THE AUSTRALIAN

Entomologist

published by

THE ENTOMOLOGICAL SOCIETY OF QUEENSLAND

Volume 37, Part 3, 30 September 2010

Price: $8.00 per part

ISSN 1320 6133

THE AUSTRALIAN ENTOMOLOGIST

ABN#: 15 875 103 670

The Australian Entomologist is a non-profit journal published in four parts annually

by the Entomological Society of Queensland and is devoted to entomology of the

Australian Region, including New Zealand, Papua New Guinea and islands of the

south-western Pacific. Articles are accepted from amateur and professional

entomologists. The journal is produced independently and subscription to the journal

is not included with membership of the society.

The Publications Committee

Editor: Dr A.G. Orr Editorial Co-ordinator

Assistant Editors: Dr C.J. Burwell Dr F. Turco

Queensland Museum Business Manager

Dr F. Turco Dr G.B. Monteith

Queensland Museum (geoff.monteith@bigpond.com)

Subscriptions

Subscription are payable in advance to the Business Manager, The Australian

Entomologist, P.O. Box 537, Indooroopilly, Qld, Australia, 4068.

For individuals: A$33.00 per annum in Australia.

A$40.00 per annum in Asia-Pacific Region.

A$45.00 per annum elsewhere.

For institutions A$37.00 per annum in Australia.

A$45.00 per annum in Asia-Pacific Region.

A$50.00 per annum elsewhere.

Please forward all overseas cheques/bank drafts in Australian currency.

GST is not payable on our publication.

ENTOMOLOGICAL SOCIETY OF QUEENSLAND

Membership is open to anyone interested in Entomology. Meetings are normally held

at the CSIRO Long Pocket Laboratories, Indooroopilly at midday on the second

Monday of March-June and August-December each year. Meetings are announced in

the Society's News Bulletin which also contains reports of meetings, entomological

notes, notices of other Society events and information on Members! activities.

Enquiries relating to the Society should be directed to the Honorary Secretary,

Entomological Society of Queensland, P.O. Box 537, Indooroopilly, Qld, Australia,

4068.

Cover: This undescribed species of Myrmecoroides (Heteroptera: Miridae) is about 5

mm in length and occurs along the Great Dividing Range from southeast Queensland

to Victoria. It is found on native grasses. The species is sexually dimorphic, with fully

-winged males and short-winged females (illustrated here).

All species of Myrmecoroides are strongly ant-mimetic. This species is being described

by Gerry Cassis of the University of New South Wales and Michael Wall of the San

Diego Natural History Museum.

Illustration by Hannah Finlay.

Australian Entomologist, 2010, 37 (3): 77-92 77

A REVIEW OF TAENARIS HÜBNER (LEPIDOPTERA:

NYMPHALIDAE: AMATHUSIINAE) IN QUEENSLAND,

TOGETHER WITH FIRST AUSTRALIAN RECORDS FOR T. MYOPS

KIRSCHI STAUDINGER AND ELYMNIAS AGONDAS MELANIPPE

GROSE-SMITH (SATYRINAE)

TREVOR A. LAMBKIN

Queensland Primary Industries and Fisheries, Department of Employment, Econòmic Development and

Innovation, 665 Fairfield Road, Yeerongpilly, Qld 4105.(Trevor.Lambkin@deedi.qld.gov.au)

© The State of Queensland (through Department of Employment, Economic Development and Innovation)

Abstract

Species of the amathusiine genus Taenaris Hübner known to occur in Australia, predominantly

from Torres Strait, are reviewed and illustrated. 7. myops kirschi Staudinger is recorded for the

first time in Australia from four male specimens collected on Dauan Island, Torres Strait. A

female specimen of the satyrine Elymnias agondas melanippe Grose-Smith also collected from

Dauan Island represents the first record of this taxon from Australia. The high degree of variation

observed in the external facies of Taenaris from Torres Strait and reliable taxonomic separation

of female specimens are discussed. Taenaris-like forms of the papilionid, Papilio aegeus

ormenus Guérin-Méneville and E. a. melanippe from Torres Strait and Dauan Island respectively

are illustrated and reviewed. The form of P. a. ormenus from Torres Strait that is most similar to

Taenaris spp is identified as form ormenus Guérin-Méneville variety onesimus Hewitson.

Introduction

The genus Taenaris Hübner, 1819 contains around 25 species (Parsons 1998)

of owl butterfly which have a wide distribution from Malaysia, through the

Moluccas and New Guinea, and out to the Solomons (Corbet and Pendlebury

1992, Parsons 1998, Tennent 2002). Despite this extensive distribution, the

majority of species are confined to New Guinea and its outer islands (Brooks

1950, Parsons 1998), i.e. east of Weber’s line (Brooks 1950). The exceptions

to this are 7. horsfieldii Swainson, 1820 (Corbet and Pendlebury 1992),

which is the only species that occurs west of Weber’s line in the Sundaland

region (Malaysia, Palawan, western Borneo, southern Sumatra and Java)

(Brooks 1950), and 7. phorcas (Westwood, 1958) which is endemic to the

Bismarck and Solomon Archipelagos (Parsons 1998, Tennent 2002). South

of New Guinea, Taenaris occurs in Torres Strait and extends south to Cape

York, Australia (Waterhouse and Lyell 1914, Johnson and Johnson 1991,

Braby 2000). Brooks (1950) provided a useful map of the Malay Archipelago

which illustrates the geographical distribution and limits of the genus.

Taenaris urania (Linnaeus, 1758) from the Moluccas is the generic type

species (Waterhouse and Lyell 1914).

Taenaris are attractive, medium to large butterflies (¢¢ forewing lengths 40-

56 mm with larger QQ; Parsons 1998), predominantly white in colour but

often with extensive areas of black, grey or tan (Brooks 1950, D’Abrera

1978, Parsons 1998). They characteristically have large round hindwings,

with striking ocelli that are predominantly yellow in colour with central blue-

black pupil areas. The ocelli are particularly obvious on the verso surface and

78 Australian Entomologist, 2010, 37 (3)

are considered to contribute to their aposematic colour patterns (Merrett

1996). The wings of Taenaris butterflies are relatively fragile and tear easily.

Both sexes of many Taenaris species are highly polymorphic, so much so

that their variability can tend to overlap between some species, making

delimitation of these difficult, especially female specimens (Brooks 1950,

Parsons 1991, 1998), although Brooks (1950) indicated that structural

differences between the genital armatures of the males could be reliably used

to identify males of Taenaris spp. Based on external facies, Taenaris males

are characterised by having the inner margin of the forewing convex or

bowed near the base, curving to the tornus, with tufts of long androconial

hairs at the base of the upper side hindwing and an androconial hair-streak

along the hindwing dorsal inner margin. Some species also have androconial

scales underlying this inner marginal hair-streak (Brooks 1950, Parsons 1998,

Braby 2000). Females typically have much broader wings with a straight

forewing inner margin and lack androconial hairs and scales. Many species

are believed to form Miillerian mimetic complexes; these complexes are also

thought to incorporate, as probable Batesian mimics, pale forms of Papilio

aegeus Donovan, 1805 (Paplionidae) and Elymnias agondas (Boisduval,

1832) (Nymphalidae) (Parsons 1984, Braby 2000).

CREP sail

266?

Sco a

Prince of Wales 3 > Fe

of

PACIFIC

OCEAN

LOCALITY PLAN

(not to scale)

Map 1. Map of Torres Strait, Queensland showing positions of main or inhabited

islands with known locations for Taenaris spp indicated.

Australian Entomologist, 2010, 37 (3) 79

Despite being large showy butterflies, Taenaris are surprisingly secretive as

they most often frequent the well shaded understorey of dense primary and

secondary forest, from low to moderate elevations (Parsons 1998, Tennent

2002). In the understorey they often perch on the upper surface of vegetation,

sometimes close to the ground (Parsons 1998, Tennent 2002) and quite often

fly only when disturbed. Consequently, individuals are not often observed.

Adult Taenaris primarily feed on rotting fruit and seeping sap within the

understorey, although Parsons (1984) reported two species imbibing juices

from damaged cycad fronds and nuts. Of the species for which larval host

plants are recorded almost all of these utilise monocotyledons, with the

gymnosperm family Cycadaceae also recorded as a host (Parsons 1998,

Braby 2000). Despite Taenaris being a well known genus, the life histories of

a number of species have only just been described over the last couple of

decades (Parsons 1984, Johnson and Johnson 1991, Merrett 1996). Larvae of

many species are gregarious (Parsons 1984). Parsons (1984) considered that

Taenaris are normally continuously brooded, but reported pupal diapause

occurring during times of prolonged dry periods.

In Australia, Taenaris are primarily confined to the northernmost and eastern

islands of Torres Strait, Queensland (Map 1), and due to the remoteness of

these islands and the secretive nature of the adult butterflies, relatively few

specimens have been collected. Early historical specimens (prior to 1911) are

principally held in the Australian Museum and Museum of Victoria, although

there are a few in the British Natural History Museum (NHM), London. After

this early period there were no further specimens collected from Torres Strait

for almost 75 years until airstrips were constructed on many of the islands

and efficient travel to these islands was possible. Since the 1980s, many more

specimens of Taenaris have been collected in Torres Strait including a

specimen from the Australian mainland (Wood 1987, Lambkin and Knight

1990, Johnson and Johnson 1991, Braby 2000).

Due to the temporally disjunct collections of Taenaris in northern

Queensland, and since almost all of the knowledge relating to these

butterflies is fragmentary or unpublished, I have attempted in this paper to

pull together this data. Therefore in this work, the Taenaris spp occurring in

Queensland, primarily from Torres Strait are reviewed. This includes

presenting and discussing the history of their collection, their current known

distributions, and some taxonomic difficulties within the group. In addition,

information on their habits, prevalence and seasonality in Queensland is

provided. I also report and illustrate for the first time the Taenaris-like form

of Papilio aegeus ormenus Guérin-Méneville, 1831 variety onesimus

Hewitson, 1858 that occurs in Torres Strait. Lastly, I report and illustrate the

first records for Australia of 7. myops kirschi Staudinger, 1887 and a

Taenaris-like form of E. a. melanippe Grose-Smith, 1894, both collected

from Dauan Island, Torres Strait.

80

Australian Entomologist, 2010, 37 (3)

Australian Entomologist, 2010, 37 (3) 81

Figs 1-8 (Left). Taenaris spp males: Torres Strait, Queensland. All figures not to

scale: upperside left, underside right [forewing lengths, in mm, in square brackets].

(1-5) T. artemis jamesi: (1) Dauan Island, 21.iv.2001 [46], AIK; (2) Murray Island,

25.iv.1989 [50], AIK; (3) Dauan, 28.11.2006 [45], AIK; (4) Dauan, 25.iv.2000 [43],

AIK; (5) Dauan, 9.iii.2006 [47], AIK; (6-8) T. myops kirschi: (6) Dauan, 6.iii.2006

[48], AIK; (7) Dauan, 10.iii.2006 [47], AIK; (8) Dauan, 24.ii.2006 [49], AIK.

Abbreviations of collectors, collections and their locations are: AIK — A.l.

Knight; AJJ — A.J. Johnson; AM — Australian Museum, Sydney; ANIC -

Australian National Insect Collection, Canberra; CEM — C.E. Meyer; CGM —

C.G. Miller; CGMC - C.G. Miller collection, Lennox Head; EH - E.

Hamacek; HE - H. Elgner; IRJ = I.R. Johnson; IRJC — LR. Johnson

collection, Brisbane; JWCD — J.W.C. d’Apice; KB — K. Beattie; KBC - K.

Beattie collection, Brisbane; KH - K. Houston; MDB - M. De Baar; MDBC

- M. De Baar collection, Brisbane; MTQ — Museum of Tropical Queensland,

Townsville; MV — Museum of Victoria, Melbourne; NHM - Natural History

Museum, London; PSV — P.S. Valentine; PSVC - P.S. Valentine collection,

Townsville; QPIFC - Queensland Primary Industries and Fisheries

collection, Brisbane; SJJ — S.J. Johnson; SSB — S.S. Brown; SSBC - S.S.

Brown collection, Bowral; TAL — T.A. Lambkin; TLF - T.L. Fenner; TLIKC

- Joint collection of T.A. Lambkin and A.I. Knight, Brisbane; WWB - W.W.

Brandt.

Taenaris artemis jamesi Butler, 1877

(Figs 1-5, 11-13)

Material examined or reviewed. MAINLAND QUEENSLAND: 16, North

Queensland (NHM); 19, Lockerbie, Cape York, 8.vi.1990 SJJ (MTQ). TORRES

STRAIT: 643, 19, Darnley Island, 13.iv.1910 (14), 18.iv.1910 (14), 21.iv.1910

(1d), 22.iv.1910 (388), 18.v.1910 (12) HE (MV); 333, 229, same data except

xii. 1909 (19), 22.iv.1910 (18, 12), 18.v.1910 (18), 19.v.1910 (18) (AM); 1, same

data except 6-12.iv.1984 JWD (ANIC); 13, 12, Murray Island (NHM); 14, same

data except 11-12.v.1995 SJJ (MTQ); 333, same data except 13-17.iv.1993 SJJ &

IRJ (MTQ); 433, same data except 2.iv.1989 (388), 3.iv.1989 (18) IRJ & AJJ

(MTQ); 13, same data except 3.iv.1989 IRJ & AJJ (IRJC); 13, 19, same data except

20.iv.1989 (19), 17.iv.1993 (18) SJJ & IRJ (PSVC); 333, same data except

17.iv.1993 (14), 17.iv.1994 (25:3) PSV (PSVC); 233, same data except 4.iv.1986

MDB (MDBC); 433, 12, same data except 30.ii1.1986 (19), 22-25.iv.1989 (388),

26.iv.1996 (13) TAL (TLIKC); 13, same data except 4.iv.1986 KB (KBC); 733,

same data except 24.iv.1989 (14), 25.iv.1989 (388), 22-25.iv.1989 (388) AIK

(TLIKC); 288, same data except 6-10.iv.2001 SSB & CEM (SSBC); 16, same data

except 1-7.vi.1986 JWD (ANIC); 388, Dauan Island, 2.iv.2004 PSV (PSVC); 346,

same data except 2.iv.2004 (23.3), 3.iv.2004 (18) SJJ (MTQ); 233, same data

except 1-8.iv.2009 MDB (MDBC); 343, same data except 5.iv.2009 CGM (CGMC);

733, 5292, same data except 6.1.2006 (19), 11.i.2006 (299), 2.iv.2009 (13),

3.iv.2009 (14, 19), 5.iv.2009 (333), 6.iv.2009 (243), 6.i.2010 (12) TAL (TLIKC);

1733, 729, same data except 25.iv.2000 (18), 7.v.2000 (19), 21.iv.2001 (12),

4.i.2006 (12), 23.11.2006 (18), 24.11.2006 (18), 28.11.2006 (263), 6.ii1.2006 (18,

82

Australian Entomologist, 2010, 37 (3)

Australian Entomologist, 2010, 37 (3) ee

Figs 9-16 (Left). Taenaris spp. females, and Taenaris-like forms of Elymnias

agondas melanippe and Papilio aegeus ormenus: Torres Strait, Queensland. All

figures not to scale: upperside left, underside right [forewing lengths, in mm, in

square brackets]; (9) 7. catops turdula: Saibai Island, 1.iii.1996 [50], TAL; (10-13) 7.

artemis jamesi: (10) Dauan Island, 11.iii.2006 [51], AIK; (11) Dauan, 11.i.2006 [51

mm], TAL; (12) Dauan, 6.iii.2006 [54], AIK; (13) Dauan, 9.iii.2006 [57], AIK; (14)

E. a. melanippe female: Dauan, 4.iv.2009 [44], MDB; (15-16) P. a. ormenus females:

(15) Dauan, 6.i.2006 [65], TAL; (16) Murray Island, 1.v.1999 [60], AIK.

12), 8.iii.2006 (18), 9.ii1.2006 (633, 19), 10.41.2006 (288), 10.11.2008 (19),

11.11.2006 (13, 19), 19.xii.2009 (19) AIK (TLIKC); 19, same data except 13-

19.iv.2001 SSB (SSBC); 3, Q, Saibai Island, 20.iv.2000 AIK (TLIKC); 488,

Stephens Island, 7.v.1985 (388), 11.v.1985 (18) CGM (CGMC)./PAPUA NEW

GUINEA: 19, Lae, 20.ix.1951, WWB, determined TLF 1975 (ANIC); 19, Subitana

(Central District) 1800 ft, 11.viii.1949 WWB, ID by WWB (ANIC); 19, same data

except 16.viii.1949.

Taenaris artemis (S.C. Snellen van Vollenhoven, 1860) occurs on the

western side of New Guinea from Gebe, Waigeo, Misool, Aru, Salawati,

Mioswaar, Biak and Japen, throughout New Guinea, including the outlying

islands of Papua New Guinea and south through Torres Strait to the tip of

Cape York Peninsula, Queensland (Brooks 1950, Parsons 1998, Braby 2000).

The type locality is New Guinea (Waterhouse and Lyell 1914, Edwards et al.

2001). In Papua New Guinea it is widespread and occurs in a variety of

habitats, including primary and secondary forest, and eucalypt savannah

(Parsons 1998). Populations from southern Papua New Guinea and Torres

Strait are assigned to 7. a. jamesi (Brooks 1950, Parsons 1998). The type

locality for T. a. jamesi (originally described as Tenaris jamesi Butler) is

Yule Island, New Guinea (Butler 1876, Parsons 1998). Prior to Wood (1987)

collecting three specimens of T. a. jamesi on Murray Island in Torres Strait in

1984/85 (Map 1), all previous known specimens from Australia predated

1911. As mentioned above, these earlier specimens include a male and

female from ‘Murray Island’ housed in the NHM (the types of T. a. zetes,

Brooks 1944) and a male also in the NHM from ‘N. Queensland’, the type of

T. a. queenslandicus (Rothschild 1916). The remainder of these specimens, 5

in the AM and 7 in the MV (although Waterhouse and Lyell [1914] indicated

a total of only 11) were all collected on Darnley Island (Map 1) by Hermann

Elgner in 1909/10 (Moulds 1977, Dunn 2007).

In Queensland T. a. jamesi is primarily restricted to the islands in the

northern and eastern sectors of Torres Strait, and is known with certainty

from one female collected from Lockerbie at Cape York (Johnson and

Johnson 1991, Braby 2000) (Map 1). The female specimen purportedly

collected at Bamaga at Cape York in October 1980 by the late F.G. Sattler

(now in the S. Ginn collection, Sydney) mentioned in Braby (2000) appears

to be mislabelled and likely originated from Mumeng in Papua New Guinea.

S. Ginn (AM) has recounted the movements of Sattler over that period and

84 Australian Entomologist, 2010, 37 (3)

Sattler was not known to have collected at any location in Queensland north

of the Claudie River, although during this same period he visited Papua New

Guinea and collected other Taenaris specimens while there. Therefore based

on this evidence, Sattler’s Bamaga record for T. a. jamesi is almost certainly

erroneous. The precise collection details for the ‘historic’ male specimen

housed in the NHM labelled ‘N. Queensland’ (Brooks 1950, Braby 2000) are

unknown and it might well have originated from the area around Lockerbie.

Current Torres Strait collection records indicate that 7. a. jamesi is restricted

to Murray, Darnley and Stephens Islands in the very east of the strait, and to

Saibai and Dauan Islands in the north, close to the southern coast of the

Western Province of Papua New Guinea (Map 1). Interestingly, there has

been only one Darnley Island specimen (14, 6-12.iv.1984 JWCD) of T. a.

jamesi collected more recently than Elgner’s records from a century ago,

which attests to the secretive habits of these butterflies, although what is

believed to be this species has been observed on the island on a number of

occasions (De Baar [1988] and unpublished data; Johnson unpublished data).

By far, the majority of specimens has been observed and collected from

Dauan and Murray Islands.

In Torres Strait, 7. a. jamesi is only known during the wetter months, viz.

January to June. Little is known of the seasonality of the species in Torres

Strait as there are no collection records for this species over dry seasons,

mainly because the region is rarely visited by entomologists during the drier

periods. Some correlation might be drawn from the work of Parsons (1984),

who studied the ecology of 7. onolaus Kirsch in Papua New Guinea. He

found that females of this species were generally short lived and that the

species was continuously brooded all year round, with large fluctuations in

population numbers directly related to extremes of wet and dry periods. In

addition, Parsons (1984) found that prolonged dry periods produced diapause

in pupae of 7. onolaus. Thus in Queensland T. a. jamesi may be continuously

brooded, with populations normally increasing in numbers during the wet

season.

In Australia, 7. a. jamesi has been observed in dense primary or secondary

vine thicket often with a prominent Pandanus S. Parkinson (Pandanaceae)

component (Johnson and Johnson 1991). In Torres Strait, many specimens

have been collected quite close to habitation, in the understorey of vine

thicket, under mango trees (Mangifera indica L., Anacardiaceae) or in

overgrown or abandoned banana (Musa spp, Musaceae) gardens. When

flying in dense vegetation, they are especially adept at weaving through

undergrowth and between tree trunks which makes them difficult to capture.

On Murray and Dauan Islands females have also been observed swiftly

traversing open ground such as over roads, between two and four metres

above ground level. The species is most easily collected while resting on

foliage or imbibing from over-ripe or fermenting fallen fruit and are most

often only observed following disturbance from their perching or imbibing

Australian Entomologist, 2010, 37 (3) 85

sites. Recorded adult fruit hosts in Torres Strait are Ficus L. (Moraceae),

Terminalia catappa L. (Combretaceae), Musa L. (Musaceae) and M. indica.

Fruit feeding by adult butterflies on Dauan Island has frequently been

observed in the mornings at around 0800hrs EST and just prior to dusk, after

1730hrs EST. Johnson and Johnson (1991) recorded Pandanus as a larval

host for T. a. jamesi from material they collected on Murray Island, Torres

Strait, while in Papua New Guinea, Parsons (1984) and Merrett (1996)

recorded P. odorus Ridl. and Cocos nucifera (L.) (Arecaceae) as hosts.

Taenaris catops turdula Fruhstorfer, 1914

(Fig. 9)

Material examined or reviewed. QUEENSLAND (TORRES STRAIT): 9, Darnley

Island, 13.v.1910 HE (AM); Q same data except 18.v.1910 (MV); 9, Saibai Island,

1.iii.1996 TAL (TLIKC).

Taenaris catops (Westwood, 1851) is known from Gebe, Waigeo, Misool,

Aru, Salawati, Mioswaar, Roon, Japen, mainland New Guinea, Torres Strait

and various outlying islands of Papua New Guinea (Brooks 1950, Parsons

1998). The type locality for 7. catops is Aru (Brooks 1950, Parsons 1998).

Despite it being the most widespread and the most frequently encountered

Taenaris species in New Guinea (Brooks 1950, Parsons 1984), there are still

only three specimens known from Torres Strait (Waterhouse and Lyell 1914,

Braby 2000), i.e. from Darnley and Saibai Islands (Waterhouse and Lyell

1914, Braby 2000) (Map 1). These three specimens are similar in facies,

being predominantly white with a pair of ocelli on each hindwing (Fig. 9).

Based on the pale colouration of the two Australian specimens (Waterhouse

and Lyell 1914) known to Brooks (1950) at that time, he tentatively assigned

them to 7. c. turdula, which is principally found in the southern provinces of

Papua New Guinea; the type specimen being from Yule Island, Papua New

Guinea. Therefore its distribution encompasses the Western Province of

Papua New Guinea which is a very short distance from Saibai Island. The

Saibai Island specimen collected in March 1996 is also pale and was

subsequently assigned to T. c. turdula by Braby (2000). Nothing is known of

the habits, biology or seasonality of 7. c. turdula in Torres Strait except that

the specimen from Saibai Island was collected in March 1996 as it flew

swiftly, about two meters above ground level along the landward side of

mangroves. In Papua New Guinea, larvae of T. c. turdula feed on a range of

host plants, Cordyline terminalis (L.) (Liliaceae), Phaius tancarvilleae

(Banks ex L’Her.) Blume (Orchidaceae), Musa sp., Areca catechu L. and

Caryota rumphiana (Arecaceae) (Parsons 1984, 1998). Despite the paucity of

material collected from Torres Strait, it is possible that the species locally

occurs on Darnley and Saibai Islands, as both are largely unexplored, have

potential host plant species and are infrequently visited by butterfly

collectors. In addition, it is possible that 7. c. turdula in Torres Strait might

not readily frequent areas near habitation, which might explain its apparent

scarcity in Torres Strait compared to T. a. jamesi (although T. catops has

86 Australian Entomologist, 2010, 37 (3)

been collected in gardens on the edge of villages in Papua New Guinea, T.L.

Fenner, unpublished data).

Taenaris myops kirschi Staudinger, 1887

(Figs 6-8, 10)

Material examined) QUEENSLAND (TORRES STRAIT): 443, Dauan Island,

23.11.2006 (3), 24.ii.2006 (3), 6.iii.2006 (3), 10.iii.2006 (3) AIK (TLIKC). PAPUA

NEW GUINEA: 19, Lae, 14.vi.1951, WWB, ID by WWB (ANIC); 19 same data

except 6.ix.1951, determined TLF 1975; 12, Angoram (Sepik District) 20 ft,

26.iv.1950, WWB, ID by WWB (ANIC).

The distribution of 7. myops (C.&R. Felder, 1860) includes Waigeo, Misool,

Aru, Salawati, Mioswaar, Biak, Japen, mainland New Guinea and various

islands outlying Papua New Guinea (Brooks 1950, Parsons 1998). The type

locality is Aru (Brooks 1950, Szent-Ivany and Barrett 1956). Parsons (1998)

described T. myops as being widespread in Papua New Guinea with the race

T. m. kirschi principally occurring along the southern coast, which includes

the Western Province, directly opposite and very close to the northern Torres

Strait islands. The type locality of 7. m. kirschi is Port Moresby, Papua New

Guinea (Parsons 1998). In Papua New Guinea it primarily occurs in the

central district of Papua in eucalyptus savannah and monsoon forest (Szent-

Ivany and Barrett 1956). As its distribution encompasses the area of Papua

New Guinea adjacent to Torres Strait, it is not surprising that four males of T.

m. kirschi were collected (AIK) on Dauan Island in February and March 2006

(Map 1). My placement of these specimens into subspecies kirschi is based

solely on geographical grounds. The four specimens were all collected,

together with 7. a. jamesi, as they imbibed on fermenting mango fruit under

mango trees at the village edge. No further specimens have been observed on

Dauan Island since, despite visits by a number of butterfly workers.

Therefore it is unknown whether 7. m. kirschi is resident on the island or

whether it sporadically invades from the Papua New Guinea mainland,

although the latter seems improbable as Taenaris spp are generally secretive

and of a frail nature, and are mostly reluctant to fly great distances. At Port

Moresby in Papua New Guinea, larvae of 7. m. kirschi were recorded feeding

on Musa spp. (Szent-Ivany and Barrett 1956).

Elymnias agondas melanippe Grose-Smith, 1894

(Fig. 14)

Material examined. QUEENSLAND (TORRES STRAIT): 9, Dauan Island,

4.iv.2009 MDB (MDBC).

Elymnias agondas (Boisduval) occurs to the west of New Guinea (Seram,

Waigeo, Aru and Salawati), throughout mainland New Guinea, including

some of its islands as E. a. melanippe; and into the northern coastal area of

Cape York Peninsula, Queensland as E. a. australiana Fruhstorfer (Parsons

1998, Braby 2000). It is found primarily in or near rainforest where its host

plants Calamus spp (rattan or lawyer palms: Arecaceae) mostly grow

Australian Entomologist, 2010, 37 (3) 87

(Braby 2000). E. a. melanippe predominantly occurs in Papua New Guinea

and on its islands, Normanby, Woodlark and Daru (Parsons 1998) and its

type locality is indicated as ‘German New Guinea’ (Parsons 1998). The

species was not known from Torres Strait until April 2009 when a female of

E. a. melanippe was collected by M. De Baar on Dauan Island, Torres Strait

(Map 1) as it flew close to the village edge, near monsoonal vine forest. This

specimen constitutes the first Australian record of E. a. melanippe and the

first record of E. agondas in Torres Strait. The specimen collected was of the

form that resembles Taenaris (Fig. 14), the same form that Parsons (1998)

reported as a ‘mimic’ of various Taenaris species throughout New Guinea.

Although Calamus spp have not been observed on Dauan, other Arecaceae

occur naturally on the island (e.g. Nypa fruticans Wurmb., C. nucifera and

Ptychosperma macarthurii H. Wendland)

Papilio aegeus ormenus (Guérin-Méneville), 1831

form ormenus Guérin-Méneville variety onesimus Hewitson, 1858

(Figs 15, 16)

Material examined. QUEENSLAND (TORRES STRAIT): 12 Dauan Island,

11.v.2001 AIK (TLIKC); 292 same data except 6.i.2006, 5.iv.2009 TAL; 12 same

data except 1-8.iv.2009 MDB (MDBC); 19 Darnley Island, 1-2.iv.1987 MDB

(MDBC); 19 Murray Island, 1.v.1999 AIK (TLIKC); 19 same data except (Mer),

29.iii.-4.iv.1986 MDB (MDBC); 19 Moa Island (St Pauls Mission), 10-16.i1.1986 KH

& EH (QPIFC).

Papilio aegeus Donovan occurs across Torres Strait as two subspecies,

demarcated roughly in the central area of the strait, with the nominate race, P.

a. aegeus Donovan found on and south of the central island group (Moa,

Badu and Maubiag Islands) and P. a. ormenus predominantly occurring north

and east of this group, although specimens considered to be P. a. ormenus are

also known from the central island group (Map 1). Braby (2000) regarded the

central area of Torres Strait to be a hybrid zone for the two subspecies, and

indicated that some specimens from this central island group can be variable,

possessing intermediate characters and consequently are difficult to place. In

general, the females of both subspecies can be highly polymorphic,

particularly so for P. a. ormenus and especially in Papua New Guinea

(Parsons 1998). Hancock (1983) reviewed the systematics and biogeography

of P. aegeus (as Princeps aegeus [Hancock 1983]) and concluded that there

were primarily three female forms of P. a. ormenus, of which one form,

ormenus Guérin-Méneville, variety onesimus Hewitson is a pale Taenaris-

like morph with white forewings and darker costal and apical borders (Figs

15, 16). Similarly Braby (2000) reported three forms in general for P. a.

aegeus, with form beatrix Waterhouse being roughly analogous to the more

northern ormenus form of P. a. ormenus, although a true Taenaris-like form

(strictly comparable to variety onesimus) is not known within females of P. a.

aegeus.

88 Australian Entomologist, 2010, 37 (3)

Discussion

Parsons (1991, 1998) and D’Abrera (1978) emphasised the extreme variation

within 7. artemis, T. catops and T. myops, and because of this variation,

Parsons (1998) considered that Brooks (1950), in his review of Taenaris,

could not justify listing 20, 26 and 13 subspecies respectively of such

variable taxa. Parsons (1998) reviewed T. artemis across its range and due to

its variability he accepted only six subspecies for Papua New Guinea, and

even placed some doubt on the validity of these six, still referring to them as

‘supposed’. In Australia, because of this variability, Wood (1987) and Braby

(2000) consequently treated 7. a. zetes Brooks from Murray Island, Torres

Strait and 7. a. queenslandicus Rothschild from north Queensland as junior

synonyms of T. a. jamesi. Additionally, Parsons (1998) recognised only 11

subspecies of T. catops and just three T. myops subspecies from Papua New

Guinea.

In support of his own review of the above three species and their subspecies,

Parsons (1998) indicated that ‘Miillerian mimetic associations’ and ‘the

existence of clinal variation’ contributed to the variability shown by these

species throughout their range. Parsons (1991, 1998) indicated that the

variation in their facies was ostensibly influenced by the overall similar facies

of their mimetic Taenaris counterparts (interspecific and intraspecific), and

other mimicry ‘models’ occurring in the same geographical area. In addition,

he proposed that this particular influence on variation even caused ‘females

to closely resemble their respective males at any given locality’ (Parsons

1991). Moreover, Brooks (1950) who first reviewed the genus also suggested

that 7. myops and T. artemis ‘both conform to a typical geographical pattern’.

Therefore, Parsons (1998) considered that mimetic and clinal influences on

local populations, combined with an overall degree of natural variability of

these three species justified his reluctance to adopt many, if not all of the

geographical races, in particular those of T. artemis.

Despite Brooks (1950) and Parsons (1998) proposing that mimicry and clines

might strongly influence the variation observed in Taenaris species, Szent-

Ivany and Barrett 1956 found that the many individuals of 7. myops that they

reared on banana at Port Moresby, Papua New Guinea showed high

variability in the extent of wing colouration and in the size of the ocelli.

Similarly in Torres Strait, 7. a. jamesi and T. m. kirschi are highly variable

(Figs 1-8), even in series of specimens from small islands such as Dauan.

Thus, this high degree of variation recorded in confined geographical areas

such as Dauan Island places some doubt on the proposition by Parsons (1991,

1998) and Brooks (1950) that mimicry and clinal variation could strongly

influence the variability of these two species.

Identification of the three Taenaris species that occur in Torres Strait is

mostly easy, and males in particular of each species can be clearly delineated,

more so than the females. Males of 7. catops have typically short, stubby

Australian Entomologist, 2010, 37 (3) 89

forewings (the wings are almost square in cross section), prominent dark

scaling along the radial, medial and cubital veins of the forewing upper

(Parsons 1998), and the forewing inner margin is always devoid of dark

scaling (which the other two species almost always have) (Parsons 1998). In

addition, both sexes of T. catops can be easily separated from other members

of the genus by the position of hindwing ocelli, which in 7. catops is set in

further from the termen in the subtornal area, while in other Taenaris species,

the ocelli sit in the tornal area of the hindwing (Waterhouse and Lyell 1914).

Taenaris myops males always possess dark androconial scales underlying the

inner marginal hair-streak on the hindwing upper (Brooks 1950, Parsons

1998) (Figs 6-8), while Parsons (1998) indicated that the males of 7. artemis,

in addition to lacking the androconial inner marginal scales of T. myops, are

generally paler and more tan-brown to grey-brown in colour as opposed to

the dark brown, almost black markings of 7. myops, although some

individuals of 7. a. jamesi from Torres Strait can have grey-black colouration

(Fig. 5). Moreover, specimens of 7. a. jamesi known from Torres Strait are

highly variable in wing colouration, extent of colouration and wing shape

(Figs 1-5).

Fig. 17. Principal wing features (not including dark androconial scales) considered

useful by Brooks (1950), D'Abrera (1978) and Parsons (1991, 1998) (see Discussion)

in identifying Taenaris artemis jamesi and T. myops kirschi. All figures not to scale

[forewing lengths, in mm, in square brackets]. A: yellow basal area of upperside

hindwing; B: hindwing costal border not terminating at wing base; C: position of

forewing inner margin; D: forewing cell; E: dark-brown almost black markings; F:

absence or reduction of hindwing costal border. Specimen data, left to right: 7. a.

jamesi female upperside, Dauan Island, 9.iii.2006 [57], AIK; T. a. jamesi female

underside, Dauan, 6.iii.2006 [54], AIK; T. m. kirschi male underside, Dauan,

24.11.2006 (49), AIK; T. a. jamesi male underside, Dauan, 9.iii.2006 [47], AIK.

Among the females of the three Taenaris species in question, only the female

of T. catops (Fig. 9) can be reliably identified (see above). A number of

workers have attempted to separate females of 7. artemis and T. myops on the

90 Australian Entomologist, 2010, 37 (3)

basis of wing colouring. Brooks (1950) indicated that the female of 7.

artemis could often be distinguished from allied species by the presence of a

yellow area below the base of the upperside hindwing (Fig. 17A), and dark

forms of 7. myops could be separated from allied dark forms by the costal

border of the hindwing underside not terminating at the wing base but

expanding around into the hindwing (Fig. 17B). In addition Brooks (1950)

reported that the overall markings of 7. artemis tend to be ‘light’ in colour.

Finally in his review of the genus he summarised the difficulty in reliably

separating the two species and stated that the ‘females of some of the races of

T. artemis and T. myops so closely resemble each other that it is impossible to

describe characters which separate them’. D’Abrera (1978) listed two

characteristics that he considered peculiar to female 7. artemis, the first was

the top edge of the forewing band in the inner marginal space which, when

present, is more or less parallel with the dorsum (Fig. 17C), and secondly,

where the marginal band enters the cell it never completely fills it but leaves

a lighter space that more or less follows the shape of the vein which closes

off the cell (Fig. 17D). Parsons (1991) considered that 7. artemis could be

distinguished from 7. myops by its paler colour, ‘more tan-brown compared

to the dark-brown, almost black markings of 7. myops’ (Fig. 17E), (even

though some individuals of 7. a. jamesi from Torres Strait can be almost

black in colouration Fig. 12), and in 7. artemis by the colour not fully filling

the forewing cell (Fig. 17D), whereas it does in 7. myops. Later, Parsons

(1998) reported that the yellow sub-basal area of the upper side hindwing of

T. artemis (as Brooks [1950] reported) (Fig. 17A) was not a useful character

to distinguish 7. artemis, as the hindwing of 7. myops is also frequently sub-

basally yellow. Nonetheless, Parsons (1998) regarded the most useful

characters that distinguish females of 7. artemis from females of T. myops

were the narrower dark brown forewing inner marginal band (or absence of

forewing inner marginal band) of 7. artemis (Fig. 17C), and the hindwing

inner margin of 7. myops being narrowly bordered with dark brown (Fig.

17B), but usually completely white in 7. artemis (Fig. 17F).

Accordingly, using information provided by Brooks (1950), D’Abrera (1978)

and Parsons (1991, 1998), my assessment of the female Taenaris specimens

reviewed in this paper, predominantly from Torres Strait, and of a small

series of six female Taenaris from southern Papua New Guinea, purportedly

identified as 7. a. jamesi and T. m. kirschi (ANIC), places doubt on the

reliability of many if not all of the above characters when used to separate

either sex of the two species. Based on my assessment I found no helpful,

consistent or reliable characters (as illustrated in Fig. 17) that could be

dependably used to identify females of the two species (including the six

specimens from Papua New Guinea). In effect, for pale morphs of both

species, even male specimens would be difficult to distinguish using only

these characters. Based on this, and the fact that only four out of a total

Australian Entomologist, 2010, 37 (3) 91

of 39 male Taenaris specimens collected on Dauan Island are 7. m. kirschi, Ï

have tentatively classified all females specimens examined from Dauan

Island and the remainder examined from Saibai, Darnley and Murray Islands,

including the specimen from Lockerbie as 7. a. jamesi.

The reporting here of female P. aegeus ormenus form ormenus, variety

onesimus from Torres Strait constitutes the first recognition of this Taenaris-

like variety from Australia and here, two specimens are first illustrated from

Torres Strait. Moreover, due to the intensity of butterfly collecting

undertaken on Dauan island since the early 2000s it seems likely to surmise

that the first capture of E. a. melanippe on the island in April 2009, in

conjunction with the specimen being quite worn, likely constitutes a vagrant

from nearby Papua New Guinea. In addition, because of the capture of four

relatively fresh specimens of 7. m. kirschi from Dauan Island in 2006 and

none since, it might be possible that vagrant populations of this species from

Papua New Guinea from time to time may become established on Dauan

Island. Conversely, due to the fragility of Taenaris butterflies and their

sedentary behaviour, and the fact that many Torres Strait islands (Map 1),

including Dauan, are still largely unexplored because of the nature of their

terrain, it is feasible that E. a. melanippe and T. m. kirschi might be

established on Dauan Island.

Acknowledgements

S.S. Brown, M. De Baar, I.R. Johnson, C.G. Miller and P.S. Valentine

allowed examination and/or review of material held in their collections. The

Australian Museum, Sydney (via J.S. Bartlett), the Museum of Victoria,

Melbourne (C. McPhee), the Australian National Insect Collection (CSIRO),

Canberra (E.D. Edwards), the Museum of Tropical Queensland, Townsville

(B. Done) and the Queensland Primary Industries and Fisheries Collection

(J.S. Bartlett) provided collection records or loan of specimens. J.S. Barlett

assisted with the preparation of the map. S. Ginn (Australian Museum)

provided personal communication in regard to the likely collection details of

specimens of 7. artemis from the F.G. Sattler collection. T.L. Fenner

provided personal communication on Taenaris spp in Papua New Guinea.

Special appreciation goes to A.I. Knight for his significant contribution in

collecting Taenaris specimens in Torres Strait, in particular from Dauan

Island. I thank the local communities of Murray, Dauan and Saibai Islands

for allowing entry into their communities, supplying lodging and providing

much cooperation during my time spent on each island.

References

BRABY, M.F. 2000. Butterflies of Australia: their identification, biology and distribution.

CSIRO publishing, Collingwood, Victoria; xx + 976 pp.

BROOKS, C.J. 1944. New species and forms of Tenaris (Lepidoptera: Amathusiidae).

Proceedings of the Royal Entomological Society London (B) 13: 109-116.

92 Australian Entomologist, 2010, 37 (3)

BROOKS, C.J. 1950. A revision of the genus Tenaris Hubner (Lepidoptera: Amathusiidae).

Proceedings of the Royal Entomological Society London 101: 179-238.

BUTLER, A.G. 1876. Descriptions of Lepidoptera from New Guinea with notice of a new

genus. Proceedings of the Zoological Society London 1876: 765-768.

CORBET, A.S. and PENDLEBURY, H.M. 1992. The Butterflies of the Malay Peninsula. 4th

edition, revised by J.N. Eliot. Malayan Nature Society, Kuala Lumpur; 578 pp.

D’ABRERA, B. 1978. Butterflies of the Australian Region. 2nd Edition. Lansdowne, Melbourne;

415 pp.

DE BAAR, M. 1988. Insects collected during a trip to Torres Strait 27 March to 10 April, 1987.

News Bulletin of the Entomological Society of Queensland 15(9): 107-117.

DUNN, K.L. 2007. The geographical movements of Hermann Elgner (d. 1913): a lepidopterist

based in Torres Strait, Australia. Calodema 10: 19-34.

EDWARDS, E.D., NEWLAND, J. And REGAN, L. 2001. Lepidoptera: Hesperioidea,

Papilionoidea. Jn Wells, A. And Houston, W.W.K. (eds), Zoological catalogue of Australia. Vol.

31.6. CSIRO Publishing, Collingwood; x + 616 pp.

HANCOCK, D.L. 1983. Princeps aegeus (Donovan) and its allies (Lepidoptera: Papilionidae):

systematic, phylogeny and biogeography. Australian Journal of Zoology 31: 771-797.

JOHNSON, S.J. AND JOHNSON LR. 1991. Notes on the life history of Taenaris artemis

(Snellen Van Vollenhoven) (Lepidoptera: Nymphalidae) in Australia and the rediscovery of 7. a.

queenslandica Rothschild. Australian Entomology Magazine 18: 85-86.

LAMBKIN, T.A. and KNIGHT, A.I. 1990. Butterflies recorded from Murray Island, Torres

Strait, Queensland. Australian Entomological Magazine 17: 101-112.

MERRETT, P.J. 1996. Life histories of three Taenaris species (Nymphalidae: Amathusiinae) in

Papua New Guinea. Journal of the Lepidopterists’ Society 50: 261-268.

MOULDS, M.S. 1977. Bibliography of the Australian Butterflies (Lepidoptera: Hesperioidea

and Papilionoidea) 1775-1973. Australian Entomological Press, NSW. 239pp.

PARSONS, M.J. 1984. Life histories of Taenaris (Nymphalidae) from Papua New Guinea.

Journal of the Lepidopterist’s Society 38: 69-84.

PARSONS, M.J. 1991. Butterflies of the Bululo-Wau Valley. Handbook Wau Ecological

Institute. No 12: 280 pp, 25 pls, 23 figs.

PARSONS, M.J. 1998. The butterflies of Papua New Guinea: their systematics and biology.

Academic Press, London; xvi + 736 pp, xxvi + 136 pls.

ROTHSCHILD, W. 1916. Notes on Amathusiidae, Brassolidae, Morphidae, etc. with

descriptions of new forms. Novitates Zoologicae 23: 299-318.

SZENT-IVANY, J.J.H. and BARRETT J.H. 1956. Some insects of banana in The Territory of

Papua and New Guinea. Papua and New Guinea Agricultural Journal 11: 40-44.

TENNENT, W.J. 2002. Butterflies of the Solomon Islands: Systematics and biogeography.

Storm Entomological Publications, Dereham, Norfolk; 413 pp, 90pls.

WATERHOUSE, G.A. and LYELL, G. 1914. The butterflies of Australia. Angus and Robertson,

Sydney; vi + 239 pp.

WOOD, G.A. 1987. The butterflies of Murray Island, Torres Strait, Queensland. Australian

Entomological Magazine 14: 39-42.

Australian Entomologist, 2010, 37 (3): 93-104 93

THE AESTIVATION SITES OF BOGONG MOTHS, AGROTIS

INFUSA (BOISDUVAL) (LEPIDOPTERA: NOCTUIDAE), IN THE

SNOWY MOUNTAINS AND THE PROJECTED EFFECTS OF

CLIMATE CHANGE.

KEN GREEN

Snowy Mountains Region,

NPWS, PO Box 2228, Jindabyne, NSW 2627.

(email: kenneth.green@environment.nsw.gov.au)

Abstract

Bogong moths constitute a keystone species in the ecology of the alpine zone of mainland

Australia with an éstimated 2.2 billion moths migrating to the Snowy Mountains annually.

During spring they are found in temporary camps and in aestivation sites in boulderfields and

among complex rock tors, with most moths moving to the higher altitude sites as summer

progresses. Aestivation sites have a higher relative humidity than in nearby meteorological

screens but similar temperatures. Crevices in boulderfields and among rock tors are open to heat

exchange with the atmosphere during the summer and consequently their temperatures fluctuated

in step with the fluctuations in the screens, but with the dampening of extremes in the aestivation

sites. This meant that the average daily temperature in aestivation sites falls with increasing

altitude at about 0.7°C per hundred metres of ascent, which approximates the regional lapse rate.

As a consequence, questions about the future viability of aestivation sites for bogong moth

populations (and that of their obligate parasites and predators) under a future climate change

scenario can be answered from general climate models.

Introduction

Larvae of bogong moths Agrotis infusa (Noctuidae) feed in the western plains

of eastern Australia from the Darling Downs in Queensland, south to the

north-western plains of Victoria (Fig. 1). There is no regular monitoring of

numbers of larvae despite their occasional outbreaks as pests, when they are

believed to damage winter cereal crops (Gregg et al. 1994). The moth can be

a univoltine or bivoltine species, with the cutworm larvae feeding on annual

plants which are unavailable over summer. Because of this, most of the adults

of the spring generation migrate to aestivate gregariously in the Australian

Alps (Common 1954). The numbers of moths involved in this annual

migration are not known, but Green (2010) calculated that total annual

mortality of moths in the 1400 km? above 1500 m asl in the Snowy

Mountains from predation, and non-predator-related causes such as

parasitism and weather amounted to 1015 million moths. These contributed

approximately 5000 GJ of energy annually together with 7 t of nitrogen and 1

t of phosphorus to the mountains, emphasising their importance as a keystone

species in the ecology of the alpine zone.

A key to our understanding of the ecology of bogong moths in an uncertain

climatic future is their summer distribution. Common (1954) tracked the

changes in numbers at one site and recorded regularly used aestivation sites

in the Brindabella Ranges. There are also a number of anecdotal records of

locations and Common (1954) suggested that the frequency of the word

‘bogong’ in place names meant that sites where bogong moths congregated

94 Australian Entomologist, 2010, 37 (3)

were likely to be quite common. The present study set out to examine the

distribution of sites used by bogong moths in the Snowy Mountains above

1500 m asl (Fig. 2). This area is important because it contains the highest

known sites and hence the ones least likely to be rendered unsuitable by

climatic warming. A further aim was to examine the characteristics of sites to

test their resilience in the face of future warming.

Queensland

Goondiwindi

e a

Great 4

Dividing 39

Rang

Ay f

r Victorian Alps 4: ^

" Oct .- Tingaringy

` O Mt. Buller

Fig. 1. Map of south-eastern Australia showing the self mulching soils (shaded) which

are thought to be the important breeding areas for bogong moths, together with

locations where larvae were collected by Froggatt (1900) (open triangles), Common

(1954) (crosses), Green (2008) (closed circles), and the sites outside of the Snowy

Mountains mentioned in the text. (After Common 1954).

Methods

Distribution of moths

In the summers of 2000/01 through to 2009/10, all large complex rocky

outcrops, blockstreams and blockfields in the Snowy Mountains above 1500

m asl were searched for evidence of bogong moths. Sites were visited on foot

with a final survey by helicopter to ensure that no obvious sites were missed.

Sites in the present study were recorded as either an aestivation site used in

Australian Entomologist, 2010, 37 (3)

+

Cabramurra

Dargal Mountain

Perisher?

u

osciuszko `

Rs

Chaflotte Pass

95

high summer or temporary

camps, following the classi-

fication of Common (1954)

of ‘camps’ and temporary

camps. Locations of sites

were recorded with a hand-

held GPS. The aspect of the

aestivation crevices was re-

corded using a hand-held

compass.

Statistical examination of the

aspect was undertaken using

the Rayleigh test (Batschelet

1981).

Movement of bogong moths

out of temporary camps and

into the higher aestivation

sites is difficult to map

because it is difficult for a

person to observe inside most

bogong moth sites. However,

foxes are much smaller, and

are driven by their need for

food, so examination of their

scats at different altitudes

facilitated this monitoring. A

study of fox diet on alpine

and subalpine transects was

under-taken from January

1996-December 1998 by

collecting scats monthly

(Green 2003). The com-

position of scats was studied

and the proportion of bogong

moths present in them was

determined for two complete

Fig. 2. Map of the contiguous

area above 1500 m altitude in the

Snowy Mountains showing moth

aestivation sites (squares) and

temporary camps (closed circles).

96 Australian Entomologist, 2010, 37 (3)

summer seasons, 1996/97 and 1997/98. These data were plotted to examine

the relative abundance of bogong moths monthly at different altitudes.

Temperature and relative humidity

Temperature loggers and temperature/RH loggers (Tinytag plus and Tinytag

Extra -Gemini Data Loggers, Chichester England) were deployed from 2001

onwards in a number of bogong moth camps and caves. Loggers were placed

above ground level in free flowing air in areas where moths congregated.

This was confirmed when moths were disturbed from behind the loggers at a

number of sites when the loggers were retrieved.

60

50

30

Percentage of remains in scats

Month

Fig. 3 Mean proportion of individual fox scats occupied by bogong moth remains at

subalpine (open symbols) and alpine altitudes (closed symbols) over two summer

seasons, 1996/97 (solid lines; diamonds) and 1997/98 (broken lines; circles).

Results

Distribution of moths

Within the study area in the Snowy Mountains there were 92 bogong moth

temporary camps and 42 aestivation sites (Fig. 2). Outside the study area,

bogong moth aestivation sites were found in isolated sites at Mt. Tingaringy

(37° 00’S. 148° 40’E) to the south-east of the Snowy Mountains on the

Victorian Border, at the Bogong Mountains to the north-west, and on the

peaks constituting the border with the Australian Capital Territory to the

north east (Fig. 1), with a camp on Dargal Mountain (Fig 2). Most bogong

moth sites were found along the highest ridge-line of the Kosciuszko Main

Range, with a group of outliers around Gungartan and mainly individual

Australian Entomologist, 2010, 37 (3) 97

outliers elsewhere. Sites were generally complex rock tors or periglacial

boulderfields and boulderstreams. Elsewhere, where rock outcropping was

rare, for example to the north of Jagungal and southwest along the Grey Mare

Range, there were no sites. The Crackenback Range, running along the

northwestern side of the Thredbo Valley from east of Perisher Valley through

to the South Ramshead contained outcroppings along the edge of the steep

drop-off to the Thredbo Valley. Many of these outcroppings appeared

suitable as aestivation sites or at least camps, but held no evidence of present

or previous occupation by bogong moths.

The aspect of the aestivation sites was dependent upon geomorphology and

the alignment of suitably exposed fragmented rock. For the test of preferred

aspect, the relevant statistic for the Rayleigh test is w. If w > 3.0, then the null

hypothesis of no clumping of bearings would be rejected. For the aestivation

sites w = 0.138, and for the temporary camp sites w = 1.792. Thus both null

hypotheses of no clumping were accepted. Because of the lack of clumping

the calculation of mean bearing is meaningless.

No fox scats were collected in the alpine zone in November because, with

melting snow still on the transect, their date of deposition (winter or spring)

could not be ascertained. Bogong moth remains in scats peaked in January,

thereafter there was a general decline during February (which at the Thredbo

Automatic Weather Station has the highest mean maximum and minimum

temperatures and is also the driest month). After February, bogong moths

were not common in scats at subalpine altitudes but increased in frequency in

scats at alpine altitudes (Fig. 3)

Temperature

At Charlotte Pass (1755 m asl), over the period November 2009 to March

2010, the average temperature in the boulder field (12.3+2.9°) was not

significantly different (t= 0.787, df=118, p=0.433) from that in the Bureau of

Meteorology Stevenson Screen (12.2+3.4°). However, at the South

Ramshead site (1950 m asl) over the period December to. March, the

temperatures in the cave and Stevenson Screen were significantly different in

both 2002/03 (t= 2.849, df=95, p<0.01) and 2003/04 (t= 4.692, df=106,

p<0.0001). Though significant, this may not be biologically important with

the temperature in the cave differing little from that in the Stevenson Screen

in 2002/03: 11.9£3.5° against 12.2+4.4°, and in 2003/04: 10.7+3.6° against

11.2+4.3°, Corrected for a temperature lapse rate of 0.77 °C per 100 m

(Galloway 1988), these differences would be equal to a difference in altitude

of 40 and 65 m respectively.

Over the period November 2009 to March 2010, there was a significant

difference (t= 16.843, df=104, p<0.0001)) between temperatures at the

Charlotte Pass Stevenson Screen (12.3+3.5°), and the bogong moth

98 Australian Entomologist, 2010, 37 (3)

Average daily temperature °C

Dec Jan Feb

Month

Fig. 4 Average daily temperatures over the period November 2009 to March 2010 in

the Bureau of Meteorology Stevenson Screen at Charlotte Pass, 1755 m (solid line)

and the bogong moth aestivation site on Mt Kosciuszko, 2210 m (broken line).

aestivation site on Mt Kosciuszko at 2210 m asl (8.6+2.5°). Although these

were significantly different, the plot of the two locations suggests that the two

were following the same general weather patterns (Fig. 4), with differences

mainly related to the air temperature lapse rate. In fact, when all locations

were plotted there was a highly significant negative relationship between

average temperature and altitude (r° = 0.943, p<0.0001) (Fig. 5), and the

expression (0.0068x) in the equation of the line y = -0.0068x + 24.278

closely approximated the expected lapse rate of 0.0077x where x is the

altitude in metres (Galloway 1988).

The winter temperatures showed differences between sites among rock tors

on the South Ramshead and in boulderfields on the flanks. In the period

leading into the full winter snowcover, the sites fluctuated in reasonable

synchrony (Fig. 6). However, once the snowcover blanketed the boulderfields

the air space between boulders was disconnected from the ambient air and in

the second half of winter there was no fluctuation in the temperature trace.

Australian Entomologist, 2010, 37 (3) 99

12

L

2

3

fou

Q

E

L

$h 10

g

2

<

9

1800 1900 2000 2100 2200

Altitude (m)

Fig. 5 Regression of average daily temperature from November 2009 to March 2010

in bogong moth aestivation sites on altitude. The formula of the line is y =-0.0068x +

24.278.

Average daily temperature °C

Month

Fig. 6 Average daily temperatures in a boulderfield at 1950 m (solid line) and rock

tors at 2015 m (broken line) on the South Ramshead over the period when obligate

parasites of bogong moths are independent of their hosts

By contrast, because of the architecture of the upright boulders in the tors,

there was no such disconnection in the sites in these tors. Summer

temperature data from the north and south aspects of the South Ramshead

and Mt Townsend showed opposite trends. For South Ramshead, the bogong

100 Australian Entomologist, 2010, 37 (3)

moth caves on the northerly aspect were warmer (10.7+3.5°) than the south

facing caves (10.4+3.4°), a significant difference (t= 2.200, df=104, p<0.05).

By contrast, for Mt Townsend the bogong moth caves on the southerly aspect

were warmer (9.5+3.1°) than the north facing caves (9.3+3.4°), again a

significant difference (t= 2.174, df=104, p<0.05). However, there were slight

altitudinal difference between the sites on north and south (80 m for Mt

Townsend and 25 m for the South Ramshead) and when this was corrected

for using a lapse rate of 0. 77° per 100 metres (Galloway 1988), there was no

significant difference between aspects on the South Ramshead (t= 433,

df=104, p= 0.666) whereas on Mt Townsend the northerly aspect was

significantly warmer than the south (t= 5.301, df=104, p< 0.0001).

Relative humidity

Over the period November 2009 to January 2010, there was a significant

difference (t= 3.621, df=51, p<0.001) between mean daily relative humidity

at the Charlotte Pass Stevenson Screen (64.5+15.2%), and the boulderfield

moth aestivation site at the same location (70.2+17.3%). Over the period

November 2009 to March 2010, there was a significant difference (t= 11.738,

df=71, p<0.0001) between mean daily relative humidity at the Charlotte Pass

Stevenson Screen (66.7+14.2%), and the moth aestivation sites at Mt

Townsend (75.9+14.4%), a similar figure to that recorded on the South

Ramshead northerly aspect cave from November 2001 to March 2002

(75.3+14.4%).

Discussion

Distribution

The apparently suitable but unoccupied rock outcrops on the Crackenback

Range may reflect a preference by bogong moths for higher, cooler sites,

even though moths were found elsewhere at lower altitudes such as at about

1200 m on the north side of Mt Tingaringy. However, it might also be

possible that along the length of the Crackenback Range, until South

Ramshead is reached, most of these sites have rolling terrain to the north and

west and the outcrops are not distinct landmarks except from the south. If

moths find the sites visually these might not stand out on dark nights. The

whole question of how bogong moths find their chosen sites and the roles and

relative importance of visual and olfactory cues is one that requires some

attention.

Common (1954) found large aggregations on south-western slopes and none

on apparently suitable sites on north-easterly slopes, and Blakers (1980)

suggested that aestivation sites were preferentially located on the south side

of peaks. Common (1954) did speculate that the chosen aspect in the

Brindabellas may be a result of the availability of outcrops and the present

study showed that throughout the Snowy Mountains overall, there was no

fixed aspect on which bogong moths congregated. The location of aestivation

Australian Entomologist, 2010, 37 (3) 101

sites and camps was more a result of geomorphology than choice of aspect.

The preference for higher sites, when available, appears to occur seasonally,

where through incremental upwards movement, or because more room

becomes available due to attrition of populations over the summer period

through mortality or migration, moths moved to higher aestivation sites

particularly from February onwards.

Numbers

There has been little information on the summer distribution of bogong moths

in the past, outside of the Brindabella Range (Common 1954). The

difficulties in calculating total numbers in aestivation sites from the unknown

proportion that are visible, means that there have been no attempts to date to

calculate a total number of bogong moths migrating annually. From a process

of allocating mortality between predation, parasitism and weather-related

mortality, Green (2010) calculated a total mortality of 985 million bogong

moths in the Snowy Mountains annually. Blakers (1980) estimated that in a

‘good year’ for moth numbers there would be a 45% reduction in numbers of

bogong moths while in aestivation sites. Based on this, a first order estimate

of the number of moths migrating to the Snowy Mountains annually would

be 2.2 billion. Allowing for early season mortality of about 285 million

moths, when moths were mainly occupying camps (Green 2010), the

estimated number of bogong moths in the Snowy Mountains at its peak

(around 1 January, see Common 1954), would be about 1.9 billion moths. At

a density of 17 000 moths m? (Common 1954) this would require 112 000

m? of usable rock face in the main aestivation sites at peak abundance. If

these were spread evenly across the 42 aestivation sites recorded here, there

would be about 45 million moths per site. On the wall of Common’s

observation cave were an estimated 144 000 moths (Common 1954). Bennett

(1834) stated that, ‘the quantity of moths which may be collected from one of

the granite groups, it is calculated would amount to at least five or six

bushels’ (180-220 litres). This would translate to about 200 000 moths at 100

per litre. These are the figures for the easily observed (and captured) numbers

and represent less than half a percent of the total calculated per site. These

easily counted moths, that by definition occupy the most exposed, and

possibly least preferred areas of suitable wall (Common 1954) would

probably not be of great value in calculating numbers of the unknown

proportion within aestivation sites. Hence the difficulties in calculating

annual variation in total numbers of migrating moths appear to be intractable

at present.

Temperature and relative humidity

Crevices in tors and boulder fields act like dynamic caves, that is, air flows

through them as part of the general air circulation with almost instantaneous

warming and cooling (Geiger 1965; Harris and Pedersen1998). This was

demonstrated for a cave on the South Ramshead, where comparisons between

102 Australian Entomologist, 2010, 37 (3)

logger temperatures in the boulders followed temperature recorded in a

nearby Stevenson screen differing only by 0.3° through summer of 2002/03

and 0.5° in 2003/04. At the Charlotte Pass boulder fields the match was much

closer (0.1°). There will be a varying relationship between measured

temperatures, perhaps dependent on the exact placement of the loggers that

cannot be standardized in the same way as measurements in a Stevenson

screen. Regardless of this, as temperatures fell with altitude so did the

temperatures in aestivation sites, and at about 0.7°C per hundred metres of

ascent it fell at close to the lapse rate calculated by Galloway (1988).

More uniform temperatures and higher humidity are characteristic of caves

(Geiger 1965). Relative humidity in caves in the present study also changed

with the general air circulation patterns, rising and falling with that recorded

in a meteorological screen. However, the relative humidity in caves was

about 10% higher in moth aestivation sites. This would probably be higher

still where the moths congregate, overlapping like tiles and further restricting

moisture transport to the circulating air (Common 1954). This could be

important in reducing water loss by the moths during aestivation because,

when moths, en masse, leave the caves to drink they may be more exposed to

predators (Green pers.obs.).

Climate change

Concerns have been expressed as to the impacts of climate change on bogong

moths and their predators, particularly the mountain pygmy-possum

(Burramys parvus) (Heinze et al. 2004). The major predators on bogong

moths in the Snowy Mountains are little ravens (Corvus mellori), bush rats

(Rattus fuscipes), Richard’s pipits (Anthus novaeseelandiae) and foxes

(Vulpes vulpes) (Green 2010). Although not an important predator on the

moths, B. parvus is the only predator that is dependent on them, in fact, high

altitude populations of B. parvus would not survive without access to

migratory bogong moths as a food resource (Heinze et al. 2004). Two species

of mermithid worms that parasitise bogong moths Amphimermis bogongae

and Hexamermis cavicola (Welch 1963) are also dependent on them

(Common 1954). These worms emerge from the moths in January and

February, causing their deaths. They remain in the caves over winter and re-

infest moths in spring. For these species that remain in the moth sites, the

wintering temperature is also important. In this case the temperatures in the

two architectures (boulderfields and caves in tors) do not correspond as well

as they do in summer. Initially, after the moths have left, temperatures are

similar but as snow begins to accumulate they become disconnected, and

with full snowcover the winter temperature in the boulderfields is fully

disconnected from ambient temperature fluctuations and remains steady at

just above 0°C (Fig. 6) while the caves in tors still maintain a connection

with outside temperature because of the upright, more open, architecture.

This, however, may have little impact on the overwintering worms if they are

Australian Entomologist, 2010, 37 (3) 103

buried deep in the moth detritus on the floor of caves, and B. parvus is

unlikely to remain in the caves in tors over winter, probably leaving for lower

altitude boulderfields once the moths have gone.

Migration accompanied by diapause may be essential for the survival of

bogong moths (Common 1954). Without the refuges in boulderfields and

caves, this may not be possible. The present study shows that while the

bogong moths occupy the highest possible altitude within their range, and

there is an upward movement through the summer season as conditions

become warmer and drier, there is still some flexibility because it is the

lowest altitude that they can occupy that is important in determining their

requirements. While loss of some lower altitude sites may have an impact on

the numbers that can aestivate, the loss of progressively higher sites with

climate warming will take some time. There will be a varying relationship

perhaps in different locations between temperature inside and outside of

aestivation sites but, as global temperatures rise so, obviously, will the

temperatures in boulderfields. Because the aestivation sites generally follow

air temperature they will become unusable at the same rate as regional

warming so that a worst case scenario of warming in alpine areas of 2.9°C by

2050 (Hennessy et al. 2008) will lead to a loss of aestivation sites in the

lowest 400 m of their range unless moths are able to adapt. Taking the Mt.

Tingaringy site (1200 m) as a possible minimum altitude for aestivation,

means that most sites above 1600 m should still prove suitable for aestivation

and this includes all of the sites recorded in the present study.

Conclusion

Crevices among boulderfields and rock tors provide a dark, cool, relatively

stable and moist environment necessary for aestivation and the long-term

survival of bogong moths in high numbers. The locations where moths

congregate are far enough from the outer surface of the rocks for aspect not

to be important. Because the temperature in aestivation sites is dependent on

shade air temperature, higher sites are cooler in accord with the regional

temperature lapse rate. Most sites used in the Snowy Mountains are above

1700 m altitude and even in a worst-case climate change scenario to 2050

should be proof against becoming unsuitable for moths.

Acknowledgments

I thank Ted Edwards (CSIRO) and the late Ian Common for discussions on

bogong moths and Col for piloting the helicopter. Ted Edwards and Mel

Schroder commented on the manuscript.

References

BATSCHELET, E. 1981. Circular Statistics in Biology. Academic Press, Orlando.

BENNETT, G. 1834. Wanderings in New South Wales, Batavia, Pedir Coast, Singapore and

China. Richard Bentley, London.

104 Australian Entomologist, 2010, 37 (3)

BLAKERS, R. 1980. Insects in the alpine zone. (Honours thesis) Department of Zoology,

Australian National University, Canberra.

COMMON, I.F.B. 1954. A study of the ecology of the adult bogong moth Agrotis infusa

(Boisd.) (Lepidoptera: Noctuidae), with special reference to its behaviour during migration and

aestivation. Australian Journal of Zoology 2: 223-263.

FROGGATT, W.W. 1900. Caterpillars: The bugong moth (Agrotis infusa Boisd.). The

Agricultural Gazette of New South Wales 10: 1252-1256.

GALLOWAY, RW 1988. The potential impact of climate changes on the Australian ski fields.

In: Greenhouse, planning for climate change. (Ed. G.I. PEARMAN). pp. 428-437. CSIRO

Publishing, Melbourne.

GEIGER, R. 1965. The Climate near the Ground (Fourth German edition). Harvard University

Press, Cambridge, Massachusetts.

GREEN, K. 2003. Altitudinal and temporal differences in the food of foxes (Vulpes vulpes) at

alpine and subalpine altitudes in the Snowy Mountains. Wildlife Research 30: 245-253.

GREEN, K. 2008. Migratory bogong moths (Agrotis infusa) transport arsenic and concentrate it

to lethal effect by gregariously aestivating in alpine regions of the Snowy Mountains of

Australia. Arctic, Antarctic and Alpine Research 40: 74-80.

GREEN, K. 2010. The transport of nutrients and energy into the Australian Snowy Mountains by

migrating bogong moths Agrotis infusa. Austral Ecology (in press)

GREGG, P.C., FITT, G.P., COOMBS, M. and HENDERSON, G.S. 1994. Migrating moths

collected in tower-mounted light traps in northern New South Wales, Australia: influence of

local and synoptic weather. Bulletin of Entomological Research 84: 17-30.

HARRIS, S.A. and PEDERSEN, D.E. 1998. Thermal regimes beneath coarse block materials.

Permafrost and Periglacial Processes 9: 107-120.

HEINZE, D, BROOME, L and MANSERGH, I. 2004. A review of the ecology and conservation

of the mountain pygmy possum Burramys parvus. In: The biology of Australian possums and

gliders. (Eds R.L. GOLDINGAY and S.M. JACKSON). Pp. 254-267 Surrey Beatty & Sons,

Chipping Norton.

HENNESSY, K.J, WHETTON, P.H., WALSH, K. SMITH, IN, BATHOLS, J.M,

HUTCHINSON, M., and SHARPLES, J. 2008. Climate change effects on snow conditions in

mainland Australia and adaptation at ski resorts through snowmaking. Climate Research 35:

255-270.

WELCH, H.E. 1963. Aphimermis bogongae sp. nov. and Hexamermis cavicola sp. nov. from

Australian bogong moth Agrotis infusa (Boisd.) with a review of the genus Amphimermis

Kaburaki and Immura, 1932 (Nematoda: Mermithidae). Parasitology 53: 55-62.

Australian Entomologist, 2010, 37 (3): 105-112 105

A REVIEW OF GYNANDROMORPHISM IN THE GENUS

ORNITHOPTERA BOISDUVAL, (LEPIDOPTERA: PAPILIONIDAE)

J. E. NIELSEN

GPO Box 858, Canberra, ACT 2601

Abstract

A review of records of gynandromorph specimens in the genus Ornithoptera Boisduval is

presented, with the causes of gynandromorphism in insects briefly discussed. Most Ornithoptera

gynandromorphs are known from O. priamus (Linnaeus), which probably reflects the number of

specimens available as opposed to an unusual tendency towards gynandromorphism in this

species. Gynandromorphs are also known of O. croesus lydius (Felder), O. victoriae regis

(Rothschild), O. goliath Oberthiir, and the Australian taxa O. priamus pronomus (Gray), O.

richmondia (Gray) and O. euphorion (Gray).

Introduction

A gynandromorph is an organism whose genotype simultaneously expresses

aspects of both male and female morphology in the phenotype.

Gynandromorphism is generally attributed to genetic errors associated with

cell division, with different errors at different stages of development believed

to produce different types of gynandromorph (Pereira et al. 2003, Richards

and Davies 1977, Wigglesworth, 1972). Gynandromorphs are exceedingly

rare in nature and are only obvious where there is strong sexual dimorphism.

The phenomenon has been most commonly observed in insects, where the

phenotypic expression of sexual difference is not mediated by the prevailing

endocrine environment. Gynandromorphism has also been recorded in birds,

where other processes have been postulated, including the suggestion that sex

chromosome genes acting within individual cells directly contribute to sex

differences in cell function (Agate et al. 2003).

The precise mechanism leading to gynandromorphism is not well understood

in butterflies, in which sex is determined by a WW/WZ system, with the

heterogametic sex (ie WZ) being the female, the reverse of the condition

found in mammals and most other insects. It is better understood for the

horka mutation of the vinegar fly Drosophila melanogaster (Meigen, 1830)

(Szbad et al. 1995). All known mechanisms of Drosophila

gynandromorphism rely on the zygote (fertilised egg that has not yet

undergone division or cleavage) having an initial chromosome component of

X'X (male), and subsequent loss of the X chromosome.

In D. melanogaster, the horka mutation produces gynandromorphs due to

nondisjunction, where chromosome inheritance to ‘daughter’ cells is

inhibited. For horka, all chromosomes except X” are unreliably inherited

during cleavage and subsequent cell divisions producing the blastula (Szbad

et al. 1995). If the X chromosome is not inherited by one of the cells

produced at cleavage, its absence will likewise be inherited by all ‘daughter’

cells. As the two cells produced at cleavage subsequently proliferate into

what later become the lateral halves of the adult organism, this type of error

106 Australian Entomologist, 2010, 37 (2)

may ultimately produce a phenotype whose lateral halves are of opposite sex

(bilateral gynandromorph). Localised loss of the X chromosome later in

development is also believed to produce gynandromorphs whose phenotype

is a mosaic of both male and female morphology (Richards and Davies, 1977;

Wigglesworth 1972) (mosaic gynandromorphism).

Additional causes of gynandromorphism in insects include fertilisation of

binucleate ova, replacement of mitotic cell division with meiosis and

fertilisation by multiple sperm, which may fuse and act as a second nucleus

(Pereira et al. 2003, Richards and Davies 1977, Wigglesworth 1972).

Gynandromorphism in Ornithoptera

Ornithoptera Boisduval, 1832 is a genus of thirteen species of swallowtail

butterflies restricted to the Australasian biogeographic region (ie. east of

Wallace’s line) (Parsons 2000). Three species, O. richmondia (Gray, 1852),

O. euphorion (Gray, 1852) and O. priamus (Linnaeus, 1758), occur within

Australian territories (Braby 2000). Along with two additional genera

(Trogonoptera Rippon, 1890 and Troides Hübner, 1819), the Ornithoptera

are popularly known as birdwings and have attracted considerable scientific

interest in the areas of taxonomy (e.g. Hancock 1983; 1991; Braby et al.

2005), conservation ecology (Collins and Morris, 1985, Sands et al. 1997,

Sands and New 2002) reproductive biology (Orr, 1988) and general ecology

(Matsuka 2001, Parsons 2000). As this genus includes some of the largest

and most spectacular of all Lepidoptera, they are much prized by amateur

collectors (Collins and Morris, 1985). All species (excluding O. alexandrae

(Rothschild, 1907)) are presently bred in ranching programs, with large

numbers sold internationally to collectors. Trade in these species is monitored

under Appendix 2 of the Convention on the International Trade in

Endangered Species (CITES) (United Nations Environment Programme,

World Conservation Monitoring Centre, 2007).

Ornithoptera is an ideal genus to review for gynandromorphs as all species

exhibit spectacular sexual dimorphism, such that extremely small areas of

male tissue may be visible on a female wing, and vice versa. Moreover, the

trade in aberrant Ornithoptera results in a high rate of reporting (if only in

sales catalogues) and there is extensive literature describing and illustrating

even the slightest variations observed in most species (e.g. D’Abrera 2003;

Otani and Kimura 2001, Schaffer 2001). A thorough review of literature

yielded a large number of Ornithoptera gynandromorph records, which are

presented in Table 1. Only specimens clearly identified as natural

gynandromorphs are presented because some fraudulent material has been

‘manufactured’ and advertised for sale on the internet in recent years (eg. a

purported O. x allottei (Rothschild, 1914) gynandromorph, consisting of the

body and left wings of a female O. victoriae regis (Rothschild, 1895) with the

right wings of a male O. priamus urvillianus (Guérin-Méneville, 1830)).

Elements of the Ornithoptera wing pattern defined by Haugum and

Australian Entomologist, 2010, 37 (3) 107

Low (1978) are used to describe the gynandrous phenotype of individual

specimens. Ornithoptera taxonomy used here follows that presented by

Parsons (2000) and Braby (2004).

Fig. 1 Mosaic gynandromorph of O. priamus pronomus collected by H. Elgner at

‘Cape York’, Queensland, on 17 February 1907. Specimen in the Australian Museum.

above: upperside, below: underside.

108 Australian Entomologist, 2010, 37 (2)

Table 1. List of known Ornithoptera gynandromorphs, including for each specimen,

collection locality, gynandromorph type, description of phenotype, literature

references and current repository. Abbreviations: LH (left half); RH (right half); FW

(forewing); HW (hindwing); D (dorsal wing surface); V (ventral wing surface); AM

(Australian Museum); British Musuem (Natural History) (BM(NH)); IFTA (Insect

Farming and Trading Agency, Papua New Guinea); P (various private collections).

Species Locality Type Phenotype Reference and

Group Taxon Repository

euphorion Australia Mosaic 3 overall with 2 Schiffer (2001)

(Kuranda, scaling on RHFW p

Queensland) and HW. Halved

genitalia.

richmondia Australia Mosaic © overall with Sands and Scott 2002

(Queensland?) partial Í radial Photograph only

band on RHF W

apex.

priamus Australia Mosaic Figure 1. Common and

pronomus (Cape York’, Waterhouse 1972

Queensland) (as O. priamus

poseidon)

AM:

priamus Papua New Bilateral LH male, RH 9; Haugum and Low

admiralitatus Guinea abdomen 1978

Rothschild, (Trobriand bilaterally P

1915 Isl.) divided.

priamus Papua New Bilateral Not figured; LH Parsons 2000

admiralitatus Guinea 3; Received by Not specified

(Admiraly Is.) IFTA.

priamus New Guinea Bilateral LH ĝ, RH 9; D’Abrera 2003

poseidon (no locality) abdomen BM(NH)

bilaterally

divided.

priamus Papua New Mosaic LH dw. 2 Otani and Kimura

poseidon Guinea scaling on HWV; 1998;

(Aseki, RH Qw.d Matsuka 2001

Morobe scaling on HW. P

Province) Abdomen w. Í

genitalia, mostly

Q dorsally, mostly

Í ventrally.

Australian Entomologist, 2010, 37 (3) 109

Table 1 continued

Species Locality Type Phenotype Reference and

Group Taxon Repository

priamus Papua New Mosaic Predominantly J; ` Matsuka 2001

poseidon Guinea (Aseki, RHFW mostly 2 P

Morobe w. partial ĝ radial

Province) band; LH Í w. Q

scaling on HW

tornus and FW

apex.

priamus Indonesia Mosaic Predominantly Q; Otani and Kimura,

poseidon (Nabire, FW w. partial Ó 1998

Papua radial streak and P

province) evidence of black

median stripe.

Overall markings

diffuse w. much

iridescence.

priamus Papua New Mosaic LH Q; RH dw. Parsons (1999)

urvillianus Guinea limited Q scaling [FTA

(Bougainville on HW. Ẹ

province) abdomen w. Í

scaling. Ranched.

croesus Indonesia Bilateral LH 3, RH 9; Parrott and Schmid,

lydius (Halmahera abdomen 1984 (in Parsons

Is.) bilaterally 2000)

divided. P

victoriae Papua New Mosaic FW generally 3 D'Abrera 2003

regis Guinea with Q markings; Howarth 1977

(Bougainville HW Í with Q BM(NH)

province) pattern and ĝ

scaling.

Several Ornithoptera gynandromorphs that lack a formal literature reference

but are otherwise well known to collectors via the internet are also presented.

They are listed separately to those discussed in literature (Table 2).

Specimens of female O. priamus poseidon and O. aesacus with iridescence

on the wings not taking the form of defined markings are not considered to be

gynandromorphs.

Australian Ornithoptera gynandromorphs

Of the three known Australian Ornithoptera gynandromorph specimens, only

one resides in an Australian collection. This specimen, a mosaic

gynandromorph of O. priamus pronomus (Figure 1), was collected at ‘Cape

York’, Queensland, by H. Elgner on 17 February 1907. It is currently lodged

110 Australian Entomologist, 2010, 37 (2)

in the Australian Museum. This specimen was incorrectly referred to as a

specimen of O. priamus poseidon (Doubleday, 1847) by Common and

Waterhouse (1971). A gynandromorphic specimen of O. euphorion held in a

private collection in Germany was figured on both surfaces by Schäffer

(2001) and is described as originating from Kuranda, Queensland. It is most

likely a captive bred specimen. A second, bred mosaic gynandromorph of O.

euphorion is also known and will be discussed in a forthcoming article

A gynandromorph of O. richmondia figured by Sands and Scott (2002; pages

8 & 46) was predominantly female with a partial male radial band. It does not

appear to have been collected and was not recognised in text as a

gynandromorph by Sands and Scott (2002). The provenance of this

photograph was not cited.

Table 2. Ornithoptera gynandromorphs sighted via the internet with no literature reference

(‘anecdotal’ records).

Taxon Locality Type Phenotype

goliath Indonesia Mosaic Predominantly 9; RHFW mostly Í w. slight

(Papua © influence; LHF W w. partial Ï radial band.

province) Ranched.

goliath Indonesia Mosaic Q w. partial Í cubital band on LHFW.

(Papua Ranched.

province)

goliath ? Bilateral LH ĝ, RH Ọ.

Other notable Ornithoptera gynandromorphs

Ornithoptera gynandromorphs have also been reported for O. priamus, O.

croesus lydius (Felder, 1865) and O. goliath Oberthür, 1888, with the

majority of specimens known from O. priamus subspecies (Appendix 1). The

higher frequency of gynandromorphs for the latter taxa probably reflects its

abundance through much of its natural range (Parsons 2000) and the quantity

of specimens collected for trade, as opposed to a genuinely higher frequency

of gynandromorphism. Trade in O. priamus subspecies represent some 47%

of all Ornithoptera trade monitored by CITES, with at least 158,369

specimens exported from Indonesia, Papua New Guinea and the Solomon

Islands between 1985 - 2005 (United Nations Environment Programme,

2007).

Three gynandromorphs of O. goliath were also examined from a series of

detailed photographs published on the internet between 2001 - 2010

(Appendix 2). Additional Ornithoptera gynandromorphs known to the author

from anecdotal reports alone have been omitted because it was impossible to

independently verify their existence or the nature of their phenotype. It is

hoped this paper will encourage those with gynandromorph specimens of any

insect taxon to publish detailed photographs and descriptions in appropriate

literature.

Australian Entomologist, 2010, 37 (3) 111

Acknowledgements

I am grateful to David Britton for allowing me to illustrate the O. priamus

pronomus from the Australian Museum, and for providing its collection data.

I wish also to thank Les Ring, David Rees, Ted Fenner and two anonymous

reviewers, who all provided useful feedback on the draft manuscript. John

Olive also provided interesting discussion regarding an additional Australian

Ornithoptera gynandromorph. I am particularly grateful for the unwavering

encouragement in all things offered by my wife, Haliz.

References

AGATE, R. J., GRISHAM, W., WADE, J., MANN, S., WINGFIELD, J., SCHANEN, C.,

PALOTIES A. and ARNOLD, A. 2003. Neural, not gonadal, origin of brain sex differences in a

gynandromorphic finch. Proceedings of the National Academy of Sciences 100: 4873-4878

BRABY, M.F. 2000. The butterflies of Australia their identification, biology and distribution.

CSIRO Publishing: Collingwood; xx + 976 pp

BRABY, M.F. 2005. The complete field guide to butterflies of Australia. CSIRO Publishing:

Collingwood. 2 Vols., xx + 980; 70 plates.

BRABY, M., TRUEMAN, J. and EASTWOOD, R. (2005) When and where did troidine

butterflies (Lepidoptera: Papilionidae) evolve? Phylogenetic and biogeographic evidence

suggests an origin in remnant Gondwana in the Late Cretaceous. Invertebrate Systematics 19:

113-143.

COLLINS, N.M., and MORRIS M.G. 1985 Threatened swallowtail butterflieds of the world: the

IUCN red data handbook. IUCN: Gland and Cambridge. viii + 402; 8 plates.

COMMON, I.F.B. and WATERHOUSE, D.F. 1981. Butterflies of Australia. Revised edition.

Angus and Robertson: Melbourne. xiv + 684; 49 plates.

D’ABRERA, B. 2003 Birdwing butterflies of the world. New and revised edition. Hill House

Publishers: Melbourne. 324 pp.

HANCOCK, D.L. 1983. Classification of the Papilionidae (Lepidoptera): a phylogenetic

approach. Smithersia 2: 1-48.

HANCOCK, D.L 1991. Notes on the phylogeny and biogeography of Ornithoptera Boisduval

(Lepidoptera: Papilionidae). Tyô to Ga 42: 17-36.

HAUGUM, J. and LOW, A.M. 1978. A monograph on the birdwing butterflies. The systematics

of Ornithoptera, Troides and related genera. Vol. 1. The genus Ornithoptera. Scandinavian

Science Press: Klampenborg. 309 pp, 12 plates.

HOWARTH, T.G. 1977. A list of the type-specimens of Ornithoptera (Lepidoptera:

Papilionidae) in the British Museum (Natural History). Bulletin of the British Museum (Natural

History), entomology. 36: 153-169.

MATSUKA, H. 2001. Natural history of birdwing butterflies. Privately published: Japan. 368

pp.

ORR, A.G. 1988. Mate conflict and the evolution of the sphragis in butterflies. PhD thesis,

Griffith University, Brisbane; xvii + 348 pp.

OTANI, T and KIMURA, Y. 1998. Birdwing butterflies Ornithoptera. Endless collection series

volume 7. Endless scientific information: Japan. ii + 116 pp.

112 Australian Entomologist, 2010, 37 (2)

PARROTT, R.E. and SCHMID F. 1984. a new subspecies of Ornithoptera croesus from Morotai

Island, Indonesia, and a gynandromorph of O. croesus lydius (Lepidoptera: Papilionidae), The

Canadian Entomologist 116: 1519-1524.

PARSONS, M.J. 1999. The Butterflies of Papua New Guinea, their systematics and biology.

Academic Press: Hawaii. xvi + 738; 262 plates.

PEREIRA, R.A.S., PRADO and A.P., KJELLBERG, F. 1994. Gynandromorphism in pollinating

fig wasps (Hymenoptera: Agaonidae). Entomological News: 114: 152-155.

RICHARDS, O.W. and DAVIES, R.G. 1977. Imm’s general textbook of entomology. Volume 1.

structure, physiology and development. Tenth Edition. Chapman and Hall: London. PP

SANDS, D.P.A. and SCOTT, S. 2002. Conservation of birdwing buterflies. SciComEd and

THECA: Queensland. 48 pp.

SANDS, D.P.A. and NEW, T.R. 2002. The Action Plan for Australian Butterflies. Environment

Australia, Canberra. 378 pp.

SANDS, D.P.A., SCOTT, S.E. and MOFFATT, R. 1997. The threatened Richmond birdwing

butterfly (Ornithoptera richmondia (Gray)): a community conservation project. Memoirs of the

Museum of Victoria 56: 449-453.

SCHAFFER, O. 2001. Butterflies of the world. Part 12 Ornithoptera. Plates and text in two

volumes. Goecke and Evers, Koltern. 14 pp. + 53 plates.

SZBAD, J., MATHE, E. and PURO, J. 1995. Horka, a dominant mutation of Drosophila,

induces nondisjunction through paternal effect, chromosome loss and genetic mosaics. Genetics

139: 1585-1599.

UNITED NATIONS ENVIRONMENT PROGRAMME, 2007. Review of trade in ranched

birdwing butterflies. Prepared for the European Commission Directorate. Available online at:

http://ec.europa.eu/environment/cites/pdf/review_butterflies.pdf (verified online 4 May 2010).

WIGGLESWORTH, V.B. 1972. The principles of insect physiology. Seventh edition. Chapman

and Hall: London. viii + 827 pp.

Australian Entomologist, 2010, 37 (3): 113-114 113

MULTIPLE STYLOPISATION OF A PAPER WASP, ROPALIDIA

ROMANDI (LE GUILLOU) (HYMENOPTERA: VESPIDAE)

‘CLYDE H. WILD and CASEY R. HALL

Griffith School of Environment, Griffith University Gold Coast Campus

Parklands Qld 4222 (E-mail: 'Clyde.Wild@Griffith.edu.au *Casey.Hall@student. griffith.edu.au)

Notes

Figure 1 shows a vespid wasp, Ropalidia romandi (Le Guillou, 1841) bearing

three extruded, parasitic stylopids (Strepsiptera: Stylopidae). It is not

absolutely clear from the photograph if these are all adult females or include

male pupae, although given the host’s behaviour it seems most likely that at

least one female is present. The photograph was taken on 8 February 2010 in

Southport, Queensland. The wasp was behaving strangely, crawling to the top

of a blade of grass, falling off and the climbing another one. It is possible the

female parasites manipulate the host’s behaviour so that it climbs to a good

position to encounter males.

Strepsipterans are rarely seen and enigmatic. In nearly all the female is a

permanent endoparasite of the host, which, depending on the species, may

belong to one of several orders. Unlike most insect parasitoids, strepsipterans

do not kill their host at pupation and some species actually prolong their

host’s life relative to the unparasitised condition (Kathirithamby 1991).

Fig. 1. Ropalidia romandi bearing three stylopid strepsipterans extruded through

intersegmental membranes (Photo: Casey R. Hall).

114 Australian Entomologist, 2010, 37 (3)

The male is a small, winged insect with greatly reduced forewings and flying

hind wings, while in nearly all species the female is a reduced simplified

permanent endoparasite in the abdomen of the host, occupying up to 80% of

the abdomen’s volume. Stylopisation renders the host sterile in most cases

and changes in the host morphology, cuticle and behaviour are seen

(Kathirithamby 1991). The male lives only a few hours after emergence from

his host, eats nothing and mating takes place on the female’s host (Pohl and

Beutel 2008). Eggs hatch within the mother and the female then produces

thousands or even millions of tiny active triungulin larvae (Pohl and Beutel

2008). These seek out suitable hosts and dissolve their way through its

cuticle, commencing their parasitic lifestyle in a sac produced from host

tissues which apparently protects the parasite from cellular defences by the

host (Kathirithamby et a/. 2003).

These very highly specialised parasites are so unlike other insects that their

placement within the Insecta is still unresolved (Pohl and Beutel 2008). They

have been hypothesised to be a family within the beetles, a separate order

closely related to the beetles, an order close to the Diptera, and, given that the

larvae of males have external wingbuds, not even included in the

Endopterygota (Whiting 1998). Molecular biology work seems to be making

some progress on “the Strepsiptera problem”, as it is referred to in the

literature, and a view is developing that these strange insects lie near the

Diptera (Whiting et al. 1997, Wheeler et al. 2001), which is not inconsistent

with their life-history, and some of their morphology and anatomy.

References

KATHIRITHAMBY, J. 1991. Strepsiptera. In Insects of Australia: a Textbook for Students and

Research Workers. 2"? Edn pp 684-695. Melbourne University Press ISBN 0 522 84454 5.

KATHIRITHAMBY, J., ROSS, L.D. AND JOHNSON, S.J. 2003. Masquerading as self?

Endoparasitic Strepsiptera enclose themselves in host-derived epidermal “bag”. Proceedings of

the National Academy of Science 100: 7655-7659.

POHL, H. AND BEUTEL, R.G. 2008. The evolution of Strepsiptera (Hexapoda). Zoology 111:

318-338.

WHEELER, W.C., WHITING, M.F., WHEELER, Q,D. AND CARPENTER, J.M. 2001. The

cladistics of the extant hexapod orders. Cladistics 17: 113-169.

WHITING, M.F. 1998. Phylogenetic position of the Strepsiptera: Review of molecular and

morphological evidence. International Journal of Morphology and Embryology 27: 53-60.

WHITING, M.F., CARPENTER, J.C., WHEELER, Q.D. AND WHEELER, W.C. 1997. The

Strepsiptera problem: phylogeny of the holometabolous insect orders inferred from 18S and 28S

ribosomal DNA sequences and morphology. Systematic Biology 46: 1-68.

Australian Entomologist, 2010, 37 (3): 115-127 115

THE LIFE HISTORY OF ATTACUS WARDI ROTHSCHILD

(LEPIDOPTERA : SATURNIIDAE) FROM THE NORTHERN

TERRITORY, AUSTRALIA

'D.A. LANE, °G. MARTIN & °R.P. WEIR

'3 Janda Street, Atherton, Old 4883, d. L.lane@bigpond.net.au

26 Rosebury Drive, Palmerston, NT 0830, g.martin@terrorbyte.com.au

°] Longwood Avenue, Leanyer, NT 0812, Richard.Weir@nt.gov.au

Abstract

The life history of Attacus wardi Rothschild is described from the Northern Territory, Australia,

and discussion is presented on the species biology, distribution and potential distribution, based

on available specimen records and the known foodplant distribution. The only known larval

foodplant is Croton habrophyllus Airy Shaw (Euphorbiaceae). Comparisons are made between

this species and Attacus dohertyi Rothschild, Attacus intermedius Jurriaanse & Lindemans,

Attacus inopinatus Jurriaanse & Lindemans, Attacus erebus Fruhstorfer and Attacus aurantiacus

Fruhstorfer, from Indonesia and East Timor. Some initial assessment is presented on the

conservation status of Attacus wardi.

Introduction

The genus Attacus Linnaeus, 1767 belongs to the Tribe Attacini, Subfamily

Saturniinae and includes most of the species popularly referred to as Atlas

Moths, due to their large size, distinctive wing markings and broad wing area

and shape. The genus is widely distributed throughout the Asian and Indian

region, from the Himalayas, south through India to Sri Lanka, southern

China, the south east Asian mainland, the Philippines, extending through

Indonesia to northern Australia. Peigler (1989) in a revision of the genus

Attacus lists fourteen species.

Attacus wardi Rothschild 1910 (Fig 15) is an Australian endemic and the

only known Attacus species from northern Australia. It was first collected

and recorded from the Northern Territory, Australia in February—May, 1909,

when F.P. Dodd collected a number of cocoons of this species from “Port

Darwin” (Peigler 1989). Dodd subsequently bred quite a number of

specimens (about 50) from wild collected cocoons taken from “Port Darwin”

and distributed most of these adult specimens to many of the world’s

museums and some private collections (Peigler 1989). Unfortunately no

information was published or provided by Dodd on any life history or

biological details of the species, although some notes on adult emergence

times were reported by Dodd (in Oberthur 1916).

Since the first collection of the species in 1909-1910, the species remained

completely unknown, with no further observations being made until a solitary

adult male was collected at light by E.D. Edwards at Black Point, Cobourg

Peninsula, Northern Territory in January, 1977 (E.D. Edwards pers. comm.,

Peigler 1989). Since 1977, no additional records or further observations were

made, until some limited evidence was observed by or presented to the

authors of the species' occurrence on Bathurst and Melville Islands,

116 Australian Entomologist, 2010, 37 (3)

(collectively known as the Tiwi Islands), N.T. during 2005, 2007 and 2008.

The presence of the species was again confirmed on Melville Island during

March, 2009, and March 2010, where adult moths, wild eggs, first, second,

and third instar larvae, and pupal exuviae (empty cocoons) were found by the

authors, and detailed observations were made of adult moth behaviour and of

its early stages. During March 2010 the species was also found at Gunn Point

near Darwin, where wild eggs were collected.

Attacus wardi was first described and subsequently recorded by other authors

as a subspecies of Attacus dohertyi Rothschild 1895, viz Attacus dohertyi

wardi, until a revision of the genus Attacus raised wardi to full species status

(Peigler 1989). The distinctive life history of Attacus wardi described below

confirms its full species status.

Life history

The known larval foodplant (Fig 14) Croton habrophyllus Airy Shaw

(Euphorbiaceae) is a tree that is endemic to Western Australia and the

Northern Territory (Hyland & Whiffin 1993), and grows to a height of 8-10

metres within monsoon forest areas and fringing forest areas along

watercourses, from sea level to approximately 100 metres elevation. Trees of

between 3-7 metres in height and growing either along the outer margins of,

or along transect tracks through the monsoon forests appeared to be

particularly favoured by ovipositing females of A. wardi. It is believed that

such situations provided adequate flight space for the large female moths -

however these observations may more reflect the authors’ observational

techniques, as many of the crowns of tall Croton trees growing within the

monsoon forest areas were more difficult to adequately access. In fact it is

likely that the upper crowns of tall Croton trees are the favoured oviposition

sites of the moths, as typical larval feeding patterns within the upper canopy

were regularly observed from the ground. All wild eggs found had been laid

in the upper crown of foodplant trees on the underside of mature leaves.

Egg (Fig 1). Oval, flattened type, approx 2.6x2.1x1.6mm high, pale brownish

white, laid singly or in a line of two, three or four adjacent but separated by

several millimetres on the underside of the foodplant leaf, lying near to the

leaf margin and usually no further than 1cm from the leaf margin. Eggs have

a coating of a pale brown secretion which appears to be an adhesive agent for

affixing the eggs to the leaf surface. Wild A. wardi eggs were observed to

have an incubation period of at least 10 - 15 days.

First instar larva (Figs 2-3). Length 5-10mm. Head, prolegs, thoracic and

abdominal segments all jet black. Each segment carries six long fleshy, erect

scoli, two dorsal, one subdorsal each side, one scoli below spiracles, all jet

black. All scoli carry a series of radial black spines at apex. Larval duration

4-5 days.

Australian Entomologist, 2010, 37 (3) 117

Second instar larva (Figs 3-4). Length 10-16mm. Head and thoracic legs

light brown. Prolegs, thoracic and abdominal segments white, but with mixed

pale brown intermittent spotting. Erect scoli much longer than those of first

instar, equally fleshy, all coloured white with apex of scoli adorned with a

ring of white “cotton wool” like fleshy appendages and short white radial

setae. Scoli on anal segments shorter than those on abdominal segments. Anal

prolegs white. Abdominal segments 1&2, 7&8 each with a lateral red area

that lies between the subdorsal scoli and the scoli below the spiracles.

Thoracic and abdominal segments are lightly coated in a fine whitish powder.

Larval duration 7-8 days.

Third instar larva (Fig 5). Length 16-37mm. Head and thoracic legs light

brown. Prolegs, thoracic and abdominal segments all white, but adorned with

intermittent dark green spotting. Long fleshy scoli coloured white, adorned at

apex as in second instar larva with “cotton wool” like fleshy appendages and

short white setae — those on abdominal segments sloping backwards.

Thoracic and abdominal segments covered with a white wax-like powder

which can be easily dislodged if the larva is touched or handled in any way.

Anal prolegs same colour as prolegs and body, but shows the first discernable

sign of a “false eyespot” lateral marking. Prolegs each with a series of short

white basal setae. Larval duration 9-10 days.

Fourth instar larva (Figs 6-8). Length 37-60mm. Head and thoracic legs pale

greenish white. Prolegs, thoracic and abdominal segments all white but

adorned with intermittent dark green spotting; spotting is larger and more

distinctive on abdominal segment 8 and anal prolegs. Scoli long and fleshy,

much thinner than those of third instar, coloured white but with upper half

pale blue — (this blue colouring is less pronounced in early fourth instar

larvae), becoming darker after two days. Scoli with short white setae arising

from various lengths along the scoli stem. Scoli on abdominal segments

sloping backwards as in third instar. Spiracles very pale blue, ringed white.

Larval body carries patches of white wax-like powder, but in lesser quantity

than that of third instar. Anal prolegs adorned with a black lateral “false

eyespot”, lightly pitted. Scoli on anal segments greatly reduced, three dorsal

rows of low domed shape, and coloured light blue; an anterior row of 4, with

two lower parallel posterior rows comprising 2 scoli each. These scoli

represent the morphological change to defensive glands, as also recorded for

A. dohertyi (Paukstadt & Paukstadt 1993), however their function as such

could not be determined. Larval duration 12-13 days.

Fifth instar larva (Figs 9-10). Length 60-85mm. Head, thoracic legs, prolegs,

and body segments all light green, closely matching the colouration of the

foodplant leaves. Dark green intermittent spotting faintly indicated on

thoracic segments, but more pronounced on abdominal segment 8 and anal

prolegs. Long thin scoli light green at base, white mid section, with dark blue

upper one third; those on abdominal segments sloping backwards as in fourth

118 Australian Entomologist, 2010, 37 (3)

Figs 1-7. Attacus wardi: (1) egg; (2) first instar larva; (3) first & second instar larva;

(4) second instar larva; (5) third instar larva; (6) early fourth instar larva; (7) fourth

instar larva, lateral view.

Australian Entomologist, 2010, 37 (3) 119

instar; each white and dark blue scoli section is also adorned with short white

and blue setae respectively. Some small amounts of white waxlike powder

present in the “folds” at segmental junctions. Spiracles coloured as on body,

ringed lighter green. Anal prolegs with black “false eyespot” lightly pitted.

Eight much reduced, low dome shaped scoli on anal segments coloured

darker green, representing defensive glands but their functionality was not

confirmed. The larval body is large and bulky. Larval duration 11-12 days.

Sixth instar larva (Figs 11-13). Length 85-115mm. Very similar to fifth

instar, but more bulky. Scoli of similar colour to fifth instar, but shorter in

length. No visible defensive gland scoli on anal segments. Intermittent dark

green spotting not distinct, and only present on abdominal segment 8 and anal

prolegs. Larval duration 8-9 days.

Pupa and cocoon (Figs 16-17, 18-19). Pupa dark brown, approaching black,

stout and ovoid in shape, length 32-38mm, width 18-22mm at wingcases.

Male antennal covers much broader than female. Hindwing wing cases

extend past that of the forewing. Cremaster blunt, rounded. The only

available (to the authors) pupae for comparison are those of Attacus erebus

Fruhstorfer from Sulawesi, Indonesia (DAL coll., legacy S. Naumann,

Germany) — the pupa of erebus is slightly larger than but similar in shape to

that of wardi, and light brown in colour. The pupa of A. dohertyi is brown,

whilst that of A. inopinatus is red brown (Paukstadt & Paukstadt 1993,1992).

The location of wild observed cocoons of A. wardi indicate that in some

situations mature larvae leave the foodplant tree canopy to reach lower

sections of the tree, or leave the tree to reach intertwining vines or

understorey shrubs, on which to pupate. Cocoons are cylindrical or broadly

cylindrical, elongated, tapering at each end, 60-90mm in length, 25-40mm

wide at the midsection, are of double walled construction, and have either a

single or several leaves wrapped around the outer cocoon, all tightly attached

with silk, which serves to camouflage the cocoon to a remarkable extent; the

impression is one of a dead hanging leaf. The leaf stalk and adjacent stem is

also wrapped in silk, effectively securing the cocoon from falling away.

Colour of cocoons ranges from light tan to coffee brown, and is slightly

darker than that of A. erebus. During the final process of pupation, A. wardi

larvae, once suitably enclosed by their newly spun white silken cocoon, and

also housed by the wrap-around leaf, appear to regurgitate a brown liquid

which initially saturates the cocoon walls, with some excess brown liquid

dripping and falling away from the cocoon. This process gives the cocoon its

distinct tan or coffee brown colour. Its purpose is not clearly understood, but

it is believed that it acts as a drying and sealing agent to protect the inner

pupa from fungal, bacterial or viral infections, and possibly dessication.

Adult moths have been observed to emerge from their cocoons after intervals

of 21 -30 days, or the cocoons may enter diapause for periods of up to twelve

months.

120 Australian Entomologist, 2010, 37 (3)

Dodd, in his 1916 recorded observations of his trip to Port Darwin in 1909-

1910, when referring to the cocoons that were collected, stated “Most of

these emerged within three months time, some up to six months after our

return to Kuranda, but the last one duly emerged after remaining in pupa for

over 14 months.”

Biological observations

Of numerous eggs found in the wild, quite a few were found to be parasitised

by a minute species of wasp (Agiommatus sp., Superfamily Chacidoidea,

Family Pteromalidae, Subfamily Pteromalinae. (identification John La Salle,

E.D. Edwards, pers. comm., specimens deposited in ANIC)). Species of wasp

from this genus have been recorded parasitising moth eggs in Antheraea

(Saturniidae), Acherontia (Sphingidae) and those of skippers belonging to

Erionata (Hesperiidae). The genus is distributed in Africa (including

Madagascar), South Asia to Australia (E.D. Edwards, pers. comm.). This

minute species of Agiommatus appears to be a primary population control

agent of A. wardi, and by our observations approximately 35% of wild eggs

had been parasitised. Each parasitised egg had only a single adult wasp

emerge, executed by cutting a minute exit hole through the egg wall.

First instar larvae appear to be fairly tenuous of life — they first devour the

greater part of the hatched eggshell, leaving only a basal plate or part of the

wall. A period of inactivity of 4 — 12 hours then follows, with those larvae

remaining motionless on the leaf underside, usually near to the remnant

eggshell. Two such first instar larvae did not begin to feed, and perished —

however after this inactive period all others began to feed by cutting irregular

patches from the leaf margin of semi juvenile leaves. By second instar, larvae

wander and become widely dispersed through the upper tree.

Larval development through the six instars is over 51-56 days, and all instars

displayed periods of continuous feeding followed by equal periods of

inactivity of up to several hours. Apart from short periods of moving around

the foodplant tree to seek fresh feeding areas, larvae of all instars mostly

remained on the underside of the foodplant leaves, or in the case of fifth and

sixth instar individuals, sometimes upside down on small twigs. Second to

sixth instar larvae displayed a high degree of protective camouflage amongst

their foodplant leaves. First instar larvae generally resembled brown or

blackish leaf blemishes that were a feature of mature foodplant leaves, and

were equally well camouflaged.

On Melville Island adult moths are often taken by Boobook Owls (Ninox

novaeseelandiae) after they have been attracted to street lights. Here the birds

appear to patrol around certain street lights, awaiting the arrival of any large

bodied moths. Our observations of the regular assemblage of remnant moth

wings including Attacus on the ground below certain street lights was

testament to the birds! activity. Green Tree Ants (Oecophylla smaragdina

(Fabricus, 1775)) are known to prey upon saturniid larvae in general

Australian Entomologist, 2010, 37 (3) 121

(DAL pers obs). Foodplant trees with any Green Tree Ant activity never

contained eggs or larvae of A. wardi.

While our field observations were conducted over a brief time interval during

March 2009 and March-April 2010, and to date the only confirmed foodplant

is C. habrophyllus, there is a strong probability that other foodplant trees

would be utilised, particularly other plants belonging to Euphorbiaceae.

Croton tomentellus F. Muell. grows in similar situations to C. habrophyllus,

though it is a slightly smaller tree in height, and also has a slightly wider

distribution, extending from Western Australia to the Northern Territory,

through to north east Queensland (Hyland & Whiffin 1993). Omolanthus

novo-guineensis (Warb.) Lauterb. & Schumann (Euphorbiaceae) also has a

similar distribution and habitat to both above Croton species, ranging from

Western Australia through to north east Queensland (Hyland & Whiffin). A

number of larvae of all instars of A. wardi were offered O. novo-guineensis as

food, and readily accepted this and developed at a comparable rate to those

on C. habrophyllus.

Discussion

Some Australian, Indonesian and Papuan species of Saturniidae are quite

sedentary; adults are not often observed, in cases due to extremely localised

distributions, extremely short flight periods that may be directly tied with

seasonal rainfall or relative humidity, and relatively short adult lifespans

(estimated 3-5 days, DAL pers. obs.). Nocturnal flight periods may be either

late evening or early morning. Attacus wardi appears to fit very well into this

category, and its flight periods, biology and sedentary behaviour can be

directly compared with Attacus dohertyi from Timor (DAL, pers. obs.).

Available records of adult A. wardi emergences (including from Dodd’s

original material, Peigler 1989) list the months of January to February,

extending to early March. Our observations of wild adult moths confirm a

nocturnal period of flight activity between 23.00 — 03.00 hours, and

combined with observations of some Tiwi Islands residents, also confirm

early to late March as the flight season, provided good monsoonal rainfall

occurs. Our observations of eggs, first, second and third instar larvae, and

empty cocoons in late March, gives further credence to March being part of

the flight season of A. wardi.

Of historical interest, during 1935 Walter Dodd published a series of

newspaper articles titled “Meanderings of a Naturalist” in the “North

Queensland Register” published out of Townsville (G. Monteith, pers

comm). In one such article dated 9 March 1935, Walter describes the trip to

Darwin with his father in 1908, and when referring to A. wardi states “We

had many pupae, and whilst the moths were emerging, males of the species

were occasionally attracted from the jungle several miles away”.

Observations of the flight activity of A. dohertyi in East Timor (DAL & M.

Lane, 2002 & 2004) confirmed that adult emergence is triggered by the first

122 Australian Entomologist, 2010, 37 (3)

substantial wet season rainfall. After good rainfall in October 2002 that

followed several months of little to no rainfall, a significant number of adult

moths were collected at lights at Bobonaro for a period of about one week

(always between 2300 — 0300 hrs), but then populations quickly tapered and

no further adult activity was observed.

Comparisons

Peigler (1989) lists the characteristics that clearly separate adults of A. wardi

from those of three geographically adjacent and closely related species, viz A.

dohertyi (Timor, Romang and Damar Is. (Lesser Sunda Islands)), A.

intermedius (Tanimbar Islands), and A. inopinatus (Flores (Lesser Sunda

Islands)). Paukstadt & Paukstadt (1992 & 1993) described and illustrated the

life histories of A. inopinatus and A. dohertyi with black and white

photographs. Peigler & Wang (1996) documented in colour photographs the

life history of A. dohertyi. Peigler (1989) stated that A. intermedius was so

named because its authors intended to convey the point that this moth appears

intermediate between A. dohertyi and A. wardi, believing that all three

species were conspecific. The life history of A. intermedius remains

unknown.

All larval instars of A. wardi are readily distinguished from those of both A.

dohertyi and A. inopinatus. Firstly the jet black colouration of the first instar

larva of A. wardi is unique amongst all known species of Attacus, and is quite

unlike the first instar larva of either dohertyi or inopinatus, which are both

coloured white with some degree of black striping. Compared with those of

dohertyi, the second, third, fourth and fifth instar larvae of wardi show

noticeable differences in colour, shape and form of scoli, prolegs, thoracic

and abdominal segments. The sixth instar larva of wardi is closer to that of

dohertyi, but differs from that species in the extent of upper blue scoli

colouration. Second to sixth larval instars of inopinatus differ considerably

from those of both wardi and dohertyi, particularly in shape, length and

colouration of scoli, and also in colouration of prolegs, thoracic and

abdominal segments. The final instar larva of Attacus aurantiacus Fruhstorfer

(Nassig & Taschner, 1996) from the Kai Islands, Indonesia differs

significantly from all of the above species, with dorsal and lateral scoli

coloured bright red and dark blue. Based on sixth instar larval characters, A.

wardi appears to have a closer affinity to A. dohertyi than to all other species,

and possibly displays close ancestral ties with dohertyi.

Distribution and conservation status

To date, Attacus wardi has remained a poorly known and understood species

in terms of its known distribution, flight times, range of foodplant

preferences, and its specific habitat requirements. Our observations,

combined with historic records, indicate that the species is restricted to the

monsoon forest areas of the northern coastal areas of the Northern Territory,

including but not limited to Darwin, Gunn Point, the Tiwi Islands, and

Australian Entomologist, 2010, 37 (3) 123

Figs 8-13. Attacus wardi: (8) fourth instar larva, dorsal view; (9) fifth instar larva,

lateral view; (10) fifth instar larva, dorsal view; (11) sixth instar larva; (12) sixth

instar larva, frontal view; (13) sixth instar larva, rear segments & anal claspers.

124 Australian Entomologist, 2010, 37 (3)

Cobourg Peninsula (Garig Gunak Barlu National Park), and adults principally

fly during the wet season months (January to March). Local adult population

numbers can be quite high but for extremely limited duration, and like its

counterpart A. dohertyi, adult nocturnal flight activity is limited from the late

evening to early morning. However, should good summer storms provide

suitable conditions, it is also quite feasible that an earlier generation of A.

wardi might occur around October to December. Based on the published

distributional data of Croton habrophyllus as a known foodplant source, and

combined with the seasonal rainfall distribution of the Northern Territory, A.

wardi may occur as far east as Nhulunbuy and Groote Eylandt, as well as the

many intervening coastal monsoon forest areas, where extensive areas of

suitable habitat remain. It is our belief that several other foodplant tree

species from several plant families will be found to be used by A. wardi, as is

the case with other species of Attacus (Peigler 1989), and also other closely

related genera, such as Coscinocera Butler (DAL pers. obs., Common 1990).

Peigler (1989), under material examined, lists two male specimens of A.

wardi held in the American Museum of Natural History (AMNH), New

York, that carry data labels citing “Cape York” and “North Cape York

Peninsula” in Queensland as localities. These localities are considered most

unlikely by the authors, as the zoogeographic region of Cape York Peninsula

is considered far more closely aligned with the Papuan/Australian faunal area

than that of Indonesia. However, a close lookout should still be maintained

by entomologists particularly those visiting the north western coastline of

Cape York. Peigler (1989) also considered Coscinocera hercules Miskin to

be sympatric with A. wardi at Darwin — this is incorrect as within Australia

C. hercules is restricted to north eastern Queensland.

The forest verges of the ‘Top End’ are a constantly changing interface, where

pioneer trees including Croton species are a component of an ever expanding

or contracting monsoon forest, dependant on rainfall intensity and

seasonality, combined with the frequency and intensity of fire regimes. It is

our opinion that any component of irregular or frequent fires that penetrate

the monsoon forest verges would be severely destructive to populations of A.

wardi. The observed tendency for mature larvae to form their cocoons on mid

to lower parts of the foodplant trees, or on understorey shrubs, would leave

them extremely vulnerable to any fire activity, especially as the species

undergoes dry season diapause as a pupa.

The observed habit of some mature A. wardi larvae leaving the upper forest

canopy to pupate at lower levels may be an adaptation to having developed

some degree of protection against high cyclonic winds. It is of particular

interest to compare this behaviour with that of several north Queensland

species of moth, whose specific habitat is periodically exposed to cyclonic

disturbance. The observed larval habits of leaving their feeding zones in the

forest canopy and forming cocoons on lower sections of their foodplant trees,

Australian Entomologist, 2010, 37 (3) 125

19 `

Figs 14-19: Attacus wardi: (14) foodplant tree, Croton habrophyllus; (15) adult

female, ex pupa; (16) male pupa, lateral view; (17) male pupa, ventral view; (18)

cocoon, cross sectional view, depicting double walled construction; (19) cocoon.

126 Australian Entomologist, 2010, 37 (2)

or on understory trees or shrubs, may well be an adaptation to avoiding the

high canopy winds of infrequent cyclones. In particular, three large moth

species that exhibit this behaviour are C. hercules, the uraniid moth Alcides

metaurus (Hopffer), and the anthelid moth Chelepteryx chalepteryx (R.

Felder). The latter species feeds as a larva within the forest canopy of the tall

wet sclerophyll and rainforest areas at high altitude on the southern and

western Atherton Tablelands. In this situation mature larvae lower

themselves onto understorey shrubs and trees to spin their cocoons (DAL

pers. obs.), whereby the large heavy pupae appear to gain some protection

from high winds.

In the Northern Territory some introduced plant and insect pest species may

prove to be of concern to populations of A. wardi — Gamba grass

(Andropogon gayanus Kunth.) is spreading across the ‘Top End’, and burns

fiercely and hot, potentially providing a threat to the monsoon forest verges

and their extent. The introduced Yellow Crazy Ant (Anoplolepis gracilipes

(Fr. Smith)) occurs in the East Arnhem region near Gove, and colonies of this

ant are notorious for destructively devouring all available food resources in

the near vicinity.

The habitat requirements of 4. wardi as observed at Gunn Point and Melville

Island are reasonable sized pockets of Monsoon Forest, of a possible

minimum size of around 8 hectares but preferably up to 20 hectares or larger,

containing good numbers of Croton foodplant trees. The pockets should stand

in reasonably close proximity to each other, so that adult moths can

intersperse readily between them. The greater the number of such Monsoon

Forest pockets, combined with their near proximity to each other, the more

the secure are A. wardi populations inhabiting them.

Historic maps of Darwin and environs suggest that at one time extensive

bands of Monsoon Forest extended almost continuously from the Darwin

Esplanade, through East Point, to Nightcliff, Casuarina and Lee Point. It is

our opinion that this band of Monsoon forest was the locality from which

Dodd’s 1909 historic specimens were collected. Sadly the greater part of this

habitat is now gone, with the remaining remnant Monsoon Forest now found

only at East Point and Lee Point. Interestingly, at both of these sites a number

of large and small Croton trees are well established, but our searches have so

far failed to find early stages of A. wardi. These areas in theory are indeed

suitable habitats, but we feel that they are currently too isolated from each

other, without intervening pockets, and hence are too small to support a

viable population of 4. wardi. However, should a suitable replanting

programme ever be instigated, to interconnect these remnant pockets with

suitable intervening Monsoon Forest pockets containing adequate Croton and

other Euphorbiaceae trees, then the opportunity may then arise to re-

introduce A. wardi to the Darwin coastal area.

Australian Entomologist, 2010, 37 (3) 127

At present, the conservation status of A. wardi seems fairly secure, with good

populations known from the Tiwi Islands and Gunn Point, coupled with adult

records from Cobourg Peninsula (Garig Gunag Barlu National Park).

Populations are almost certain to be found at intermediate localities between

Gunn Point and Cobourg Peninsula where extensive habitat remains, and

possibly through to the Nhulunbuy area. The protected species status of A.

wardi does not appear justified, particularly as such little research has been

undertaken on its biology and habitat requirements. A conservation status

listing as Data Deficient would be more appropriate — with a recommendation

that much more detailed research be undertaken into the species biology and

distribution. Habitat protection is far more relevant to the long term tenure of

A. wardi, particularly in consideration of the current extremely poor

management of fire regimes across the ‘Top End’.

Acknowledgements

Thanks are extended to the generous support of many Tiwi Island residents.

E.D. Edwards (ANIC, Canberra) is sincerely thanked for identification of the

parasitic wasp species, for searching numerous literature records, and for his

always generous and constructive advice. Staff of the Darwin Herbarium are

thanked for foodplant identifications and advice as to plant distributions. Dr.

Stefan Naumann, Berlin, also searched and supplied historic literature

references, and generously offered much technical advice. Dr. G. Monteith is

thanked for making available some important historic literature. Thanks are

extended to the Department of Natural Resources, Environment and the Arts,

as this research was carried out under permit No 35168.

Bibliography

COMMON, I.F.B. 1990. Moths of Australia. Pp xxxii + 535. Melbourne University Press,

Melbourne.

DODD, F.P. 1916. Faune entomologique de I’ Australie, in C. Oberthur, Et. Lepid. Comp. 11

bis:27,28.

HYLAND, B. & WHIFFIN, W.H. 1993. Australian Tropical Rainforest Trees. An Interactive

Identification System. CSIRO Australia.

NASSIG, W.A. & TASCHNER, F. 1996. Beschreibung einer Altraupe von Attacus aurantiacus

Rothschild 1895 von den Kai-Inseln, Indonesien (Lepidoptera: Saturniidae). Nachr. Entomol.

Ver. Apollo, Frankfurt/Main, N.F. 17 (2): 153-159.

PAUKSTADT, U. & PAUKSTADT, L.H. 1992. Die Praimaginalstadien von Attacus inopinatus

Jurriaanse & Lindemans 1920 von Flores, Indonesien, sowie Angaben zur Biologie und

Okologie (Lepidoptera:Saturniidae). Entomol. Z. (Essen) 102(12):213-222.

PAUKSTADT, U. & PAUKSTADT, L.H. 1993. Die Praimaginalstadien von Attacus dohertyi

W. Rothschild 1895 von Timor, Indonesien, sowie Angaben zur Biologie und Okologie

(Lepidoptera:Saturniidae). Entomol. Z. (Essen) 103(16):281-293.

PEIGLER, R.S. 1989. A revision of the Indo-Australian genus ATTACUS. The Lepidoptera

Research Foundation, Inc. Publication. Beverley Hills, California. 167pp. 4 plates.

PEIGLER, R.S. & WANG, H.Y. 1996. Saturniid moths of Southeastern Asia. Published by the

Taiwan Museum, Taipei. 262pp. c.pl.

iza Australian Entomologist, 2010, 37 (2)

RECENT LITERATURE

Compiled by Max Moulds (msmoulds@bigpond.net.au) & Editor

An ongoing selection of literature published since Daniels' Bibliography of

Australian Entomology 1687-2000. Details of your publications for inclusion

are always welcome.

DUNN, K.L.

2009 Observations on carrying pair behaviour among Asia-Pacific butterflies: part III

(personal field records since 2005). Victorian Entomologist 39: 12-20.

ENDERSBY, I.

2009 Mecoptera of Victoria. Victorian Entomologist 39: 2-4.

FAITHFULL, I.G

2009 Native Budworm, Helicoverpa punctigera (Lepidoptera: Noctuidae) destroys fruit of

Sticky Bartsia, Parentucellia viscosa (Scrophulariaceae), with a discussion of the

beneficial status of Helicoverpa species in weed management. Victorian

Entomologist 39: 7-11.

FORBES, G.

2008 Butterfly species observed at Stanage, April 2008. Metamorphosis Australia,

Magazine of the Butterfly and Other Invertebrates Club 50: 11-14.

FRANKLIN, D.C.

2007 Dry season observations of butterflies in the "Gulf Country" of the Northern

Territory and far north-west Queensland. Northern Territory Naturalist 19: 9-14.

FRANKLIN, D.C. AND BISA, D.

2008 Field key to the lycaenid butterflies of the Top End and Kimberley. Northern

Territory Naturalist 20: 1-18.

GRUND, R.

2009 New range extensions and other data for selected butterflies and sun-moths from

the Maralinga and far westcoast areas of South Australia. Victorian Entomologist

39: 108-114.

HARRIS, K.

2008 A voracious assassin. Victorian Entomologist 38: 79-84.

HENDRY, P.

2008 The Anthelidae. Metamorphosis Australia, Magazine of the Butterfly and Other

Invertebrates Club 50: 27-31

LANE, D.A. EDWARDS, E.D. AND NAUMANN, S.

2010 A revision of the genus Syntherata Maassen, 1873 (Lepidoptera: Saturniidae)

within Australia, with the description of three new species, and descriptions of

their life histories. The European Entomologist 3: 1-39.

MILLER, R.

2008 Further notes on "The osmeterium-type structure found on the larva of Phaedyma

shepherdi". Metamorphosis Australia, Magazine of the Butterfly and Other

Invertebrates Club 50: 26-27.

NAUMANN, S., LANE, D. AND LOFFLER, S.

2009 Some new species of the Indo-Australian genus Syntherata from the island of New

Guinea (Lepidoptera: Saturniidae). Přírodovědné studie Muzea Prostějovska, 2009

(10/11): 43-65.

ENTOMOLOGICAL NOTICES

Items for insertion should be sent to the editor who reserves the right to alter,

reject or charge for notices.

NOTES FOR AUTHORS

Manuscripts submitted for publication can either be submitted as hardcopies

or electronically. Three copies (double spaced text and illustrations) of

hardcopy manuscripts should be submitted. Manuscripts submitted in digital

format should be sent in Microsoft Word. Digital illustrations should be sent

initially as low resolution images in a separate Word file, as low resolution

JPEGs, or as low resolution PDF files, with figure numbers indicated clearly

for each figure. Digital manuscripts may be sent via email to

federica.turco@qm.qld.gov.au Hardcopy manuscripts and digital manuscripts

on floppy disk or CD-ROM should be sent to:

The Editor

The Australian Entomologist

P.O. Box 537,

Indooroopilly, Qld, 4068

Authors should refer to recent issues for layout and style. All papers will be

forwarded to two referees and the editor reserves the right to reject any paper

considered unsuitable.

It is Magazine editorial policy that usage of taxonomic nomenclature will

comply with the mandatory provisions of the International Code of

Zoological Nomenclature.

Papers longer than twenty printed pages will normally not be accepted.

Publication costs are as follows: cost per printed page, or part thereof, $20

(black & white) and $50 (colour). These costs include the supply of a pdf

copy of the paper and 10 hardcopy reprints to the senior author. Papers

occupying one printed page or less may be accepted without charge if no

reprints are required. Reprints may be supplied for one page papers at the

normal cost, by arrangement. Page charges may be reduced at the discretion

of the Publications Committee. An application for reduction must be made,

with reasons, at the time of acceptance of the manuscript.

Printed by ABC Printing, 2 Dorsey Street, Milton, 4064.

THE AUSTRALIAN

Entomologist

Volume 37, Part 3, 30 September 2010

CONTENTS

LAMBKIN, T.A.

A review of Taenaris Hübner (Lepidoptera: Nymphalidae: Amathusiinae)

in Queensland, together with first Australian records for 7. myops kirschi

Staudinger and Elymnias agondas melanippe Grose-Smith (Satyrinae). Z

GREEN, K.

The aestivation sites of bogong moths, Agrotis infusa (Boisduval)

(Lepidoptera: Noctuidae), in the Snowy Mountains and the projected

effects of climate change.

NIELSEN, J.E.

A review of gynandromorphism in the genus Ornithoptera Boisduval

(Lepidoptera: Papilionidae). 105

WILD, C.H. AND HALL, C.R.

Multiple stylopisation of a paper wasp, Ropalidia romandi (Le Guillou)

(Hymenoptera: Vespidae). : 113

LANE, D.A., MARTIN, G. AND WEIR, R.P.

The life history of Attacus wardi Rothschild (Lepidoptera: Saturniidae)

from the Northern Territory, Australia. 115

RECENT LITERATURE 128