oligocene species of Ektopodon ( E. stirtoni ) was described by

Pledge (1986) from Mammalon Hill, Lake Palankarinna (fig.

L), and specimens also from Lake Tarkarooloo were also

referred to E. stirtoni.

Rich (1986) described Darcius duggani from Hamilton,

western Victoria, a deposit radiometrically determined to be

Pliocene in age (ca. 4.46 Ma), and E. paucicristata from Pliocene

sediments near Portland, also in western Victoria (Rich et al.,

2006). By 1986 the rich deposits of Riversleigh, northwestern

Queensland, were beginning to yield many new species including

an unidentifiable ektopodontid, and a second species of Ektopodon

was found in the Kutjamarpu Local Fauna of South Australia

(Pledge et al. 1999). New specimens from Lake Palankarinna

referable to Chunia illwninata and Ektopodon stirtoni are

described in this paper. The status of the Tarkarooloo specimens

referred to E. stirtoni is re-evaluated and a new species erected.

Age

Although Stirton (Stirton et al., 1961) initially assessed the age

of the Etadunna Formation and its faunas to be Oligocene, it

became customary to interpret these central Australian

deposits as middle Miocene in age (approximately 12-15 Ma;

Woodburne et al., 1985), following preliminary analyses by

W. K. Harris of pollen, wrongly identified as grasses, from the

Etadunna Formation; this was amended to Restionacea (sedge)

pollen by Martin (1990). Later work with foraminiferans

(Lindsay 1987) suggested that at least some of these deposits

may be as old as Late Oligocene. This assessment was based

on the abundant presence of the foraminiferan Buliminoides

sp. cf. B. chattonensis. The older age is also supported by

Truswell et al. (1985) on the basis of pollens from the Geera

Clay (which appears to be a lithological equivalent of basal

Namba Formation, itself correlated with the Etadunna

Formation), and by Norrish and Pickering (1983) who reported

a Late Oligocene Rb-Sr date for authigenic illite from the

Etadunna Formation.

Collection methods

Specimens from the Lake Eyre Basin were found by

excavation of the fossiliferous horizons or by screen-washing

the dried sediment through window screen with mesh of 6 x 6

wires per cm 2 .

174

N.S. Pledge

Figure 1. Locality map. Ektopodontid localities.

Terminology

Molar homology follows Luckett (1993). The terminology for

the crown morphology of ektopodontid teeth used here is that

of Woodburne & Clemens (1986a) with modifications to cusp

homology as recommended by Tedford & Woodburne (1987).

Museum abbreviations

SAM, South Australian Museum; NMV, Museum Victoria;

UCMP, University of California, Berkeley; UCMP V,

University of California Museum of Paleontology locality

number; UCR, University of California, Riverside; SAM PL,

South Australian Museum Palaeontological locality.

Systematic Palaeontology

Marsupialia (Illiger, 1811) sens. Cuvier (1817)

Diprotodontia Owen, 1866

Phalangeroidea (Thomas, 1888) sens. Aplin & Archer (1987)

Ektopodontidae Stirton, Tedford & Woodburne, 1967

Chunia Woodburne & Clemens, 1986

Chunia illuminata Woodburne & Clemens, 1986

(Figs. 3A-D)

Holotype. SAM P17997. In their original description of Chunia

illuminata, Woodburne & Clemens (1986b) described the RM 2

holotype SAM P17997 (= RM 3 in their notation following Archer

1978), together with QM F10641 (maxilla fragment with partial LM 2 ),

UCR 15228 (LM 2 ) and UCR 15227 (LM 2 ).

New Specimens. SAM P29081, left M 1 (figs. 3A-D) collected by J.

McNamara 9 July 1987; also SAM P33944, left M 3 , collected by J.

Case and J. Clemitson, June 1992, from Tedford Locality.

Locality. White Sands Basin (SAM locality PL 7719), 200 m

south of Tedford Locality (UCMP V5375, the Type Locality for

the species).

Stratigraphy and Age. Stratigraphically, White Sands Basin is

about one metre below the sandy clay layer in Tedford Locality

that produced the holotype. This fauna from the Etadunna

Formation is considered to be Late Oligocene in age (see above).

Revision of specific diagnosis. In addition to the features noted

by Woodburne & Clemens (1986b), including the upper dental

formula I ? , C 1 , P 13 , M 1 4 , the M 1 of Chunia illuminata differs

from that tooth in all other ektopodontids in being smaller,

having fewer cusps on its lophs and in having a parastyloph that

is less loph-like and more cusp-like than that structure in any

known species of Ektopodon. The face is ‘longer’ and less

obtuse than that of Ektopodon stirtoni.

Description. The M 1 of Chunia illuminata has the same basic

outline as that tooth in other species of Ektopodon but differs

in several ways noted below. The cusps on the ‘protoloph’

and ‘metaloph’ are all worn apically, so some detail has

been obscured.

The parastyloph (“paraloph” of Pledge, 1986) is simpler

than that structure in other species, being an oblique blade

confluent with the buccal face of the tooth and having three

cusps. The minute lingual cusp is offset posteriorly from the

end of the loph and gives rise to a pair of postcristae that

initially diverge, then converge slightly linguad. A weak

precrista descends basally from the point of inflection of the

loph. The central cusp is by far the largest of the three but,

apart from the loph crest, gives rise only to a short postcrista

at right angles to it. However, a strong postcrista arises from

near the buccal end of the cusp and continues transversely to

almost meet the converging postcristae from the lingual cusp.

The buccal cusp is separated from the median cusp by a

narrow crevice. There are two postcristae, a buccal one

forming part of the “parastyloph” and a lingual one that curves

transversely and extends half the width of the loph.

The protoloph has four distinct cusps and a complex

structure at the buccal end that could represent either two or

three smaller cusps. The lingual cusp, the protocone, is a

trigonal pyramid with the precrista being stronger than either

the postcrista or the lingual extension of the loph crest. Cusp 2

is the smallest with a short precrista cut off by converging

precristae from the protocone and cusp 3. Its postcrista is the

New specimens of ektopodontids (Marsupialia: Ektopodontidae) from south australia

175

EPOCH

VICTORIA

Otway Basin

SOUTH AUSTRALIA

Callabonna Sub-basin Tirari Sub-basin

NORTHERN

TERRITORY

Pleistocene

Pliocene

10 —

Miocene

15 —

25 —

Oligocene

30 —

- jfr Nelson Bay r

Childers uovo

% Portland |-

Hamilton

(Grange Burn Fm)

#Kutjamarpu

(Wipajiri Fm)

- ? -- 9 -

-X-Tarkarooloo

Ericmas

Pinpa

Treasure

^ Ngama

Ngapakaldi

% Ditjimanka

Minkina

Kangaroo Well

(Namba Fm)

_ 9 -? _

(Etadunna Fm)

- 9 -- 9 -

OUlta

Limestone")

BMcH

Figure 2. Stratigraphic distribution of named ektopodontid species.

176

N.S. Pledge

Figure 3. Chunia spp. teeth: a-d. Chunia illuminata, Woodburne and Clemens, 1986. a. right maxilla. QM F10641, mirror-imaged to show angle

of face, Tedford Locality; b. M 1 SAM P29081, White Sands Basin; c. SAM P17997, (type) M 2 Tedford Locality; d. M 3 SAM P33944, Tedford

Locality, Lake Palankarinna, Ditjimanka Local Fauna; e. Chunia omega Woodburne and Clemens, 1986, half of M3? (type) SAM P23065, Tom

O’s Quarry, Lake Tarkarooloo, Tarkarooloo Local Fauna. Abbreviations: mel, metaconule; pastl, parastyloph; pr, protocone; prl, protoloph; 3,

cusp 3; 5, cusp 5.

New specimens of ektopodontids (Marsupialia: Ektopodontidae) from south australia

177

simplest (after that of the protocone) and extends to the

transverse valley; a weak lingual spur arises about halfway

along its length and the distal (posterior) end curves lingually to

parallel the transverse valley. Cusps 3 and 4 are basically

similar, each being large and having a pair of subparallel

precristae and postcristae that are angled slightly linguad. Cusp

3, however, also has a third slightly sinuous precrista, two weak

ribs on the buccal face of the outermost precrista and a

bifurcation of the buccal postcrista. Cusp 5 shows two strong

postcristae reaching the transverse valley and a shorter,

bifurcating buccal postcrista. Anteriorly, two sinuous precristae

are linked by two or more near-apical struts and a basal strut.

The lingual precrista bifurcates just below a strut linking it with

the buccal precrista of cusp 4. All cusps (1-5) are linked apically

by a fine transverse strut. Cusp 6 is displaced posteriorly and is

a small trigonal structure with pre-, post- and transverse cristae.

The transverse valley is deep, trenchant, slightly curved

and anteriorly convex. No structures cross it except at the

lingual end where there is a notched cingulum. Buccal to this

are three weak irregular postcristae at the base of the protocone

and four irregular precristae on the metaconule, none of which

cross the valley. The lingual cingulum curves and extends up

the lingual face of the metaconule (i.e. the posterolingual cusp

of diprotodontians previously called the hypocone; see Tedford

& Woodburne, 1987).

The metaconule is stronger and more bulbous than the

protocone. Besides the irregular basal precristae, there are two

short postcristae that form a small pocket on the posterior face.

The main precrista is aligned with the main postcrista and with

the postcrista of cusp 2 of the protoloph. Cusp 2 of the metaloph

is similar in size to cusp 3 of the protoloph. Besides the strut

linking it to the metaconule, both of the pre- and postcristae bear

several transverse ribs on the lingual face of the lingual cristae

and the buccal cristae bifurcate basally. Cusp 3 is similar but with

fewer and weaker ribs; only its buccal postcrista bifurcates.

However, the apex of the cusp appears to be a triangle of short

crests with its base aligned with the buccal cristae. Buccally from

here, the structure is unclear. Cusp 4 appears to be a relatively

simple structure with a single precrista and a postcrista bearing

several weak lingual ribs. Near its apex, however, a short crest

leaves posterobuccally to join one coming anterolingually from

another, posteriorly offset cusp. A short postcrista arises at the

junction of these crests but does not reach the posterior cingulum.

This posterior cusp also bears four other radiating crests, two

being aligned longitudinally and the others antero- and

posterobuccally. Buccal to cusp 4 are three or four weak cusps

defined by four simple precristae alternating with three short

postcristae that do not meet the cristae from the posterior cusp. At

the base of these precristae a low transverse crest parallels the

transverse valley. A distinct postcingulum extends from the

buccal-most postcrista to the metaconule. Pledge (1986) attempted

to equate these buccal structures (in species of Ektopodon ) with

the cusp and crest patterns common in diprotodontians but it now

appears, even in this relatively plesiomorphic species, that the

homology of these cusps in ektopodontids is unclear.

M 3 . The new specimen found by Case and Clemitson SAM

P33944 is almost identical to, but the mirror image of, the

holotype SAM P17977, and paratype SAM P22722 M 2 s from

the same locality. It differs in being slightly longer and

narrower, with slightly more prominent equiradial development

of crests and struts, and is thus accorded here a more posterior

position. This tooth bears some resemblance to the incomplete

type specimen of Chunia omega (fig. 3E) from the Tarkarooloo

Local Fauna (Woodburne and Clemens 1986).

Ektopodon Stirton, Tedford & Woodburne, 1967.

Distribution. Kutjamarpu Local Fauna, Wipajiri Formation,

Lake Ngapakaldi; Ngama Local Fauna, Etadunna Formation,

Lake Palankarinna; Tarkarooloo Local Fauna, Namba

Formation, Lake Tarkarooloo; Portland Bay Local Fauna,

Whalers Bluff Formation, Portland.

Age. Late Oligocene to early-mid Miocene, Pliocene.

Diagnosis (revised). As for Woodburne and Clemens (1986b)

but with the revision of the dental formula which is now

understood to be: I?/l, Cl/1, P3/3, Ml-4/1-4. This revision

involves recognition of the presence of only a single premolar.

Ektopodon serratus Stirton, Tedford & Woodburne, 1967.

Holotype. SAM P 13847. (Fig. 6B).

Ektopodon tommosi sp. nov.

Ektopodon sp. cf. E. stirtoni Pledge 1986, pis 3.ID, F, G; 3.2;

3.3D-E (fig. 4).

Holotype. SAM P19962 (RM 1 ).

Paratypes. NMV P48750-48751 (LM 1 ), SAM P19963 (RM 2 ), SAM

P19950 (RM,), NMV P48752 (LM 2 ), NMV P48753 (LM 3 ), NMV

P48757 (LM 4 ), NMV P48764 (LM 2 ), NMV P48765 (RM 3 ), NMV

P48766 (RM 4 ).

Referred specimens. NMV P48758 (right lower incisor), NMV P48769

(partial LP 3 ), SAM P19965 (RP 3 ).

Locality. Tom O’s Quarry, western shore of Lake Tarkarooloo,

Callabonna Basin. 31°8.5'S., 140°6.3'E.

Horizon. A sandy channel deposit within the Namba Formation

(Callen & Tedford, 1976).

Age. Late Oligocene, Tarkarooloo Local Fauna. Biocorrelation

suggests this is faunistically equivalent to the Ngapakaldi Local

Fauna of the Lake Eyre Basin (Rich and Rich, 1987), which is a

little older than the latest Oligocene Ngama L.F. of Lake

Palankarinna, but younger than the Pinpa L.F. from Lake Pinpa.

Diagnosis. The molar teeth are generally 5-10% smaller than

comparable elements of E. stirtoni, and the loph(id)s generally

have one less cusp. The protostyloph on M 1 is shorter and less

loph-like with two cusps, one less than in E. stirtoni. The P 3 is

larger than that tooth in E. stirtoni. The mandible is slightly

larger than in other species, with a longer diastema.

Etymology. The species name reflects the source of these

specimens, Tom O’s Quarry site, (discovered by Cpl. John

Thompson - ‘Tom O’ to distinguish him from Tom Rich - of 3 rd

RAAME which provided logistic support for Rich’s 1974

expedition; Rich and Archer, 1979), at Lake Tarkarooloo.

178

N.S. Pledge

4 mm

Figure 4. Dentition of Ektopodon tommosi nom. nov. a. composite upper dentition, anterior at left: LM 1 fragment (NMV P48750), partial RM 1

(SAM P19962), LM 2 (NMV P48752), RM 2 (SAM P19963), LM 3 (NMV P48753), LM 4 (NMV P48757); b. LM 3 (NMV P48753); c. RM 2 (SAM

P19963); d. RIj (NMV P48758); e. RMj (SAM P19950. Tom O’s Quarry, Lake Tarkarooloo; Tarkarooloo Local Fauna. Abbreviations: mel,

metaconule; pastd, parastylid; pastl, parastyloph; pr, protocone.

New specimens of ektopodontids (Marsupialia: Ektopodontidae) from south australia

179

Figure 5. Ektopodon stirtoni Pledge, 1986, holotype dentary and new material, a-c. new maxilla (SAM P35309): a. lateral; b. dorsal; c. palatal;

d. RM 2 (P23854); e. LM 3 (P30175); f. LM 3 (SAM P30156); g. right dentary (SAM P29577), lateral aspect; h. right dentary with M 2 4 (SAM

P29577) stereo pair; i, j. holotype right dentary with P 3 , M t 3 (SAM P19509); k. LM 3 (P31638); 1. RM4 (SAM P33451). Ngama Quarry, Mammalon

Hill, Lake Palankarinna; Ngama Local Fauna.

180

N.S. Pledge

Figure 6. Ektopodontid spp.; comparison of first upper molars, a. Chunia illuminata SAM P29081 (left); b. Ektopodon serratus SAM P13847

(left); c. Ektopodon stirtoni SAM P22504 (right); d. Ektopodon litolophus SAM P30176 (right); e. Ektopodon tommosi NMV P48750-1 (left); f.

Ektopodon tommosi SAM P19962 (right); g. Ektopodon ulta (from Megirian et al. 2004:719, fig. 15A); h. Ektopodon paucicristatus (from Rich

et al. 2006:137, fig. 3D). Scale bar approximately 1 cm, for a-d; others about same scale. Abbreviations: ca, canine alveolus; fo, infraorbital

foramen; mjs, maxillojugal suture; pastl, parastyloph.

New specimens of ektopodontids (Marsupialia: Ektopodontidae) from south australia

181

Description. See Pledge (1986: 53-60) for a more-complete

description. The following is abbreviated and has updated

terminology.

P 3 . There are still only two fragments known; the incomplete

specimen (NMV48769) referred to this taxon by Pledge (1986)

is correctly ascribed. It agrees morphologically with the P 3 of E.

stirtoni (see below) as far as can be compared, but is noticeably

larger (length >5.6, width 4.1 vs. length 5.5, width 3.4). The

curving longitudinal crest extends from a conspicuous anterior

cusp at the anterolingual corner of the tooth. At the midway

point, a deep angular saddle divides the crest, separating the

anterior cusp from two closely-linked larger posterior cusps.

The latter are separated by a deep narrow crevice. The crest

ends at the posterolingual corner of the tooth. Its lingual face

has a weak basal cingulum with two small cusps.

M 1 (fig. 4A). Only three fragmentary teeth have been

found, the holotype, SAM P19962 being one of them. The

latter is quite worn and lacks the lingual margin, but compares

well with P22504, the Ml of E. stirtoni from Mammalon Hill.

However, it differs, apparently, in having one less cusp on each

loph and a shorter protostyloph with only two well-developed

cusps and the trace of an incipiently-developed third cusp.

M 2 (figs. 4A, C). This molar is broad, roughly trapezoidal,

and the largest of the upper molars having only two lophs. The

protoloph, with eight cusps, is slightly longer than the

metaloph, which has seven, and the crests of the lophs slightly

twisted rather than being in the same plane. The crests are also

not as sharp as those structures on M 1 , nor the transverse

valley as deep. On the protoloph, the protocone bears a pair of

deep grooves on its anterior and posterior sides, but the

resulting ridges do not bifurcate. Similarly, the cristae of the

second protoloph cusp do not bifurcate, but those of the 3 rd , 4 th ,

5 th and 6 th cusps do (the last only on the posterior side).

Similarly on the metaloph, the metaconule has a pair of deep

grooves, and the 2 nd , 3 rd , 4 th and 5 th cristae bifurcate, the last

only on the anterior side. The posterior groove of the protocone

joins with the anterior one on the metaconule to form a deep

buccal pocket. All cusps on a loph are joined by a fine, deep-

set apical ‘strut’ along the axial plane of the loph. All cristae

are cut off by the pre- and post-cingula.

M 3 (figs. 4A, B). This tooth is of similar length to M 2 but is

noticeably narrower, and therefore trapezoidal in outline. The

protoloph has eight cusps and extends beyond the metaloph at

each end. The metaloph probably has seven cusps, but the

count is uncertain because of the difficulty in distinguishing

between primary cristae and bifurcations. Fine linking struts

between cristae increase in number buccally to at least four at

the paracone.

M 4 (fig. 4A). This tooth appears to be a stunted version of

M 3 . It is triangular, and bears only the protoloph which is of

low relief and has seven or eight cusps (number uncertain

because of the irregular bifurcation of the cristae). The missing

metaloph is replaced by the third corner of the triangle: an

inchoate network of low crests. The metaconule appears to

have merged into the protocone, and that combination presents

as a square extension of the protoloph.

Dentaries. Three dentaries are known, one retaining a

fragment of the hypolophid of M 3 ; the others may be a pair. They

are relatively massive, are slightly larger than those of E. stirtoni.

Compared with the latter, they have a more convex ventral

profile, and a longer diastema, but the alveolar cheek-tooth length

is shorter. A minute alveolus (for a canine?) immediately follows

that of the incisor on its dorso-lateral corner.

(fig. 4D). An isolated right incisor is referred to this

position, based on size and its semicircular cross-section to fit

the alveolus. It is short, high and somewhat spatulate, but it

cannot be proven to relate to this species.

P 3 . This tooth is represented by three specimens. It is smaller

than the P 3 of E. stirtoni, and more subdued in its features, but

otherwise similar. It is ovate in outline, and has on the lingual

side a longitudinal cristid that bears four cuspids. An isolated

anterior cuspid is followed by a central slightly larger one, then

a third even larger one to which is appressed a smaller fourth

cuspid. A posterobuccal cingulum forms a small pocket.

M, (Fig. 4E). SAM P19950 is the only complete specimen

of this tooth known. It can be recognised by its distinctive

‘parastylid’. The tooth is roughly rhomboid in outline and

wider than long. The ‘parastylid’, at the anterolingual corner

of the tooth, is more prominent than that structure in E.

stirtoni. The lingual face of the tooth is relatively flat; the

buccal ends of the lophids are quite swollen. Lophid crests are

sharp and parallel to themselves, but not to the anterior and

posterior faces of the tooth. The six protolophid cuspids are

not as well-graded as in E. stirtoni. The two most buccal are

very closely appressed, while the other four are well-spaced.

The same pattern of distribution of cuspids characterizes the

hypolophid. The precristids of cuspids 2-4 on the protolophid

bifurcate. The same occurs for cuspids 1-3 on the hypolophid.

M 2 . This tooth is relatively longer than M l and has a more

open transverse valley. Each lophid has seven cuspids, the

inner-most two being combined. On the protolophid, only the

precristid of the protoconid divides, and its lingual branch

divides again. On the hypolophid, the hypoconid and cuspid 2

divide on both sides. Although this occurs on cuspid 3, it does

so only on the precristid. The precristid of the metaconid has

a small notch, as it does in M p but it does not develop a

‘parastylid’ at the precingulum.

M 3 . This tooth is not represented by a complete specimen.

A fragment of hypolophid is preserved in the dentary NMV

P48767, Although another M 3 , NMV P48765, is more

complete, it lacks the protoconid and hypoconid. Based on

what is preserved, these specimens show a morphology similar

to that of E. stirtoni, with low lophids and a wide transverse

valley. On both lophids, the postcristids of cuspids 2 and 3

divide, and on the hypolophid, the precristids also. The lingual

part of the crown develops into a network of fine cristids and

struts, which differs in the extent of this network development

from that in the M 3 of E. stirtoni.

M 4 . Despite their superficially different appearances, two

specimens, SAM P19966 and NMV P48766, have been

identified as M 4 s. The former is incomplete and slightly more

worn than the other; both are from right dentaries. They are

roughly triangular as a result of reduction of the entoconid, and

are low-crowned with very low, broad lophids. There are six to

seven cuspids on each lophid, the number uncertain because of

similarities and irregularities of cristids and ribs. The transverse

182

N.S. Pledge

valley is an irregular network of anastomosing cristids and

ribs. These teeth are elongate (length:width ratio about 1.14).

NMV P48766 fits the alveolus of jaw NMV P48767.

Ektopodon stirtoni Pledge, 1986.

(Fig- 5)

Pledge (1986) named this taxon on the basis of a right dentary

with P 3 M 13 (SAM P19509) and an isolated RM 1 (P22504).

The new material described here adds considerably to

knowledge about this species.

New specimens. SAM P23854, RM 2 with lingual root, collected

by J. McNamara, 13 July, 1981; SAM P24541, LM 2 hypolophid,

collected by N. Pledge, August, 1983; SAM P23989, RM 3

hypolophid with root, collected by N. Pledge on 5 September,

1982; SAM P23988, LM 4 crown, collected by N. Pledge, 5

September, 1982. All the above were noted (per measurements

only) in Pledge (1986). The following have not previously been

noted. SAM P29577, right dentary with M 2 4 , collected by D.

Williams, 3 July, 1988; SAM P30156, LM 3 , collected by H.

Aslin, 1 September, 1989; SAM P30175, LM 3 , found by J.

Thurmer, 1 November, 1989 (fig. 4); SAM P31637, left maxilla

with M 2 , found by J. McNamara, 27 July 1990; SAM P31638, a

left M 2 found by B. McHenry, August 1990; SAM P33451, a

right M 4 collected by G. Aldridge, October 1991; and SAM

P35309, a left maxilla, with P 3 M 13 but missing the canine and

M 4 , collected by N. Haines, 8 August, 1995.

Locality. Mammalon Hill (SAM locality PL7611; the Type

Locality for the species), northwestern shore of Lake

Palankarinna, South Australia.

Age. The Ngama Local Fauna comes from the Mammalon Hill

beds (zone D) of the Etadunna Formation. The age of this

horizon is uncertain but considered to be Late Oligocene

(Pledge, 1984; Woodburne, 1986; Woodburne et al., 1994).

Revision of specific diagnosis. In addition to the diagnostic

features noted by Pledge (1986), these additional specimens of

Ektopodon stirtoni demonstrate that this species differs from

E. serratus, in greater size, greater length/width ratio of the

molars and fewer cusp(id)s. It differs from E. litolophus (Pledge

et al., 1999) in its smaller size, relatively narrower M 1 , relatively

less regular parastyloph with less uniformly sized cusps, fewer

and less uniformly-sized cusps on main lophs with obvious ribs

and struts, presence of posterior cingulum. The face is blunter

than that of Chunia illuminata (compare figs. 3A and 5B, C).

The original description of E. stirtoni (Pledge 1986) dealt

with the holotype dentary and an M 1 (ibid, plate 3.IB). In the

dentary, only the P 3 and Mj were complete. The present work

includes descriptions of the other lower molars (M 2 4 ) as well

as that of P 3 , M 2 and M 3 . These descriptions incorporate

specimens previously noted only in the table of measurements

(ibid, table 3.1).

Descriptions. Maxilla. Although the (mostly edentulous)

maxilla of Chunia illuminata had been known for some time

(Woodburne and Clemens, 1986b), it did not fully prepare us

for the morphology shown by the new specimen of Ektopodon.

SAM P35309 is virtually complete, lacking only the thin bone

on the medial edge of the palate at the midline suture, the

dorsal wing of the maxilla with the nasal contact, the small

canine, and the last molar M 4 (figs. 5A-C).

However, part of the jugal which forms the lower border of

the orbit and part of the zygomatic arch is also present. This

enables us to form a picture of the face of Ektopodon. The

orbits had an estimated diameter of up to 15 mm, and were

directed forwards and upwards in a wide (est. 50 mm across

cheekbones) flat face with a short, narrow muzzle. The lateral

face of the maxilla is gently convex, almost flat, from canine to

malar process of the zygomatic arch. There is no malar

depression or fossa. The maxillo-jugal suture is gently sinuous

from just behind the malar to the anterior ‘corner’ of the orbit,

with the jugal tapering from less than 3 mm below the orbit and

being a uniform 3 mm wide inside the edge of the orbit. The

bottom edge of the eye socket is 5.5 mm above the molar

occlusal surface. The infraorbital foramen is ovate and situated

about 4 mm behind the canine and 3-4 mm above the diastemal

crest; it emerges 12 mm posterior in the orbital floor.

The molar occlusal surface shows a slight torsion along its

length. The combined length of the cheektooth row, P 3 to M 3 is

23.2 mm. The maximum bone length, from canine alveolus to

posterior extremity of the palate, parallel to the molar tooth-

row, is 37.3 mm. The width of the maxillary palate to the

lingual margin of the molars is (2x) 12.7 mm. The maximum

palate width measured to the buccal edge of the molars is 22.2

mm. M 4 is represented by three alveoli.

Canine. Not much can be said of this tooth based on its

alveolus, apart from its existence and its gingival diameter:

maximum 2 mm. The alveolar depth of 5 mm suggests it was

neither a large nor particularly functional tooth. It is situated at

the extreme corner of the maxilla, next to the premaxillar suture.

P 3 (fig. 5C). Previously, the ektopodontid P 3 was known

only from two referred fragments from Lake Tarkarooloo

(Pledge, 1986). These have here been reinterpreted (below) to

represent E. tommosi n. sp.

The P 3 of SAM P35309 is somewhat recumbent, with its

anterior root extending well into the diastema. However, it is

almost transverse to the long axis of the molar-row, and its

posterolingual corner is tucked neatly into the angle formed by

the ‘protostyloph’ and the protoloph of the first molar. The tooth

is fairly typical of the permanent premolars of many

diprotodontans: somewhat rectangular with a longitudinal ridge

just buccal of the centre-line bearing two major cusps, and a

shorter, lower lingual cingular ridge with an anterior expansion

that forms a prominent anterolingual corner and is the base for

a strong transverse ridge ascending to the anterior cusp. Anterior

to this ridge is a slightly weaker and shorter one, midway

between the transverse ridge and the trenchant anterior extension

of the longitudinal ridge. The longitudinal ridge is crossed by a

deep valley that separates the anterior cusp from the slightly

lower, double, posterior cusp; the crest from the hind part of this

cusp curves around to connect with the posterior end of the

lingual cingulum. There are two or three small transverse

crenulations in the cingular valley. The longitudinal crest is

slightly convex buccally, and is almost perfectly aligned with

the anterior transverse crest (the ‘parastyloph’) of M 1 .

New specimens of ektopodontids (Marsupialia: Ektopodontidae) from south australia

183

M 1 . This tooth in SAM P35309 is almost identical to the

paratype SAM P22504 described by Pledge (1986).

M 2 . In Pledge (1986), the M 2 of Ektopodon tommosi from

Lake Tarkarooloo was depicted for purposes of illustrating

morphology in Text Fig 3.5 (ibid.) as that of E. stirtoni, the

tooth having not then been found at Mammalon Hill. This

deficiency has now been rectified with discovery of specimen

SAM P23854 - an RM 2 still partly in alveolo in a fragment of

maxilla (it lacks only the buccal face of the tooth), SAM

P31637 with a complete LM 2 in most of the maxilla, and SAM

P35309, an almost complete maxillary dentition (figs. 5A-C).

This bilophodont tooth is wider than long. The protoloph is

slightly wider transversely than the metaloph. The tooth’s

length is 6.8 mm; its maximum width is 9.5 mm.

The protoloph has eight distinct cusps and an indication of

at least one more (as does the metaloph). This is more than in

E. tommosi (from the slightly older Tarkarooloo LF). The

apices of the protocone and metaconule are equal in height,

but the lingual extremity of the base of the former is more

acute and thus extends slightly farther in a lingual direction.

There is a large, deep groove on the anterior face of the

protocone and a slightly weaker posterolingual one that

extends into the pocket formed by the short but strong lingual

cingulum at the end of the transverse valley. Strong pre- and

postprotocristae extend in a slightly buccal direction (in the

same way that comparable crests extend from the metaconule)

giving a hint of remnant selenodonty. The preprotocrista

meets the lingual end of the anterior cingulum. The other pre-

cristae do not join or only just contact the precingulum. Cusp

2 is smaller and simpler than either the protocone or cusps 3 or

4. It has a single undivided longitudinal crest and is linked to

the protocone by an apical strut and a basal strut from the

postcrista. Cusp 3 has a strong pre- and postcrista, with a

somewhat weaker parallel set arising lower on the lingual

face. It is linked to cusps 2 and 4 by a fine low apical strut.

Cusp 4 is similar but the lingual pair of cristae is slightly

stronger. Cusps 5 to 8 bear undivided pre- and postcristae

which are linked by 2 or 3 subapical struts and several short

basal ribs. The extent of the median valley indicates at least

one more cusp and possibly two.

In occlusal view, only the metaconule on the metaloph is

opposite its counterpart cusp (the protocone) on the protoloph.

The metaconule is rounder than the protocone and the two

grooves diverge antero- and posterolingually, the anterolingual

groove running into the cingular pocket. The

postmetaconulecrista runs into a very narrow postcingulum to

which the other postcristae are weakly joined. Cusp 2 is large

and well-spaced from both the metaconule and cusp 3. Its

precrista bifurcates basally, with the new rib extending

lingually towards the thickened basal ends of the

premetaconulecrista and the postprotocrista. Short struts link it

anterobasally and apically to the metaconulecrista and there is

a short lingual rib from the postcrista. Cusp 3 is similar to that

of the protoloph. Cusp 4 is finer with a bifurcating precrista and

a short basal lingual rib from the postcrista. It is linked by an

apical strut to cusps 3 and 5. Cusps 5 and 6 are similar with

undivided cristae that bear a few short irregular ribs. Cusp 6

joins apically to cusps 5 and 7 with a fine strut. Cusp 7 is

irregular, having a fine wavy crista bearing several short ribs or

broken struts that link with the remnants of cusp 8.

Of the roots, only the double lingual one supporting the

protocone and metaconule is preserved, although its tapered

tip is missing. Anterior and posterior transverse roots are

represented by their bases which support the buccal ends of

the protoloph and metaloph respectively.

SAM P 23854 (fig. 5D) is similar in appearance and

construction to the M 2 referred by Pledge (1986) to E. sp. cf. E.

stirtoni ( -E. tommosi ) from the Tarkarooloo LF. It differs in

two obvious respects: (1) the relatively and absolutely greater

width of the Mammalon Hill M 2 resulting from (2) at least one

extra cusp at the buccal end of each loph. In these features, it

appears to be autapomorphic within the genus. In P 31637, by

some aberrant occlusal wear or damage, cusps 2-4 on the

protoloph and cusp 3 on the metaloph are exceedingly worn,

far more so than the other cusps.

M 3 . This tooth is represented by two specimens, SAM P

30156 (fig. 5F) and SAM P 30175 (fig. 5E), both slightly

damaged by the loss of enamel on the buccal face. The former

is the better preserved. M 3 has been described from E. tommosi,

e.g. NMV P48753, at Lake Tarkarooloo, hence a direct

comparison is possible with these two additional specimens.

The M 3 of Ektopodon stirtoni is slightly larger than that of

E. tommosi and in occlusal outline is less tapered posteriorly

since the metaloph is relatively longer and less acutely

truncated. The detailed structure of cusp ribs and struts is also

less complex. The tooth is 6.7 mm long.

The protoloph is 8.9 mm wide and has eight distinct cusps.

It is similar in most respects to that of M 2 , differing in that the

posterolingual groove of the protocone does not flow into a

pocket formed by a lingual cingulum, and in that there are

more links joining the crests of cusps 7 and 8. By the same

token, the protoloph differs from that of E. tommosi, in having

fewer subapical links and struts between the crests of cusps 5

to 8. They are similar in lacking the lingual basin.

The metaloph is 7.6 mm wide with six distinct cusps and a

buccal complex. The metaconule is situated somewhat more

buccally than the protocone, but not level with cusp 2 of the

protoloph. However, the crests of the metaconule do align with

those of protoloph cusp 2. Similarly the cristae of metaloph

cusps 2 and 3 align with those of protoloph cusps 3 and 4. The

pre- and post-cristae of cusp 2 each have an accessory crista

that is rather sinuous and arises from further down on the

lingual face, parallel to the main crista. This is similar to cusp

3 of the protoloph. Cusp 3 is a smaller version of cusp 2; both

have a basal bifurcation of the precrista. Cusp 4 is smaller

still, with both pre- and post-cristae bifurcating. The cristae of

cusp 5 do not bifurcate, but have several struts and/or ribs on

the buccal face alongside the fine apical strut linking the cusp

to cusp 6. The cristae of cusp 6 are rather zig-zag because of

the several struts linking them to cusp 5 and to the missing

buccal face. The putative apical strut linking to cusp 7 is

displaced anteriorly and there is a short simple precrista from

the apex. Parallel to this is an even shorter precrista (half the

length of that of cusp 7), which would arise from the buccal

edge of the tooth.

184

N.S. Pledge

The metaloph thus differs from that of M 2 in ways

consistent with the posterior narrowing of the tooth and the

molar gradient. It differs from that of E. tommosi in its greater

relative and absolute width, and lesser buccal angular

truncation, with consequent better development of the post-

cristae of cusps 5 to 7.

The maxilla of Chunia illuminata (QM F10641) (fig. 3A)

suggests from its alveoli that there is a steep molar gradient in

that species. The dentary of Ektopodon stirtoni (SAM P19509)

indicates that the gradient is less in this species. The new

specimens of M 3 described in this paper confirm that it is also

less than in E. tommosi, although the alveoli preserved in

maxilla P31637 are too damaged to enable tooth gradients to

be determined with confidence.

The maxilla SAM P31637 is damaged on all sides. The

root of the zygoma is split and the maxilla is broken anteriorly

through the alveolus of M 1 . The alveolus of M 3 is damaged

buccally and that of M 4 is broken across, while the medial

edge of the fragment is irregular with no part of the median

suture being preserved. While the endocranial surface of the

palatal wing of the maxilla is smooth and relatively flat

(slightly concave anteriorly), the oral (ventral) surface is

noticeable convex anteroposteriorly, rather as it is in

Phascolarctos cinereus. This characteristic is not evident in

Chunia illuminata where the palate is relatively flat.

Lower dentition. Dentary, SAM P29577, (figs. 5G, H) is

broken off just anterior to the position of M 1? which is missing.

It preserves molars M 2 4 in good but more-worn condition than

the holotype. The horizontal ramus is torted, and the tooth-

row lies at an angle of about 30° to this portion of the dentary.

M 2 . In the holotype, this tooth is incomplete, having been

reconstructed from numerous tiny fragments. Dentary SAM

P29577 presents a complete but worn M 2 , while SAM P24541

is a slightly worn hypolophid of an M 2 and SAM P31638 a

perfect M 2 crown. The following description is based on the

last, with additional comments on the others where appropriate.

The M 2 is somewhat rhomboidal in occlusal outline with

protolophid and hypolophid having about the same transverse

width. In P29577, both protoconid and hypoconid are

extremely worn with the enamel breached and the dentine

deeply excavated, but only a few of the adjacent cuspids have

been breached. In contrast, P31638 is virtually unworn.

The protolophid in P31638 has seven cuspids (eight in

P29577) with the protoconid being much larger than the

others. Its crest is oblique (at about 80°) to the lingual face.

The protoconid in P29577 is so deeply worn that it merges

with the second cuspid and the two are difficult to distinguish.

In P31638, as in the holotype SAM P19509, cuspid 2 is a

single, simple plate closely appressed to the protoconid. On

the protoconid, the anterobuccal groove is flanked buccally by

a low crest and lingually by a parallel crest that seems (in

P29577) to arise from the precristid of cuspid 2. The precristid

curves buccally to merge with the remnant of the precingulid.

Posteriorly the posterobuccal groove of the protoconid is

flanked by a fine cristid arising from the postprotocristid. The

postcristid of cuspid 2 has a short buccal rib. The enamel of

cuspids 3-6 of P29577 is breached, leaving a thicker and

higher ridge on the buccal side. Precristids of cuspids 3-5 are

simple but divide slightly as they merge with the precingulid.

Their postcristids expand slightly in the base of the transverse

valley and each has a minor buccal rib. Cristids of cuspid 6

(and 7 in P29577) are simple. The innermost cuspid (7 of

P31638 and 8 of P29577) has a bifid precristid with a basal

cuspule (homologous with the larger structure in M 1 of the

holotype) developed at the lingual corner of the precingulid.

The hypolophid parallels the protolophid. The hypoconid

has deep anterobuccal and posterior grooves separating the

bulbous buccal part from the rather sinusoidal cristid. The

precristid curves buccally and the postcristid lingually to

merge with the postcingulid. All six cuspids of P29577 have

breached enamel. Precristid 2 has a weaker buccal rib and an

expanded base; postcristid 2 is bifid at its base. On cuspid 3,

the precristid is trifid with a weak buccal rib, a stronger lingual

rib and the main crest expanded in the bottom of the transverse

valley. The postcristid is bifid at its base. Cuspids 4 and 5 are

similar, with simple undivided cristids. Cuspid 6 is complex

with three parallel anterobuccal cristids which decrease

rapidly in size lingually as they are truncated by a low,

transverse cristid in the transverse valley. There are two

parallel postcristids, the lingual one of which is shorter and

cut off by the curving end of the postcingulid. The postcingulid

is well developed.

The hypolophid (SAM P24541) has a transverse width (i.e.

normal to the lingual face) of 6.4 mm. Characteristically for

lower molars of Ektopodon spp., the lophids are oblique to the

tooth row with an acute anterolingual corner. There are seven

cuspids, the inner two being combined. The hypoconid is

large, its apex just buccal of the mid-line of the tooth. It has a

deep anterobuccal groove that swings out basally and an

almost longitudinal posterobuccal groove. The cristid obliqua

curves buccally and divides basally. The posthypocristid is

longitudinal. Cuspid 2 is on the midline of the tooth. Its pre-

and postcristids parallel those of the hypoconid and give rise

to shorter, basal supplementary cristids on the buccal side. A

notched apical strut links the cuspids. Cuspid 3 is more

complex with the precristid bifurcating and the postcristid

trifurcating. Cuspids 4 and 5 are similar, simple cuspids,

linked apically by fine struts. Their cristids do not divide.

Cuspids 6-7 are complicated in being almost inseparable but

having two diverging pre- and postcristids. The lingual-most

precristid is short and notched to produce a basal cusp that

extends as a “cingulum” along the transverse valley to the

precristid of cuspid 4. All but the penultimate postcristid

merge into the postcingulum.

This tooth fragment is similar to that of E. tommosi (NMV

P48764) but its ornamentation is less developed. It is smaller

than both that specimen and the holotype of E. stirtoni but

larger than M 3 of E. stirtoni.

M 3 . This tooth is incomplete in the holotype and poorly

known in E. tommosi. It is now represented, however, by a

complete (but worn) tooth in dentary SAM P29577 and by

hypolophid SAM P23989.

The M 3 resembles that of P29577 except for its lesser degree

of wear. Only cuspids 1-3 of the protolophid and cuspids 1-5 of

the hypolophid have been breached. There are eight protolophid

cuspids and six or seven hypolophid cuspids. The anterolingual

New specimens of ektopodontids (Marsupialia: Ektopodontidae) from south australia

185

cuspule is smaller than in M 2 but slightly larger than in M 4 . The

tooth is smaller than M 2 and less rhomboid in outline, with the

hypolophid narrower than the protolophid.

Hypolophid SAM P23989 is rectilinear and orientated

obliquely at about 75-80° to the longitudinal axis of the tooth.

The width of the hypolophid is 5.6 mm. There are seven

cuspids. The buccal-most cuspid, the hypoconid, is at about

one quarter the distance from the buccal end. It has a broad,

shallow anterobuccal groove and a shallower posterior groove.

The cristid obliqua is longitudinal but swings buccally at the

base as it joins a basal strut from the second precristid. The

posthypocristid curves lingually and joins the well-developed

postcingulum. Cuspid 2 has simple undivided pre- and

postcristids; the precristid of cuspid 3 divides halfway; and

cuspids 4 and 5 have simple undivided cristae. Cuspid 6 lacks

a postcristid but has a short precristid, parallel to that of cuspid

5, that divides basally. Cuspid 7 is displaced anteriorly and

gives rise to two slightly diverging postcristids, the lingual one

of which merges with the postcingulum. Its precristid is short

and apparently joins the last postcristid of the protolophid.

This specimen preserves the posterior root, along, tapering

transverse fang-like structure, 9.8 mm long on the lingual side.

M 4 . Three teeth represent M 4 . Apart from its high degree

of wear, which has reduced the crest angle of the protolophid

to 140° or more and almost flattened the hypolophid, SAM

P29577 from the dentary (fig. 5H) is virtually identical to the

unworn SAM P23988 and P33451 (fig. 51). Our description of

this tooth will be based on the second specimen.

Previously, M 4 was only known from E. tommosi from

Lake Tarkarooloo. The new specimens should, therefore, be

compared with NMV P48766 (Pledge, 1986, plate 3.2H),

NMV P160517 and SAM P19966. SAM P23988 is less worn

than the Tarkarooloo specimens and shows a similar occlusal

outline although with more buccal constriction at the

transverse valley. The lophids are low and broad, though

possibly higher than in E. tommosi. The primary cuspids, the

protoconid and hypoconid, stand markedly higher than their

subordinate cuspids.

The protolophid is much wider than the hypolophid and its

crest is orientated at about 60° to the lingual face. The anterior

edge of the tooth is convex and has a widening precingulid

extending from the anterolingual corner, curving backwards,

almost to the inner preprotocristid. The protolophid has seven

cuspids, as in E. tommosi. The protoconid is situated about

one third the distance from buccal end of the lophid. It is

relatively high and lacks the obvious buccal grooves present

on the protoconids of more anterior teeth. Instead, it has a

strong anterobuccal preprotocristid and a subordinate lower

cristid parallel to it on the lingual side. A pair of subparallel

postprotocristae extends posterolingually to the transverse

valley, with the lingual one having two buccally directed ribs.

Cuspid 2 is plate-like. Its precristid almost meets the

precingulum at a cuspule but instead swings buccally and

parallels it (as a rib) almost reaching the lesser preprotocristid.

The postcristid bifurcates basally. Cuspid 3 is also plate-like,

its precristid being simple and undivided and its postcristid

only thickening at the posterior end. Cuspid 4 is a weaker

irregular plate, thick anteriorly, with several minor ribs. It is

shorter than cuspid 3 and larger than cuspid 5, which is

otherwise similar although posterobuccally curved. Most

cuspids are linked by fine, deep-set, apical struts, but on cuspid

6 these are as strong as the pre- and postcristids. The precristid

is short and rather bulbous while the postcristid is finer and

longer and curves buccally to join distally at the transverse

valley the postcristids from cuspids 4 and 5. Cuspid 7 is low

and on the extreme edge of the tooth. It is linked to cuspid 6 by

a strut stronger than its pre- and postcristids. The precristid is

short and merges into the precingulum. The postcristid is little

more than a lingual cingulum with several strong basal ribs

that increase in size towards the transverse valley. No

postcristids actually cross the transverse valley.

The hypolophid is short with only five distinct cusps -

fewer than in E. tommosi. The hypoconid is rounder than the

protoconid and has a small anterobuccal groove that joins a

larger posterobuccal groove on the protoconid to form a

shallow pocket. There is no buccal cingulid. The cristid

obliqua curves almost in a semicircle to a small basal cusp in

the transverse valley, where it joins the precristid from cuspid

2. The posthypocristid curves lingually to join the short

postcingulid. Cuspid 2 is a simple cristid with no discernable

apex. Posterolingually, it joins the postcingulid at a small

cuspule. Cuspid 3 is also a simple cristid bifurcating at the

anterior end and not reaching the postcingulid. Cuspid 4 is

irregular and links with a transverse cristid in the transverse

valley. It is posteriorly short. Cuspid 5 is very low; its cristid is

irregular and buccally curved. It converges with but does not

meet ribs from the lingual cingulum and posteriorly parallels

the postcingulid.

Morphologically SAM P23988, P29577 and P33451 are

similar to the M,s of E. tommosi but the fine ornamentation of

cristids, ribs and struts is simpler. The teeth are also

marginally larger.

Remarks. Overall, these newly described teeth of E. stirtoni

confirm the distinction of this taxon from E. tommosi from the

Tarkarooloo LF and appear to represent a relatively derived

species. This is in keeping with the perceived greater age of the

Tarkarooloo LF (e.g. Woodburne et ah, 1985).

Ektopodon litolophus Pledge et al. 1999.

Holotype. SAM P30176, an isolated right M 1 (fig. 6D).

Locality. Leaf Locality (UCMP V-6213), Wipajiri Formation,

eastern edge of Lake Ngapakaldi, South Australia.

Local Fauna and Age. The Kutjamarpu Local Fauna is

estimated to be approximately early Miocene (see above).

Remarks. This unique specimen is noticeably larger than its

contemporary E. serratus, and has a simpler morphology, with

a longer, more loph-like parastyloph, suggesting that it is the

most autapomorphic member of the family.

Discussion

There are now at least five known species of Ektopodon (E.

serratus, E. stirtoni, E. litolophus, E. tommosi, a Riversleigh

taxon, and E. paucicristata)-, two or three species of Chunia

186

N.S. Pledge

(C. illuminata, C. sp. cf. C. illuminatei, C. omega)-, and the

monotypic Darcius duggani. No ektopodontid species, except

E. paucicristata and D. duggani, occurs in more than a single

local fauna, and in all but two local faunas (Tarkarooloo and

Kutjamarpu) there is only a single ektopodontid species.

Further, most are not common in their respective local faunas

and three ( E. litolophus, C. omega and the Riversleigh taxon)

are known only from a single tooth. In the faunas where more

than one species occurs, the sympatric forms are distinct in

terms of size and morphology, a situation that may indicate

ecological partitioning of species into feeding guilds. Food

preferences for these possums are unclear but possibilities

include grains, nuts and insects/grubs (Pledge, 1982, 1986,

1991).

Species of Ektopodon range in age from late Oligocene to

late Pliocene. The oldest (E. tommosi ) occurs in the

Tarkarooloo LF of late Oligocene age. Ektopodon serratus

and E. litolophus in the Kutjamarpu LF are probably Early to

Middle Miocene in age. Ektopodon ulta (Megirian et al.,

2004; fig. 6G) from the Kangaroo Well LF of the Northern

Territory may be slightly younger than E. stirtoni. The age of

the Riversleigh ektopodontid has been interpreted by Archer

et al. (1989) to be most probably Early Miocene. The youngest

reported Ektopodon is E. paucicristata (Rich et al., 2006)

from the Pliocene Whalers Bluff Formation (Dutton Way,

Portland) and Childers Cove of southwestern Victoria. This

species is a temporal anomaly, in so far as its authors regarded

it to be relatively plesiomorphic. In its short protostyloph with

few cusps, low number of deeply bifurcated cusps on the

lophs, and equidimensional molars, it resembles species of

Chunia. While the original authors considered the teeth of E.

paucicristata to be upper molars, their outline suggests they

may in fact be lower molars. Species of Chunia range in age

from Late Oligocene (Ditjimanka LF) to latest Oligocene

(Tarkarooloo LF) in age. Darcius duggani (Rich, 1986) is

known from the Early Pliocene Hamilton LF, and from the

apparently early Pleistocene Nelson Bay Formation (Rich et

al., 2006), both in southwestern Victoria. Rich (1986) regards

D. duggani to be structurally intermediate between the species

of Chunia and Ektopodon.

The fragmentary Ektopodon serratus- like ektopodontid

tooth from the Wayne’s Wok LF (Pledge et al., 1999) confirms

previous suggestions that Riversleigh’s Faunal Zone B local

faunas share taxa with the Kutjamarpu LF (Archer et al.,

1989). A Darcius-like, form, also from one of the numerous

Riversleigh localities, is noted by Long et al. (2002: 142) but is

yet to formally described.

Woodburne & Clemens (1986c) interpreted the phylogenetic

relationships of ektopodontids known at the time. Now,

Ektopodon litolophus (Pledge et al., 1999) is best regarded as a

sister taxon of E. serratus, and both may have been derived

from an E. stirtoni- like ancestor. This conclusion follows from

the observation that E. litolophus and E. serratus share the

evidently synapomorphic condition of transversely widened

lophs which also contain a relatively larger number of cusps.

Considering that both of the more derived species are apparently

contemporaneous, neither is likely to be the other’s ancestor.

Acknowledgements

I thank the owners and managers of Etadunna Station for

permission for us and our predecessors (R. A. Stirton and his

students/colleagues) to work on their land, and for their

forbearance of our activities over the years. Thomas H. Rich

allowed me to participate in his early expeditions to Lake

Tarkarooloo, which resulted in some of the material described

here. Numerous volunteers helped me in the field; notable

amongst them have been J. McNamara and J. Thurmer. A.

Tindall helped with some photography and digital processing,

and H. Hamon got the figures into presentable order.

References

Aplin, K.P. and Archer, M. 1987. Recent advances in marsupial

systematics with a new syncretic classification. Pp. xv-lxxii in: M.

Archer (ed) Possums and Opossums: Studies in Evolution. Surrey

Beatty & Sons Pty Ltd and the Royal Zoological Society of N.S.W.,

Sydney.

Archer, M., Godthelp, H., Hand S. and Megirian D. 1989. Fossil

mammals of Riversleigh, Northwestern Queensland: Preliminary

overview of biostratigraphy, correlation and environmental change.

Australian Zoologist 25: 29-65.

Callen, R.A. and Tedford, R.H. 1976. New late Cainozoic rock units and

depositional environments. Lake Frome area. South Australia.

Transactions of the Royal Society of South Australia 100: 125-167.

Lindsay, J.M. 1987. Age and habitat of a monospecific foraminiferal

fauna from near-type Etadunna Formation, Lake Palankarinna, Lake

Eyre Basin. South Australian Department of Mines Report 87/93.

Long, J. M., Archer, Flannery, T. and Hand, S. 2002. Prehistoric

mammals of Australia and New Guinea. One hundred million years

of evolution. University of New South Wales Press, Sydney. 244 pp.

Luckett, W. P. 1993. An ontogenetic assessment of dental homologies in

therian mammals. Pp. 182-204 in: Szalay F. S., Novacek M. J., and

McKenna M. C. (eds). Mammal Phylogeny: Mesozoic

Differentiation, Multituberculates, Monotremes, Early Therians and

Marsupials. Springer-Verlag, New York.

Martin, H. A. 1990. The palynology of the Namba Formation in the

Wooltana-1 bore, Callabonna Basin (Lake Frome), South Australia,

and its relevance to Miocene grasslands in central Australia.

Alcheringa 14: 247-255.

Megirian, D., Murray, P., Schwartz, L. and von der Borch, C. 2004 Late

Oligocene Kangaroo Well Local Fauna from the Ulta Limestone

(new name), and climate of the Miocene oscillation across central

Australia. Australian Journal of Earth Sciences 15 (5):701-741.

Norrish, K. and Pickering, J.G. 1983. Clay minerals; pp. 281-308 in

Soils: an Australian viewpoint. CSIRO/Academic, Melbourne.

Pledge, N.S. 1982. Enigmatic Ektopodon: a case history of

palaeontological interpretation. Pp. 479-488 in: Rich, P.V., and

Thompson, E.M. (eds), The Fossil Vertebrate Record of Australasia.

Monash University Offset Printing Unit, Clayton, Victoria.

Pledge, N.S. 1984. A new Miocene vertebrate faunal assemblage from

the Lake Eyre Basin: a preliminary report. Australian Zoologist 21:

345-355.

Pledge, N.S. 1986. A new species of Ektopodon (Marsupialia;

Phalangeroidea) from the Miocene of South Australia. University of

California Publications in Geological Sciences 131: 43-67.

Pledge, N.S. 1991. Reconstructing the natural history of extinct animals

- Ektopodon as a case history. Ch. 9, pp. 247-266 in: Vickers-Rich,

P, Monaghan, J.M., Baird, R.F. and Rich, T.H. (eds). Vertebrate

Palaeontology of Australasia. Pioneer Design Studio. Monash

University, Melbourne.

New specimens of ektopodontids (Marsupialia: Ektopodontidae) from south australia

187

Pledge, N.S., Archer, M., Hand, SJ. and Godthelp, H. 1999. Additions to

knowledge about ektopodontids (Marsupialia: Ektopodontidae):

including a new species Elctopodon litolophus. Records of the

Western Australian Museum Supplement No. 57: 255-264.

Rich, T.H.V., 1986. Darcius duggani, a new ektopodontid (Marsupialia;

Phalangeroidea) from the early Pliocene Hamilton Local Fauna,

Australia. University of California Publication in Geological

Sciences 131: 68-74.

Rich, T.H.V. and Archer, M. 1979. Namilamadeta snideri, a new

diprotodontan (Marsupialia, Vombatoidea) from the medial Miocene

of South Australia. Alcheringa 3: 197-208.

Rich, T.H., Piper, K.J., Pickering, D. and Wright, S. 2006. Further

Ektopodontidae (Phalangeroidea, Mammalia) from southwestern

Victoria. Alcheringa 30: 133-140.

Rich, T.H.V. and Rich, P.V. 1987. New specimens of Ngapakaldia

(Marsupialia: Diprotodontoidea) and taxonomic diversity in medial

Miocene palorchestids. Pp. 467-476, in: Archer M. (ed.) Possums

and Opossums: Studies in Evolution. Surrey Beatty & Sons Pty Ltd

and the Royal Zoological Society of New South Wales, Sydney.

Stirton, R.A., Tedford, R.H. and Miller. A.H. 1961. Cenozoic stratigraphy

and vertebrate palaeontology of the Tirari Desert, South Australia.

Records of the South Australian Museum 14: 19-61.

Tedford, R.H. and Woodbume, M.O. 1987. The Ilariidae, a new family

of vombatiform marsupials from Miocene strata of South Australia

and an evaluation of the homology of molar cusps in the

Diprotodontia. Pp. 401-418, in: Archer, M. (ed.) Possums and

Opossums: Studies in Evolution. Surrey Beatty & Sons Pty Ltd and

the Royal Zoological Society of New South Wales, Sydney.

Truswell, E.M., Sluiter, I.R. and Harris, W.K. 1985. Palynology of the

Oligocene-Miocene sequence in the Oakvale-1 corehole, western

Murray Basin, South Australia. Bureau of Mineral Resources

Journal of Australian Geology and Geophysics 9: 267-95.

Woodbume, M.O. 1986. Biostratigraphy and biochronology. University

of California Publications in Geological Sciences 131: 87-93.

Woodbume, M.O. and Clemens, W.A. (eds). 1986a. Revision of the

Ektopodontidae (Mammalia; Marsupialia; Phalangeroidea) of the

Australian Neogene. University of California Publications in

Geological Sciences 131: 1-114.

Woodbume, M.O. and Clemens, W.A. 1986b. A new genus of

Ektopodontidae and additional comments on Ektopodon serratus.

University of California Publications in Geological Sciences 131:

10-42.

Woodbume, M.O. and Clemens, W.A. 1986c. Phyletic analysis and

conclusions. University of California Publications in Geological

Sciences 131: 94-102.

Woodbume, M.O., MacFadden, B.J., Case, J.A., Springer, M.S.,

Pledge, N.S., Power, J.D., Woodbume, J.M. and Springer, K.B.

1993. Land mammal biostratigraphy and magneto stratigraphy of

the Etadunna Formation (Late Oligocene) of South Australia.

Journal of Vertebrate Paleontology 13: 483-515.

Woodbume, M.O., Tedford, R.H., Archer, M., Turnbull, W.D., Plane,

M.D. and Lundelius, E.L. 1985. Biochronology of the continental

mammal record of Australia and New Guinea. Special Publication

South Australian Department of Mines and Energy 5: 347-363.

Table 1. Measurements (in mm) of recently discovered ektopodontid

molar teeth. Pstl = parastyloph width.

Specimen

Number

tooth

length

Pstl

width

Posterior

width

Ektopodon stirtoni

SAM P35309

LP 3

5.5

3.4

(left maxilla)

LM 1

8.8

3.2

8.5

9.1

“

LM 2

6.9

9.3

8.7

“

LM 3

6.3

8.2

6.5

SAM P31637

LM 2

6.8

9.3+

8.7

SAM P23854

RM 2

7.9

9.5+

9.2+

SAM P30156

LM 3

6.7

8.9

7.6

SAM P30175

LM 3

6.7

8.5+

SAM P31638

lm 2

7.7

7.0

6.7

SAM P29577

rm 2

8.1

7.8

7.5

(right dentary)

rm 3

8.0

6.9

6.1

“

RM,

7.5

5.5

4.3

SAM P24541

lm 2

6.4

SAM P23989

rm 3

5.6

SAM P23988

lm 4

6.8

5.95

5.0

SAM P33451

rm 4

6.7

6.0

5.1

Ektopodon litolophus

SAM P30176

RM 1

10.6

6.0

11.7

12.1

Cimnia illuminata

SAM P29081

LM 1

5.0

1.9

4.5

4.8

Table 2. Dimensions (in mm) of cheek teeth in maxilla, Ektopodon stirtoni SAM P35309. Molar row length, 26.0; preserved molar length, 21.7;

preserved cheek row length, 23.5; half-palate width, 22.0; clear half palate, 13.2; angle of premolar crest to inner line of molar row, 65°.

‘Stylewidth’ = parastyloph width.

Tooth

Length

Stylewidth

Ant. Width

Post. Width

Ant. Cusps

Post, cusps

P 3

5.00

N/A

3.20

1.00

M 1

8.50

3.2

8.50

9.10

7.00

8.00

M 2

6.50

N/A

9.10

8.25

9.00

8.00

M 3

5.70

N/A

8.00

6.70

8.00

7.00

M 4