HARVARD UNIVERSITY

•8

Library of the

Museum of

Comparative Zoology

QUAESTIONES ENTOMOLOGICAE

ISSN 0033-5037

A periodical record of entomological investig^j/J^JJHblished at the

Department of Entomology, University of Alberta, gcji0jp^fl^[^11^rta.

Volume 26

JAN 1 8 1991

1990

HARVARD

UNIVERSITY

CONTENTS

Kohlmann and Halffter — Reconstruction of a specific Example of Insect

Invasion Waves: the Cladistic Analysis of Canthon (Coleoptera:

Scarabaeidae) and Related Genera in North America . 1

Ball and McCleve — The Middle American Genera of the Tribe Ozaenini with

Notes about the Species in Southwestern United States and Selected

Species from Mexico . 30

Book Review — McAlpine and Wood (Editors). 1989. Manual of Nearctic

Diptera. Volume 3 . 117

Book Review — Trautner and Geigenmiiller. 1987. Carabid Beetles, Tiger

Beetles . 131

Book Notices . 133

Third International Conference on Classification, Phylogeny, and Natural

History of Hydradephaga (Coleoptera): Proceedings . 137

Clark — Revision of the Anthonomus Subgenus Anthonomocyllus Dietz

(Coleoptera: Curculionidae) . 559

Askevold — Reconstructed Phylogeny and Reclassification of the Genera of

Donaciinae (Coleoptera: Chrysomelidae) . 601

Hilchie — Classification, Relationships, Life History and Evolution of

Erebia magdalena Strecker (Lepidoptera: Satyridae) . 665

Commentary: Cooper — Linear, longitudinal markings on the outer elytral

surface of beetles: intemeurs or striae? . 695

Note: Brown — New Nearctic Region Records of Palearcatic Megaselia

Species (Diptera: Phoridae) . 701

Editor's Acknowledgements and Farewell . 703

yip £' ♦

YTia: - v\< ■>

'

300

MCZ

LIBRARY

Quaestiones m 1 5 1390

Entomologicde''-'^'' '

I

A periodical record of entomological investigations

published at the Department of Entomology,

University of Alberta, Edmonton, Canada.

VOLUME 26

NUMBER 1

WINTER 1990

Publication of Quaestiones Entomologicae was started in 1965 as part of a

memorial project for Professor E.H. Strickland, the founder of the Department

of Entomology at the University of Alberta in 1922.

It is intended to provide prompt relatively low cost publication for compre¬

hensive accounts of entomological research of greater than average length.

However, shorter papers about insects in the Prairie Provinces of Canada are

acceptable. Page charges are normally levied, the rate determined by printer’s

charges. For information about current page charges, consult the Editor.

Copy for all types of papers should conform to the Style Manual for

Biological Journals, published by the American Institute of Biological Sci¬

ences, Second Edition, 1964, except that title of periodicals should be given in

full. For style of taxonomic papers, the Editor should be consulted. Two copies

of a manuscript are requested. All manuscripts will be reviewed by referees.

Abstracts are required, one in English and one in another language, prefer¬

ably French.

Copy for illustrations must accompany the manuscript, and be of such char¬

acter as to give satisfactory reproduction at page size (less 1/2 inch, or 1.2 cm

on plates of full page size [7X4 1/4 inches or 17.8 X 10.7 cm]). Reprints must

be ordered when proofs are returned, and will be supplied at cost.

Subscription rates are the same for institutions, libraries and individuals,

$20.00* per volume of four issues, normally appearing at quarterly intervals;

single issues $5.00**. Back volumes and issues are available at the same cost.

These prices supersede those previously indicated, and are subject to change as

required by inflationary pressure on the value of money.

Communications regarding subscriptions should be addressed to the

Subscription Manager, and regarding manuscripts to the Editor.

* Same rate in US$ for non-Canadian subscriptions.

** Single issues of more than 100 pages: $10.00; exception: volume 21(4) - $20.00.

Published quarterly by:

Department of Entomology

University of Alberta

Edmonton, Alberta, CANADA

T6G 2E3

Second Class Mail Registration Number 5222

Return Undeliverable mail to the address above. Return Postage Guaranteed

Issued March 1990

QUAESTIONES ENTOMOLOGICAE

ISSN 0033-5037

A periodical record of entomological investigation published at the

Department of Entomology, University of Alberta, Edmonton, Alberta.

Volume 26 Number 1 1990

CONTENTS

Kohlmann and Halffter — Reconstruction of a specific Example of Insect

Invasion Waves: the Cladistic Analysis of Canthon (Coleoptera:

Scarabaeidae) and Related Genera in North America . 1

Ball and McCleve — The Middle American Genera of the Tribe Ozaenini with

Notes about the Species in Southwestern United States and Selected

Species from Mexico . 30

Book Review — Me Alpine and Wood (Editors). 1989. Manual of Nearctic

Diptera. Volume 3 . 1 1 7

Book Review — Trautner and Geigenmiiller. 1987. Carabid Beetles, Tiger

Beetles . 1 3 1

Book Notices . 133

RECONSTRUCTION OF A SPECIFIC EXAMPLE OF INSECT

INVASION WAVES: THE CLADISTIC ANALYSIS OF CANTHON

(COLEOPTERA: SCARABAEIDAE) AND RELATED GENERA

IN NORTH AMERICA'

Bert Kohlmann

Gonzalo Halffter

Instituto de Ecologi'a, A C.

Apartado Postal no 63

Xalapa, Veracruz, 91000 Quaestiones Entomologicae

MEXICO 26: 1—20

ABSTRACT

The historical biogeography of insects of the Mexican Transition Zone is

extremely complex. Holarctic, Nearctic and Neotropical lines seem to have

invaded and speciated in the area at different times. The neotropical lines in

particular are thought to have invaded the Mexican Transition Zone from South

America on two occasions, one during the Miocene, the other during Plio-

Pleistocene. The subgenus Canthon and its closely related taxa, the subgenus

Boreocanthon and the genus Melanocanthon, as well as the subgenus

Glaphyrocanthon, have been proposed as an example for the analysis of this

process.

The implications from such an event allow predictions regarding the

phylogenesis and ecological evolution of these three taxa, as follows. First, the

proposed Miocene invasion lineage(s) shared a common history and diversified in

the biomes that evolved in North America during the Miocene and Pliocene.

Second, those species which diversified from a common ancestor form a

phylogenetically related group that reflects the history of the group in accordance

with the history of the area and biomes which they occupy. Third, the species

stemming from the proposed South American Plio-Pleistocene invasion evolved

under a different set of ecological and biogeographic conditions, and as such are

distantly related in phylogenetic terms to the members of the first invasions wave.

Fourth, consequently, in a cladogram including species of both invasion waves, the

Plio-Pleistocene elements branch at the base of the tree. Fifth, moreover, the

branching sequence of the cladogram is not in concordance with the estimated age

of appearance of the different biomes which the species occupy. These five

predictions were supported in the present study, by the species cladogram. The

'This paper is a contribution from the "Ecology and Animal Behaviour" project with the support

of the "Studies on the fauna of Coleoptera Lamellicornia from Sierra Madre del Sur"

(P220CCOR - 880061) project, contribution number 21, sponsored by the Direccion Adjunta de

Desarrollo Cientifico of the Consejo Nacional de Ciencia y Technologia (Conacyt), Mexico.

2

Kohlmann & Halffter

results of this analysis therefore lend support to the hypothesis of two waves of

invasions of the Mexican Transition Zone by Neotropical elements from South

America.

RESUMEN

La zona de Transition Mexicana es un area de gran complejidad ecologica, geologica y

biogeografica. En relacion a su dinamica zoogeografica, Halffter (1962, 1964, 1972, 1974, 1976,

1978 y 1987) ha propuesto una serie de hipotesis basadas en las relaciones taxonomicas, riqueza

especffica, historia geologica y patrones de distribution de diferentes grupos de insectos, en las

que senala que diferentes linajes de origen holartico, neartico y neotropical invadieron esta zona

en diferentes epocas. En el caso especffico de los linajes neotropicales, considera dos invasiones

mayores, una durante el Mioceno, la segunda del Plio-Pleistoceno al actual. El subgenero

Canthon y otros dos taxa muy emparentados, el subgenero Boreocanthon y el genero

Melanocanthon, al igual que el subgenero Glaphyrocanthon , son buena evidencia de estas dos

invasiones. Los invasores miocenicos se encuentran en biomas que comenzaron a originarse en

este perfodo, como es el caso de formaciones aridas, pastizales, bosques de encino-pinon y

bosques templados decfduos; o en epocas mas recientes, pliocenicas, como bosques de pino y

pino-encino (Axelrod, 1975, 1979). Las lfneas propuestas como invasoras plio-pleistocenicas o

posteriores se encuentran distribuidas principalmente en biomas de penetration reciente, como

las selvas tropicales, sobre todo la selva alta perennifolia (Germeraad et al., 1968; Graham, 1973,

1981; Toledo, 1976, 1982; Gentry, 1982; Prance, 1982). En base a estudios biogeograficos y

taxonomicos anteriores (Halffter, 1958, 1961, 1962, 1972, 1974, 1976; Halffter y Martinez,

1966, 1967, 1968, 1977) se propone que las siguientes especies del subgenero Canthon :

humectus, pilularius, imitator, vigilans, chalcites y obliquus asf como el subgenero Boreocanthon

y el genero Melanocanthon han derivado de la invasion miocenica. Las especies indigaceus,

cyanellus y morsei del subgenero Canthon y el subgenero Glaphyrocanthon se consideran parte

de la invasion plio-pleistocenica.

El objetivo de este estudio es aportar una nueva evidencia que apoye la existencia de los

dos procesos invasores por elementos neotropicales a la Zona de Transition Mexicana, a partir de

un unalisis cladfstico del subgenero Canthon y taxa cercanos. El analisis cladfstico puede apoyar o

no las predicciones que emanen de las hipotesis biogeograficas.

Con respecto a las invasiones neotropicales pueden establecerse dos hipotesis

complementarias. La primera considerarfa la posibilidad de que a partir de la lfnea miocenica

ocurriera una diversification congruente con la aparicion de zonas adaptativas disponibles

(Simpson, 1953), creadas por el surgimiento de nuevos biomas. La diversification de especies

originada a partir de un ancestro comun resultana en un grupo filogeneticamente emparentado

(Hennig, 1966; Wiley, 1981), cuyo cladograma reflejarfa la historia evolutiva del mismo y

presentarfa un major o menor grado de concordancia con la historia del area y de los biomas que

ocupan las especies. La segunda hipotesis supondria que los invasores plio-pleistocenicos, al

tener una historia ecologica y evolutiva diferente a la de los miocenicos, presentaria relaciones

filogeneticas distintas y una distribution geografica y ecologica tambien diferente. Por ello, al

construirse el cladograma, los invasores plio-pleistocenicos se encontraran en la base la

ramifacacion del arbol y no existira congruencia directa entre la secuencia de ramification del

cladograma y la secuencia de aparicion de los diferentes biomas en Norteamerica.

El analisis cladfstico (Fig. 2) se basa en 29 especies y 29 caracteres (Cuadro I) y emplea el

paquete de analisis filogenetico PAUP. Los caracteres no fueron codificados en forma ordenada,

Quaest. Ent., 1990, 26(1)

Insect Invasion Waves

3

ya que estamos de acuerdo con Meacham (1984) en que la polarizacion de caracteres no se

puede definir con certitud. La base del cladograma fue determinada utilizando el subgenero

Glaphyrocanthon como grupo externo. La decision de elegir a este subgenero como grupo

externo se encuentra fundamentada por los resultados de la estimation del grado de similitud de

todos los generos y subgeneros de los componentes americanos de la tribu Canthon'i na indicado

en la Fig. 1. Cincuenta cladogramas igualmente parsimoniosos fueron obtenidos.

Posteriormente se utilizo el programa CONTREE para establecer un cladograma de consenso

"estricto" (Rohlf, 1982) (Fig. 2). Tambien se obtuvo un cladograma siguiendo el metodo de

consenso de Adams (1972), pero no fue inclufdo en este estudio por ser muy semejante al

primero.

El cladograma de consenso obtenido fue correlacionado con los diferentes biomas (Fig. 2)

en donde se encuentran distribufdas las especies (Mapas 1-7) y su analisis concuerda con las

predicciones de la existencia de dos lmeas invasoras neotropicales en Norteamerica.

TABLE OF CONTENTS

Introduction . 3

Assumptions . 5

Predictions . 5

Material and Methods . 6

Results . 1 0

Tests of Zoogeographic Predictions . 1 1

Concluding Statement . 2 2

Acknowledgements . 22

References Cited . 2 3

Appendix . 26

INTRODUCTION

The dynamics of insect biogeography in the Mexican Transition Zone

(hereafter referred to as MTZ; Halffter, 1976), which includes the southwestern

United States, all of Mexico and a large part of Central America extending to the

Nicaraguan lowlands, are of great complexity. The MTZ is species-rich because

of the great variety of environments and ecological refuges available and presents a

complex and varied overlap of the Neotropical and Nearctic faunas. The area is

also an important part of the north-south dispersal corridor for faunas and floras of

diverse origin that have dispersed during different geological eras between North

and South America (Stehli and Webb, 1985).

Halffter (1962, 1964, 1972, 1974, 1976, 1978, 1987) has developed a set of

hypotheses to explain comprehensively the distribution of insects in the MTZ.

His inferences, based on taxonomic relationships, species richness, geological

history and distribution patterns derived for several insect groups, support the

notion that the present insect fauna of the MTZ has originated from multiple

invasions and in situ speciation at different times from Nearctic, Holarctic and

Neotropical lines, conferring on this region a mixed transitional character in insect

Quaest. Ent., 1990, 26(1)

4

Kohlmann & Halffter

composition. The Neotropical lines which invaded North America from South

America are thought to be represented by a Miocene and by a Plio-Pleistocene

component (Halffter, 1972, 1974, 1976). These components show distinctive

geographic patterns, the distribution cores of which are centered on areas defined

by actual ecological conditions.

Geological support for the possibility of migration between the American

land-masses has been summarized recently. Donnelly (1988) presents a scenario

constructed from the concordant features of the diverse and conflicting analyses

made for the Caribbean and Central American Bridge and concludes that faunal

movements requiring short (tens of kilometers) overwater dispersal conditions

might have occurred during brief intervals during the late Cretaceous through a

proto-Antillean arc and during the middle Cenozoic through Central America.

Overland dispersal became possible when the Central American Land Bridge was

finally consolidated at the beginning of the Pliocene, about 5.7 m.y. B.P. (Kaneps,

1979), or late Pliocene, 3 m.y. B.P. (Webb, 1977, 1978; Keigwin, 1978; Marshall

et al., 1982). Biological evidence presented by Savage (1982) suggests a

connection between North and South America early in the Cenozoic, based on

inferred concordant dispersal to Central America of South American organisms such

as angiosperms, fishes, amphibians and reptiles. Mammals, however, did not

exhibit the same patterns. Nevertheless, some evidence indicates that mammals had

a limited dispersal between North and South America in the late Cretaceous and

again in the Oligocene, that increased during the late Miocene, and culminated in an

extensive faunal interchange in the Pliocene (Stehli and Webb, 1985).

One of the examples given by Halffter (op. cit.) of a Neotropical invasion into

North America in two waves, one during Miocene, the other during Plio-

Pleistocene, is the genus Canthon. This taxon belongs to the subtribe Canthonina,

which, together with three other subtribes, is grouped into the tribe Scarabaeini. It

is characteristic of adults of the tribe Scarabaeini to have fine, long and bowed

metatibiae, used by most species to roll food balls made from droppings (mostly

mammalian) or small carcasses, the latter being more common in the Neotropical

Region. Members of this group do not show marked sexual dimorphism, a fact

that contrasts strikingly with the situation characteristic of other tribes. Most of

these species form balls at the food source. Subsequently, this ball is rolled and

buried at a shallow depth. This same process is followed by many of the species

for nesting; ball-rolling is then carried out by a bisexual pair.

The subtribe Canthonina has a Gondwanian distribution. More than half of all

the species of Canthonina are Neotropical, as well as 27 of 28 American genera

(Halffter and Edmonds, 1982). The subtribe is also richly represented in Australia,

South Africa and Madagascar. The other Ethiopian and Oriental faunas are less

rich, a consequence of strong competition or perhaps ecological replacement from

other tribes of ball-rolling beetles.

Quaest. Ent., 1990, 26(1)

Insect Invasion Waves

5

ASSUMPTIONS

The biogeographic analyses of the MTZ by Halffter (op. cit .) and the

taxonomic study of the American Canthonina by Halffter (1958, 1961) and Halffter

and Martinez (1966, 1967, 1968, 1977), suggests that some North American

species of the subgenus Canthon (C. humectus, obliquus, chalcites, imitator,

pilularius, and vigilans), all species of the closely related subgenus Boreocanthon,

and those of the genus Melanocanthon originate from an ancestral Miocene

invasion from South America, and that the species C. indigaceus, C. cyanellus and C.

morsei± of subgenus Canthon , as well as the subgenus Glaphyrocanthon, represent

recent Plio-Pleistocene invaders into North America. We accept these

suggestions.

PREDICTIONS

The MTZ biogeographic scenario as proposed by Halffter (op. cit.) allows

predictions for Canthon, Boreocanthon and Melanocanthon, concerning

phylogenesis and ecological evolution in North America. These predictions can be

tested by Methods outlined below.

Prediction 1.

Beginning with the proposed migration wave from South into North America

by an ancestral Canthon component in Miocene time along a discontinuous corridor

formed by islands (Briggs, 1987; Donnelly, 1988), this area would have varied in

its ability to allow the passage of some elements into North America, and thus

would have isolated these elements from their ancestral lineages in South America.

Therefore, this Miocene component would have been subjected to the same

macroecological pressures for a prolonged period of time, would have lived under

the same physiographic conditions and would have had a common biogeographic

history in North America.

Prediction 2.

Several biomes originated in North America during Miocene times, such as

arid zones, grasslands, pinon-oak woodland and hardwood forest (Axelrod, 1975,

1979). Other biomes started evolving in North America in the Pliocene, such as

pine and pine-oak forests (Axelrod, 1975, 1979). These emerging biomes would

represent newly-available habitats and one could expect species diversification to

occur in the invading line(s) (Simpson, 1953). If this species diversification

originated from one common Miocene ancestor, one can consider also on the basis

of phylogenetic tenets (Hennig, 1966; Wiley, 1981) that the derived species form

groupings which are phylogenetically related to each other and reflect the historical

course of speciation.

Quaest. Ent., 1990, 26(1)

6

Kohlmann & Halffter

Prediction 3.

Those species that are thought to be derived from this Miocene invasion and

that have diversified in the new adaptive zones, exhibit relationships that reflect

some degree of congruence with the historical sequence of biome appearance in

which the species now live, as has been postulated for area cladograms (Rosen,

1978, 1979).

Prediction 4.

Species C. cyanellus, C. indigaceus and C. morsei, of subgenus Canthon, are

considered to be derived from a Plio-Pleistocene invasion from South America

(Halffter, 1961, 1962, 1964, 1972, 1974, 1976; Halffter and Martinez, 1977)

associated with several types of tropical forest, and in particular with rainforest.

The association with the tropical forest suggests that those three Canthon species

are South American taxa which have enjoyed range expansions during and after the

closing of the Isthmus of Panama (Liebherr, 1988). These tropical species should

be distantly related in phylogenetic terms to the members of the Miocene invasion

waves, because they evolved under a different set of ecological and biogeographic

conditions. This permits that these species branch at the base of the species-tree

and cause the cladogram branching sequence to be in disorder in relation to the age

of appearance of the different biomes in which the species now live and therefore

be incongruent with the biogeographic history of the area, since they may represent

different phyletic lines. This prediction and those put forward for the Miocene

lineage of invasion can be compared with the results of the concordance between

the cladogram and the biome evolution sequence as a test of the MTZ insect

biogeographical hypothesis.

MATERIAL AND METHODS

Material

This study is the result of the examination of several thousands of specimens

of American Canthonina. All North American species of the subgenera Canthon and

Boreocanthon, the majority of Glaphyrocanthon, as well as the totality of the

M el ano canthon, were studied for the cladistic analysis Many representative

species of American genera and subgenera of Canthonina were studied for the

phenetic analysis.

All this material comes from the G. Halffter collection, Xalapa, Mexico, which

is one of the best collections for this group.

Terminal Taxa.. — We accept the diagnoses given by Halffter and Martinez

(1977) for the genus Canthon and the subgenera Canthon , Boreocanthon and

Glaphyrocanthon as well as the genus Melanocanthon. With the exception of some

mentioned later on, the species included in these genera and subgenera in North

America are those assigned in the earlier papers of Halffter and Martinez (1977)

and Halffter (1958, 1961).

Terminal taxa so defined are listed in Table II. Halffter and Martinez (1977)

consider Canthon ( Boreocanthon ) bisignatus Balthasar as a doubtful member of the

Quaes t. Ent., 1990, 26(1)

Insect Invasion Waves

7

subgenus Boreocanthon. This species is not included in the analysis as only a

scanty description of it is available and we have been unable to obtain additional

material. A second species of subgenus Boreocanthon that has been excluded is B.

nyctelius Bates, since Howden (1966) considers it to be conspecific with C.

puncticollis Le Conte. Lastly C.forreri Bates is almost identical to C. integricollis

Schaeffer, and for this reason it is not included in our analysis.

We consider Glaphyrocanthon as a convenient external group to root the

cladogram. This subgenus is composed mainly of species of South American origin

with a limited penetration into North America. A phenetic analysis (Fig. 1 and

Appendix) shows that Glaphyrocanthon has rather distant relationships with the

other genera and subgenera considered in this analysis and can be considered safely

as an outgroup.

Cladistic Methods.

The cladistic analysis based on the data presented in Tables I and II was

carried out using the PAUP (Phylogenetic Analysis Using Parsimony) computer

program, version 2.4.0 (1985), distributed by Dr. L. S wofford (Illinois Natural

History Survey, 607 East Peabody Drive, Champaign, Illinois 61820), and run on a

VAX 8700 computer employing the following options: NOTU=29; NCHAR=29;

ROOT=OUTGROUP; GO/SWAP=GLOBAL; MULPARS; CONFILE;

MAXTREE=50; ALL CHARACTERS UNORDERED. Subsequently Swofford's

CONTREE program (Version 1/3/86, distributed with PAUP) was used to

calculate Adams and strict consensus trees from the multiple trees that resulted from

the PAUP analysis.

Characters. — Of the twenty-nine characters employed, twenty-three were

binary and six multistate. Unordered multistate characters represent no problem for

a cladistic analysis using PAUP. The majority of characters are derived from

Halffter (1958, 1961) but, some of them derive from personal (B.K.) observations.

For the present analysis all characters were coded as unordered, since sister

groups in and outside Canthonina are not yet defined, thus rendering the character

transformation series unknown. Character polarities can be determined

subsequently by rooting the tree with an outgroup using the parsimony criterion

(S wofford, 1985). However, Meacham (1984) has argued that character polarities

are not known with certainty. We consider that an analysis of character polarity

under such circumstances would not be very informative, since we are analyzing

only a few species and the results could be misleading. For this reason we have not

attempted to determine character polarities.

Quaest. Ent., 1990, 26(1)

8

Kohlmann & Halffter

Fig. 1. Phenogram of the genera and subgenera of the American Canthonina. The scale

measures the dissimilarity (D) between taxa based on the Manhattan metric. The phenogram has

been constructed according to the UPGMA clustering procedure.

Quaest. Ent., 1990, 26(1)

Insect Invasion Waves

9

Old Miocenic Invasion

Evolution in North America

Modern Neotropical

Invasion— Pliocene

And Onwards

Melanocanthon

C. (Boreocanthon)

C. (Canthon)

Fig. 2. Strict consensus cladogram of the North American species of the subgenus Canthon and

related taxa of the subgenus Boreocanthon and the genus Melanocanthon. The distribution of the

taxa in different biomes is indicated as well as the inferred age of appearance of the biome in

North America. The age of appearance of biomes correlated with the greatest number of

associated species is in italics. The proposed invasion time of the different lines into North

America are also indicated. Species marked with a star represent outgroups used for rooting the

cladogram. The length of the cladogram branches have no meaning.

Quaest. Ent ., 1990, 26(1)

10

Kohlmann & Halffter

Zoogeographic Methods.

The subject of this paper is to examine further evidence regarding the

interpretation of two different dispersal waves in the speciation process of the

subgenus Canthon and its closely related taxa Boreocanthon and Melanocanthon in

relation to the biogeographical history of the MTZ as put forward by Halffter.

Broadly, the biogeographic history of the dung-rollers under study in North

America is inferred from a cladistic analysis and related to the sequence of the age

of appearance of the different biomes where they are distributed now. More

particularly, the cladogram of all species pertaining to a taxon (including or not

other related taxa) which is distributed in defined study areas is constructed. Then

we relate the cladogram with the sequence of events suggested by the age of

appearance of the different biomes where the dung-rollers live now and compare it

with the predictions stemming from the MTZ hypotheses, in order to gain some

insight into vicariant and dispersal events that have taken place in the area.

Two other approaches for reconstructing the biogeographic and ecological

history of taxa using a phylogenetic-tree analysis have been proposed (Brooks,

1985, Legendre, 1986). Brooks' method treats species as characters of the areas in

which they occur and lineages of species are thus considered transformation series

linking different areas in an historical pattern. Legendre (1986) reconstructs the

dispersal of a community into adjacent territories by using data of species

presence/absence, obtaining tree-like structures of dispersal from a single trunk.

Connor (1988) analyzes and gives a summary of the bases for inferring the

historical dynamics of biogeographic distributions using phylogenetic methods.

For vicariance biogeography (Nelson, 1973, 1975, 1978) incongruent patterns

between the area cladograms of two or more groups may be interpreted as

dispersal, but different modes of speciation may also be invoked as an explanation

(Wiley, 1981; Wiley and Mayden, 1985).

RESULTS

Cladogram Construction.

A maximum limit was set of 50 equally parsimonious trees to be retained for

analysis. The analysis resulted in 50 such trees with length=57. Consensus trees

using the strict method (Rohlf, 1982) and the method of Adams (1972) were

obtained. The result of the strict consensus tree is depicted in Fig. 2. It has a

consensus fork index (Colless, 1980) of CF=0.704. The consensus tree of Adams

is not shown, because it is very similar to the first one. The only difference is in

species C. lecontei and C. melanus forming a trichotomy with the rest of the

Boreocanthon species group, instead of being part of it, after having branched from

C. simplex.

Correlation with Biomes

Maps of the different taxa recognized by Halffter (1958, 1961) and Halffter

and Martinez (1977) have been elaborated based on personal collections and

publications (Halffter, 1958, 1961; Halffter, and Martinez, 1977; Howden, 1966,

Woodruff, 1973). The biomes where the different taxa are distributed generally

Quae st. Ent., 1990, 26(1)

Insect Invasion Waves

11

have been indicated in the cladogram as well as the probable age of biome

appearance (Fig. 2).

TESTS OF ZOOGEOGRAPHIC PREDICTIONS

Taxonomic Congruence

The results of the consensus cladogram have been compared with the accepted

classification of the North American groups (Halffter 1958, 1961). These original

groups were not based solely on phylogenetic considerations, but represent a

classification combining phylogenetic information, morphological distinctiveness

and ease of identification. Therefore, it is not surprising that the cladogram does

not recover the exact grouping. However, the species sequence of the cladogram

agrees in terms of grouping. The cladogram reveals that Canthon (C.) obliquus and

Canthon ( B .) coahuilensis have a somewhat isolated position regarding the old

invasion line of American Canthonina. This situation will be considered in

taxonomic terms in a subsequent systematic treatment of the group.

Historical Congruence.

The consensus cladogram (Fig. 2) shows two distinct groups, one composed

by the species of the subgenus Glaphyrocanthon, which have been used as

outgroups for rooting the cladogram and secondly, the ingroup, formed by the

subgenus Canthon and related taxa, the subgenus Boreocanthon and the genus

Melancanthon. We will start our discussion with the ingroup.

We observe, as expected from our predictions based on the biogeographic

hypotheses of the MTZ, that species C. morsei, C. cyanellus, and C. indigaceus

branch out at the base of our ingroup and that they are distributed in tropical

biomes of Plio-Pleistocene invasion into North America (Fig. 2, Map 1). Their

branching order is therefore incongruent with the rest of the ingroup sequence of

events if these three species had derived from the same Miocene ancestor that

invaded North America from South America and had diversified in biomes that had

originated (not invaded) in North America. This lack of congruence supports the

assumption that these three species represent separate invaders into North America

(Halffter, op. cit.). As can be seen from the consensus cladogram (Fig. 2), none of

the three taxa are grouped within a clade but form isolated lines in the study area.

They represent actually, northern ends of groups whose species richness is centered

in South America. Canthon morsei (Map 1) is a member of the juvencus line

(Halffter and Martinez, 1977), all other species of which are South American;

Canthon cyanellus (Map 1) stems from Colombia, Peru and Venezuela (Halffter,

1961) and its line has a major diversification in South America. These two taxa

more probably invaded North America in conjunction with rainforest dispersal into

the area and from there C. cyanellus expanded its range into other types of tropical

forests. Canthon indigaceus (Map 2) stems from a diversification process in

tropical Mexico (Halffter, 1961). Its three subspecies are limited to tropical

conditions, in accordance with the pattern that Halffter (op. cit.) has designated

typical Neotropical and which is essentially modern in the biogeographic history

Quae st. Ent ., 1990, 26(1)

12

Kohlmann & Halffter

of the MTZ. All these species are isolated from the Mexican and North American

Canthonina (Fig. 2) and are only distantly related to them.

The outgroup, subgenus Glaphyrocanthon (Map 3), shows the same type of

distribution and biome association as the three above-mentioned invading species.

The species C. (G.) subhyalinus is distributed from northern Amazonia to southern

tropical Mexico and together with C. (G.) euryscelis, which is distributed in

Central American rainforests, they belong to a line with greatest species diversity

in the northern part of Amazonia (Martinez, Halffter and Kohlmann, unpubl.).

Canthon (G .) femoralis stems from Colombia (Martinez and Halffter, 1972), and C.

viridis belongs to a species group whose greatest richness is South American

(Martinez, Halffter and Halffter, 1964). The northern presence of Glaphyrocanthon

suggests a concordant expansion with the rainforest, sometimes associated with

subspecies formation and penetration into eastern North America (C. viridis ;

Martinez, Halffter and Halffter, 1964). Several other genera of Canthonina, such as

Deltochilum, Cryptocanthon, Pseudocanthon, Malagoniella, and Megathoposoma

(Halffter and Martinez, 1966, 1967, 1968, 1977; Howden, 1973) follow a pattern

of association with tropical biomes, particularly rainforest, reinforcing the Plio-

Pleistocene insect invasion hypotheses into the MTZ from South America

(Halffter, op. cit.}.

The consensus cladogram (Fig. 2) shows a recognizable group formed by two

main branches, one with Canthon species and the other with mostly Boreocanthon

and Melanocanthon species. This group encompasses all those species with an

origin stemming from a postulated Miocene invasion into North America. The

group conforms also with our prediction that most of the species diversification

coincides with the sequence of biome appearances in North America, Miocene

events at the root of the group and subsequently a trend toward species

association with Pliocene biomes.

The Canthon branch includes the " humectus " and " pilularius " lines of

Halffter (1961) and relates them mainly to Miocene events. Canthon humectus is

distributed in the Mexican Highland in grasslands and arid areas with the

exception of the most arid zones, and the highlands of Oaxaca, Chiapas and

Guatemala (Map 4). As Halffter (op. cit.) indicates, Neotropical species of recent

migration have not invaded the highlands; those that do, arrived at the area before

its actual rising, a phenomenon that started during the Miocene. The later

disruption of their area is attributed by Halffter (op. cit.) to the aridity process,

which is more recent (Heine, 1973). The pilularius line must have colonized the east

of the USA during Miocene or somewhat later, but not in recent times. Its actual

distribution covers the grasslands of the Great Plains and the forests of the east

and south of the country; its distribution nucleus is centered on deciduous

hardwood forests (Map 5). Axelrod (1979) indicates that during Middle

Oligocene cold winters had eliminated the majority of evergreen dicotyledons in

eastern USA, leaving only a deciduous hardwood forest. It was not until the Plio-

Pleistocene that prairies and pine forests started to spread (Axelrod, 1979). It is

Quae st. Ent., 1990, 26(1)

Insect Invasion Waves

13

possible therefore, that the pilularius line invaded this sort of habitat from the

deciduous hardwood forest, explaining the correspondence with the present-day

distribution of all four species.

The other branch of the proposed Miocene invasion group includes Canthon

obliquus, the subgenus Boreocanthon and the genus Melanocanthon. Canthon

obliquus is unique (Map 6). It is confined to a relict, deciduous tropical forest

(Rzedowski, 1978; Arriaga and Ortega, 1988) at the Sierra de la Laguna in Baja

California, having become isolated by the drift process of the Baja California

Peninsula. The rift started by late Miocene (Karig and Jensky, 1972) and by early

Pliocene separation was very advanced. This species is a relict, since the tropical

deciduous forest in which it now lives, was well established in south-central USA

during Miocene (Axelrod, 1979) and from that moment it started retreating

towards Mexico, its area being steadily reduced. Canthon obliquus thence would

have become isolated in a remaining island of deciduous tropical forest. Later on,

all of the peninsula, with the exception of the southernmost mountains, suffered from

desertification and invasions of biota adapted to these conditions, as for C. ( B .)

puncticollis. A similar scheme is known for the Bursera (Burseraceae) species from

Baja California (Kohlmann and Sanchez-Colon, 1984).

The ancestor of Melanocanthon-Boreocanthon would have been distributed

originally in northern Mexico and south-central USA, while the area was covered

by tropical deciduous forest. By the end of Miocene the forest started to be

replaced by grasslands and pinon-oak woodland, a process that was advanced in

the Pliocene by the spread of pine forest and very recently the appearance of

deserts. Most Boreocanthon species (Map 6) live in this area, which leads us to

think that this was the main evolutionary arena for this group. Nowadays, many of

the species live in grasslands of the Great Plains (C. simplex, C. lecontei, C.

integricollis, C. mixtus, C. praticola, and C. ebenus) or in arid zones (C. puncticollis

and C. ateuchiceps). This last species seems to be a recent invader of the arid

zones of Puebla, Morelos and Oaxaca following the scheme proposed by Axelrod

(1979), where endemic species of the southern arid zones of Mexico are recent

relicts, which have been pushed into these areas by climatic events of the late

Cenozoic (Heine, 1973). One species, C. melanus, apparently is confined largely to

the Arizona mountains in pinon-oak woodland,, although there is one record from

Guaymas, Sonora, for the coastal desert. Two species, C. probus and C.

depressipennis, follow a similar pattern to Melanocanthon ; in addition to being

present in the Great Plains, both penetrate into eastern USA through the conifer

forest corridor. This distribution could represent an invasion from the grasslands

into the conifer forest corridor. With the exception of these two species, all the

rest of Boreocanthon is associated with biomes that originated in the Miocene.

The last group, Melanocanthon (Map 7) is associated mostly with Pliocene

biomes. The only species associated exclusively with a Miocene biome is M .

nigricornis, which is distributed in the grasslands of the Great Plains. The

remainder occur in the conifer forest corridor in the south and east of the USA (M .

Quae st. Ent., 1990, 26(1)

14

Kohlmann & Halffter

bispinatus ); in the grasslands and conifer forests of Texas and northern Florida

(M. granulifer)-, or else in northern Florida (M. punctaticollis). The invasion and

speciation process seems to have shifted to the east in this group, relative to the

other south-central diversification areas and follows the pine and pine-oak forest

corridors that established themselves around the deciduous forests. It seems to be

the only member of the Miocene group whose speciation events are recent

(Pliocene), whereas in Canthon and Boreocanthon some species seem to have

invaded but not speciated in Pliocene biomes. The invasion and speciation pattern

in Melanocanthon is somewhat similar to that of Ateuchus (Coleoptera:

Scarabaeidae), since it is proposed that the latter genus invaded and diversified in

an approximately similar zone in Plio-Pleistocene time (Kohlmann and Halffter,

1988).

Quaest. Ent., 1990, 26(1)

116 114 112 110 108 106 104 102 100

Insect Invasion Waves

15

Quaest. Ent 1990, 26(1)

represent the known localities for Canthon morsei.

102 100

16

Kohlmann & Halffter

Quaest. Ent., 1990, 26(1)

lap 2. Geographical distribution of the C. ( Canthon ) indigaceus line, in northern Middle America and in southern United States of America.

Insect Invasion Waves

17

Quaest. Ent 1990, 26(1)

Map 3. Geographical distribution of C. ( Glaphyrocanthon ) in northern Middle America and in southern United States of America.

108 106 104 102 100

18

Kohlmann & Halffter

CO TJ-

CM CVJ

8

8

8

a

Quaest. Ent., 1990, 26(1)

Man 4. Geoeraohical distribution of the C. ( Canthon ) humectus line in Mexico and Guatemala.

Insect Invasion Waves

19

Quaest. Ent 1990, 26(1)

Map 5. Geographical distribution of the C. ( Canthon ) pilularius line in Mexico and in the United States of America.

20

Kohlmann & Halffter

J

Quaest. Ent., 1990, 26(1)

Map 6. Geographical distribution of C. ( Boreocanthon ) in Mexico and in the United States of America. The square at the bottom end of the peninsula of

Baja California represents the distribution of C. ( Canthon ) obliquus.

125 120 115 110 105 100

Insect Invasion Waves

21

Quaest. Ent., 1990, 26(1)

115 110 105 100 95 90 85

Map 7. Geographical distribution of Melanocanthon in eastern United States of America.

22

Kohlmann & Halffter

CONCLUDING STATEMENT

We observe from the previous analysis that the concordance of the consensus

cladogram with the age of appearance of the different biomes is congruent with our

predictions based on the biogeographical hypotheses of the MTZ. Therefore this

fact may be considered supportive of the supposition of two different insect

invasion waves from South into North America. Moreover, the analysis points to a

strong correlation between a great diversification of the Miocene invasion lineage

with biomes that started to originate in the Miocene. Far fewer species are

associated with Pliocene biomes, and in several of such seem to represent

secondary invasions. Finally, no species stemming from the proposed Miocene

invasion have been found in Pleistocene biomes (tropical rainforest). This would

suggest that ball-rollers of the Miocene invasion have not had enough time to

diversify in this new environment, or that most probably ecological replacement

stemming from the many South-American-derived species living in tropical forests

in North America have precluded species diversification effectively.

This analysis may be corroborated by other studies. However, not many

insect examples with a suspected similar history are known. The scarabaeid genus

Phanaeus seems to follow a related pattern (Halffter, 1962; Edmonds, 1972). This

genus is being revised by Edmonds (pers. corns.), and a similar analysis may be

very illuminating. Savage (1982) already has presented evidence for a late

Cretaceous and a Plio-Pleistocene dispersal of South American fishes, amphibians

and reptiles into Central America. For floras. Gentry (1982) also proposes two

migrations of Neotropical floristic elements from South to North America. Gentry

(1982) considers that the two main waves occurred at the end of the Cretaceous

and then again in Pliocene. The history of the dung-rollers seems to be more similar

to the one postulated for mammals (Stehli and Webb, 1985), where an increasing

frequency of connection started in Miocene and culminated in Plio-Pleistocene. At

any rate the existence of two different invasion processes is supported by the

present analysis for the dung-rollers, although the exact timing of the first wave

may be debatable still. In conclusion we would consider that the approach

presented here may be of help in other situations where several invasion or

dispersal waves in different taxa are suspected.

ACKNOWLEDGEMENTS

We thank Professor B. John who read an early draft of the manuscript and

D. Colgan for reviewing and commenting on this paper. Special thanks go to D.

Colless for his detailed reading of the manuscript, for providing much help during

the cladistic analysis and introducing one of us (B.K.) to the use of PAUP.

Constructive criticism of two anonymous reviewers is also thankfully

acknowledged.

Quaest. Ent., 1990, 26(1)

Insect Invasion Waves

23

REFERENCES CITED

Adams, E.N. 1972. Consensus techniques and the comparison of phylogenetic

trees. Systematic Zoology. 21: 390-397.

Arriaga, L. and A. Ortega (eds.) 1988. La Sierra de la Laguna de Baja California

Sur. Centro de Investigaciones Biologicas de Baja California Sur, La Paz, Baja

California Sur, Mexico. 203 p.

Axelrod, D.I. 1975. Evolution and biogeography of Madrean-Tethyan sclerophyll

vegetation. Evolution 62: 280-334.

Axelrod, D.I. 1979. Age and origin of Sonoran Desert vegetation. Occasional

Papers of the California Academy of Sciences. 132: 1-74

Briggs, J.C. 1987. Biogeography and Plate Tectonics. Developments in

Paleontology and Stratigraphy, 10. Elsevier, Amsterdam. 204 p.

Brooks, D.R. 1985. Historial ecology: A new approach to studying the evolution

of ecological associations. Annals of the Missouri Botanical Garden. 72:

660-680.

Colless, D.H. 1980. Congruence between morphometric and allozyme data for

Menidia species: A Reappraisal. Systematic Zoology. 29: 288-289.

Connor, E.F. 1988. Fossils, phenetics, and phylogenetics: Inferring the historial

dynamics of biogeographic distributions, p. 254-269. In: Zoogeography of

Caribbean Insects, J. K. Liebherr (ed.). Cornell Univ. Press, Ithaca, N.Y. 285

P-

Donnelly, T.W. 1988. Geologic constraints on Caribbean biogeography p. 15-37.

In: Zoogeography of Caribbean Insects, J.K. Liebherr (ed.). Cornell Univ.

Press, Ithaca, N.Y. 285 p.

Edmonds, W.D. 1972. Comparative skeletal morphology, systematics and

evolution of the Phanaeinae dung beetles (Coleoptera: Scarabaeidae).

University of Kansas Science Bulletin. 49: 731-874.

Gentry, A.H. 1982. Neotropical floristic diversity: Phytogeographical

connections between Central and South America, Pleistocene climatic

fluctuations, or an accident of the Andean orogeny. Annals of the Missouri

Botanical Garden. 69: 557-593.

Germeraad, J.H., C.A. Kopping and J. Muller, 1968. Palynology of Tertiary

sediments from tropical areas. Reviews of Paleobotany and Palynology. 6:

189-348.

Graham, A. 1973. History of the arborescent temperate element in the Latin

American biota, p. 301-314. In: Vegetation and Vegetational History of

Northern Latin America, A. Graham (ed.). Elsevier, Amsterdam. 393 p.

Graham, A. 1981. Diversification beyond the Amazon Basin, p. 78-90. In:

Biological Diversification in the Tropics, G. Prance (ed.). Plenum Press, N.Y.

714 p.

Halffter, G. 1958. Dos neuvos generos de Canthonini (Coleoptera: Scarabaeidea).

Ciencia (Mex.). 17: 207-212.

Halffter, G. 1961. Monografia de las especies norteamericanas del genero Canthon

Hoffsg. (Coleoptera: Scarabaeidae). Ciencia (Mex.). 20: 225-320.

Hallfter, G. 1962. Explicacion preliminar de la distribucion geografica de los

Scarabaeidae mexicanos. Acta Zoologica Mexicana. 5: 1-17.

Halffter, G. 1964. La entomofauna americana, ideas acerca de su origen y

distribucion. Folia Entomologica Mexicana. 6: 1-108.

Halffter, G. 1972. Elements anciens de l'entomofaune neotropicale. Ses implications

biogeographiques. In: Biogeographie et liaisons intercontinentales un cours

Quaest. Ent., 1990, 26(1)

24

Kohlmann & Halffter

du Mesozoique. XVII eme Congres International de Zoologie, Monte Carlo.

40 p.

Halffter, G. 1974. Elements anciens de l'entomofaune neotropicale. Ses implications

biogeographiques. Quaestiones entomologicae. 10: 223-262.

Halffter, G. 1976. Distribucion de los insectos en la Zona de Transicion Mexicana.

Folio Entomologica Mexicana. 35: 1-64.

Halffter, G. 1978. Un nuevo patron de dispersion en la Zona de Transicion

Mexicana: El Mesoamericano de Montana. Folia Entomologica Mexicana.

39-40: 210-222.

Halffter, G. 1987. Biogeography of the montane entomofauna of Mexico and

Central America. Annual Review of Entomology. 32: 95-1 14.

Halffter, G. and W.D. Edmonds. 1982. The Nesting Behaviour of Dung Beetles

(Scarabaeidae). An Ecological and Evolutive Approach. Pubis. Instituto de

Ecologfa 10, Mexico, D.F. 176p.

Hallffter, G. and A. Martinez. 1966 Revision monografica de los Canthonina

americanos (la. Parte). Revista de la Sociedad Mexicana de Historia Natural.

27: 89-171

Halffter, G. and A. Martinez. 1967. Revision monografica de los Canthonina

americanos. (2a. Parte). Revista de la Sodiedad Mexicana de Historia

Natural. 28:79-116.

Halffter, G. and A. Martinez. 1968. Revision monografica de los Canthonina

americanos (3a. Parte). Revista de la Sociedad Mexicana de Historia Natural.

29: 209-290.

Halffter, G. and A. Martinez. 1977. Revision monografica de los Canthonina

americanos (4a. Parte). Folia Entomologica Mexicana. 39: 29-108.

Heine, K. 1973. Variaciones mas importantes del clima durante los ultimos 40,000

anos. Communicaciones Proyecto Puebla-Tlaxcala (Fundacion Alemana para la

Investigacfon Cientffica). 7: 51-58.

Hennig, W. 1966. Phylogenetic systematics. University of Illinois Press, Urbana.

263 p.

Howden, H.F. 1966. Notes on Canthonini of the "Biologia Centrali-Americana" and

descriptions of new species (Coleoptera: Scarabaeidae). Canadian

Entomologist. 98: 725-741.

Howden, H.F. 1973. Revision of the New World genus Cryptocanthon Balthasar

(Coleoptera: Scarabaeidae). Canadian Journal of Zoology. 60:2713-2724.

Kaneps, A.C. 1979. Gulf Stream: Velocity fluctuations during Late Cenezoic.

Science. 204: 297-301.

Karig, D.E. and W. Jensky, 1972. The proto-Gulf of California. Earth and

Planetary Science Letters. 17: 169-174.

Keigwin, L.D. 1978. Pliocene closing of the Isthmus of Panama, based on

biostratigraphic evidence from nearby Pacific Ocean and Caribbean sea cores.

Geology. 6: 630-634.

Kohlmann, B. and G. Halffter. 1988. Cladistic and biogeographical analysis of

Ateuchus (Coleoptera: Scarabaeidae) of Mexico and the United States. Folia

Entomologica Mexicana. 74: 109-130.

Kohlmann, B. and S. Sanchez-Colon. 1984. Estudio areografico del genero Bursera

Jacq. ex L. (Burseraceae) en Mexico: Una smtesis de metodos. p. 41-120. In:

Metodos Cuantitativos en la Biogeograffa, E. Ezcurra and B. Kohlmann

(eds.). Pubis. Instituto de Ecologfa 12, Mexico, D.F. 125 p.

Legendre, P. 1986. Reconstructing biogeographical history using phylogenetic-

tree analysis of community structure. Systematic Zoology. 35: 68-80.

Quaest. Ent., 1990, 26(1)

Insect Invasion Waves

25

Liebherr, J.K. 1988. The Caribbean: Fertile ground for zoogeography, p. 1-14. In:

Zoogeography of Caribbean Insects, J.K. Liebherr (ed.). Cornell Univ. Press,

Ithaca, N.Y. 285 p.

Marshall, L.G., S.D. Webb, J.J. Sepkoski and D.M. Raup. 1982. Mammalian

evolution and the great American interchange. Science. 215:1351-1357

Martinez, A. and G. Halffter. 1972. New taxa of American Canthonina (Coleoptera:

Scarabaeidae). Entomologische Arbeiten des Museums Georg Frey. 23: 33-

66.

Martinez, A., G. Halffter and V. Halffter. 1964. Notas sobre el genero

Glaphyrocanthon (Coleoptera: Scarabaeidae) Acta Zoologica Mexicana. 7:

1-42

Meacham, C.A. 1984. The role of hypothesized direction of characters in the

estimation of evolutionary history. Taxon 33: 26-38.

Nelson, G.J. 1973. Comments on Leon Croizat's biogeography. Systematic

Zoology. 22: 312-320.

Nelson, G.J. 1975. Historical biogeography: An alternative formulation.

Systematic Zoology. 23: 555-558.

Nelson, G.J. 1978. From Candolle to Croizat: Comments on the history of

biogeography. Journal of Historical Biology. 11: 293-329.

Prance, G.T. 1982. A review of the phytogeographic evidences for Pleistocene

climate changes in the neotropics. Annals of the Missouri Botanical Garden.

69: 594-624.

Rohlf, F.J. 1982. Consensus indices for comparing classifications. Mathematical

Biosciences. 59: 131-144.

Rosen, D.E. 1978. Vicariant patterns and historical explanation in biogeography.

. Systematic Zoology. 27: 159-188.

Rosen, D.E. 1979. Fishes from the uplands and intermontane basins of Guatemala:

Revisionary studies and comparative geography. Bulletin of the American

Museum of Natural History. 162: 267-376.

Rzedowski, J. 1978. Vegetacion de Mexico. Editorial Limusa, Mexico, D.F. 432

P-

Savage, J.M. 1982. The enigma of the Central American herpetofauna: Dispersals

or vicariance? Annals of the Missouri Botanical Garden. 69: 464-547.

Simpson, G.G. 1953. The Major Features of Evolution. Columbia Univ. Press. N.Y.

434 p.

Stehli, F.G. and S.D. Webb, eds. 1985. The Great American Biotic Interchange.

Plenum Press, N.Y. 532 p.

Swofford, D.L. 1985. PAUP phylogenetic analysis using parsimony. Version

2.4.0. Illinois Natural History Survey, Champaign, Illinois.

Toledo, V.M. 1976. Los cambios climaticos del Pleistoceno y sus efectos sobre la

vegetacion natural calida y humeda de Mexico. M.Sc. Thesis, Faculty of

Sciences, UNAM Mexico. 74 p.

Toledo, V.M. 1982. Pleistocene changes of vegetation in tropical Mexico, p. 92-

111. In: Biological diversification in the Tropics, G.T. Prance (ed.).

Columbia Univ. Press, N.Y. 714 p.

Webb, S.D. 1977. A history of savanna vertebrates in the New World. Part I:

North America. Annual Review of Ecology and Systematics. 8: 355-380.

Webb, S.D. 1978. A history of savanna vertebrates in the New World. Part II

South America and the great interchange. Annual Review of Ecology and

Systematics. 9: 393-426.

Wiley, E.O. 1981. Phylogenetics. The Theory and Practice of Phylogenetic

Systematics. Wiley-Interscience, N.Y. 439 p.

Quaest. Ent., 1990, 26(1)

26

Kohlmann & Halffter

Wiley, E.O. and R.L. Mayden. 1985. Species and speciation in phylogenetic

systematics, with examples from the North American fish fauna. Annals of the

Missouri Botanical Garden. 72: 596-635.

Woodruff, R.E. 1973. The scarab beetles of Florida. Florida Department of

Agriculture and Consumer Services. Division of Plant Industry, Gainesville,

Florida. 220 p.

APPENDIX

The phenetic analysis of the American canthonine genera and subgenera of

Canthon (Fig. 1) is based on the work of Halffter and Martinez (1966, 1967, 1968

and 1977). Seventy-one characters of external and internal structures have been

taken into consideration; thirty-two of them are binary and thirty-nine are

multistate. The distribution of these characters and their description are available

from the authors.

Some problems exist in the analysis. The male of Canthotrypes is not known.

The aedeagus of Zonocopris and Deltepilissus could not be examined.

The phenogram (Fig. 1) was elaborated using the UPGMA method. As a

distance measure the Manhattan coefficient was used. This coefficient implies the

existence of orthogonal axes. The character correlation matrix (not reproduced)

has low correlation values, therefore this metric is acceptable.

The analysis of the phenogram indicates various groups:

Tetraechma-Xenocanthon-P seudepilissus-Canthonidia; Vulcanocanthon-

Holocanthon; Melanocanthon-Boreocanthon -Nesocanthon-Canthorr,

Anisocanthon-Trichocanthon-Scybalocanthon; Deltepilissus-Francmonrosia-

Goniocanthon; Hansreia-P eltecanthon-Scybalophagus ; Zonocopris-

Pseudocanthon; Sylvicanthon-Glaphyrocanthon; Sinapisoma-Paracanthon-

Canthonotrypes-Agamopus ; Megathopa-Megathoposoma; and the following

isolated lines - Canthochilum, Canthonella, Cryptocanthon, Deltochilum,

Eudinopus and the most removed one, Streblopus. It should be mentioned here that

Deltochilum is a very heterogeneous entity.

The ladder structure in the phenogram suggests that the relationships between

taxa is a very gradual one, implying that the group is very homogeneous in its

morphological characteristics. The low morphological character correlation seems

to indicate that there has not been any tendency to form groups of associated

characters.

Until a detailed cladistic analysis is performed, the present study should be

considered as preliminary.

Quaest. Ent., 1990, 26(1)

Insect Invasion Waves

27

TABLE 1

Morphological Characters Used in the Study of North American Canthonina

1 Number of elytral striae. Usually eight: a; usually nine: b.

2. - First article of the metatarsi. Usually bigger than the second: a; usually shorter

than the second: b; equal: c.

3. - Anterior margin of the metafemur. Without margin: a; with margin: b.

4. - Elytra. Translucent: a; opaque: b.

5. - Clypeo-genal suture. With broad external notch: a; with narrow external notch:

b.

6. - Subhumeral stria. Usually not keeled: a; usually keeled: b; usually slightly

keeled: c.

7. - Number of clypeal teeth. No teeth: a; bidentate: b; quadridentate: c.

8. - Clypeal teeth separation. Narrowly separated: a; widely separated: b.

9. - Separation between proepisternum and proepimeron. Without a keel: a;

keeled: b; slightly keeled: c.

10. - Proepisternum. Shallowly excavated: a; excavated: b.

1 l.-Prescutellar impression. Usually not marked: a; usually marked: b.

12. - Dorsal surface. Usually not granulated: a; usually granulated: b.

13. -Granules. Simple: a; like shining spots: b; flattened: c; elongated: d.

14. -Pronotum. Without punctuation: a; with punctuation: b.

15. -Protibia straightly truncate: a; obliquely truncate: b.

16. -Eyes. Narrow: a; wide: b.

17. -Metatibia. With one spine: a; with two spines: b.

18. - Separation between gula and submentum. Narrow "V": a; wide "V": b; arched:

c.

19. - Elytral external striae. As deep as the internal ones: a; deeper than the internal

ones: b

20. - Dorsal eye margin. With border: a; without a border: b.

21. -Dorsal eye margin. Bordered by a wide flat band: a; bordered by a raised

fold: b.

22. - Ventral clypeal structure. With a tooth: a; without a tooth: b.

23. -Protibia. Not widened along its internal margin: a; widened along its internal

margin: b.

24. -Mesostemum. Short: a; very wide; b.

25. -External margin of the clypeo-genal suture. Straight or slightly pointed: a;

with an evident tooth: b.

26. -Lateral pronotal margin. Usually not serrated: a; usually serrated: b.

27. -Posterior angle. Pointed: a; blunt: b.

28. -Elytral margins. Slightly curved upwards behind the humeral angle: a; strongly

curved upwards behind the humeral angle: b.

29. -Elytral colour. Orange: a; not orange: b.

?= Character not applicable.

Quae st. Ent., 1990, 26(1)

28

Kohlmann & Halffter

TABLE 2

Character distribution in the North American Canthonina species

Melanocanthon punctaticollis (Schaeffer)

Melanocanthon nigricomis (Say)

Melanocanthon granulifer (Schmidt)

Melanocanthon bispinatus (Robinson)

C. (Boreocanthon) simplex LeConte

C. (Boreocanthon) ateuchiceps Bates

C. (Boreocanthon) depressipennis LeConte

C. (Boreocanthon) ebenus (Say)

C. (Boreocanthon) praticola LeConte

C. (Boreocanthon) integricollis Schaeffer

C. (Boreocanthon) mixtus Robinson

C. (Boreocanthon) lecontei Harold

C. (Boreocanthon) melanus Robinson

C. (Boreocanthon) probus Germar

C. (Boreocanthon) puncticollis LeConte

C. (Boreocanthon) coahuilensis Howden

C. (Canthon) obliquus Horn

C. (Canthon) chalcites (Haldeman)

C. (Canthon) vigilans LeConte

C. (Canthon) imitator Brown

C. (Canthon) pilularius (Linnaeus)

C. (Canthon) humectus (Say)

C. (Canthon) indigaceus Harold

C. (Canthon) cyanellus LeConte

C. (Canthon) morsei Howden

C. (Glaphyrocanthon) subhyalinus Harold

C. (Glaphyrocanthon) euryscelis Bates

C. (Glaphyrocanthon) viridis (de Beauvois)

C. (Glaphyrocanthon) femoralis (Chevrolat)

BAABAACBAAABBBAABAAABABAAABAB

BAABAACBAAABCBAABAAABABAAABAB

BAABAACBAAABABAABAAABABAAABAB

BAABAACBAAABDBAABAAABABAAABAB

BAABAACBAAABBBAAAAAABBBABABAB

BAABBBCBAAAA7BAAAAAAABBABABAB

BAABBACBAAABAAAAAAAABBBABBBAB

BAABBACBAAABAAAAAABABBBABBBAB

BAABBBCBAAABAAAAAABABBBABBBAB

BAABBBCBAAABBBAAAAAABBBABABAB

BAABBBCBAAABBBAAAAAABBBABABBB

BAABABCBAAABBAAAAAAABBBABABAB

BAABABCBAAABBBAAAAAABBBABABAB

BAABBCCBAAABBBAAAAAABBBABABAB

BAABBBCBAABBBBAAAAAABBBABABAB

BAABAACBAAAA7BAAAAAABBBAAABAB

BABBAAA7AABA7BAAAAAAAABBAABAB

BABBAABBAAABAAAAABAAABBAAABAB

BABBAABBAAABAAABABAAABBAAABAB

BABBAABAAAABBAAAABAAABBAAABAB

BABBAABBAAABBAAAABAAABBAAABAB

BABBAABBBBABABAAACAAABBAAABAB

BABBAABAAAAA7BAAACAAABBAAABAB

BABBABCACABA7BBAACAAABBAAABAB

BBBBAABACAAA7BBAACAAABAAAABAB

BBAAAABABBAA7BBAACAAABAAAAAAA

BBAAAABABBAA7BBAACAAABAAAAAAB

BBABAABABBAA7BAAACAAABBAAAAAB

ABABAACABBAA7AAAACAAABAAAAAAB

Quaest. Ent., 1990, 26(1)

Frontispiece. Photograph of habitus of Entomoanty x cyanipennis (Chaudoir), dorsal aspect.

Mexico, Veracruz, NE Catemaco, Los Tuxtlas Biological Station (CNCI). Standardized Body

Length = 4.4 mm.

THE MIDDLE AMERICAN GENERA OF THE TRIBE OZAENINI

WITH NOTES ABOUT THE SPECIES IN SOUTHWESTERN

UNITED STATES AND SELECTED SPECIES FROM MEXICO

George E. Ball

Department of Entomology

University of Alberta

Edmonton, Alberta

Canada T6G 2E3

and

Scott McCleve

2210 13th Street

Douglas, Arizona Quaestiones Entomologicae

85607 U.S.A. 26: 30—116

ABSTRACT

Based on structural features of adults, the following new taxa are described:

Entomoantyx, new genus (type species — Ozaena cyanipennis Chaudoir, 1852);

and Pachyteles (sensu stricto) enischnus, new species (type locality — Mexico,

Jalisco, near Ixtapa). Combined in a single genus, but ranked as subgenera are:

Pachyteles (s. str.) Perty, 1830 (type species — P. striola Perty, 1830);

Goniotropis Gray, 1832 (type species — G. braziliensis Gray, 1832), with its

junior synonym, Scythropasus Chaudoir, 1852 (type species — S. elongata

Chaudoir, 1852); and Tropopsis Solier, 1849 (type species — T. marginicollis

Solier, 1849). The following species-level synonymy is proposed, with the senior

synonym and thus valid name listed first for each combination: Pachyteles

(Goniotropis,) parca LeConte, 1884 (type area — U.S.A., Arizona) = P. beyeri

Notman, 1919 (type locality — Mexico, Baja California Norte, San Felipe);

Pachyteles (s. str.) gyllenhali (Dejean, 1825) (type area — Cuba) = P. verticalis

(Chaudoir, 1848 ) (type area — Colombia) = P. testaceus Horn, 1868 (type

locality — Fort Grant Arizona U.S.A.); and Ozaena lemoulti Banninger, 1931 (type

locality — French Guiana, St. Jean du Maroni) = O. halffteri Ogueta, 1965b (type

locality — Mexico, Veracruz, Tlapacoyan). The genera are characterized in terms of

adults, using defensive secretions and structural features, including chaetotaxy,

antennae, mouthparts (labrum, mandibles, maxillae, and labium), antenna cleaner of

the fore tibia, male genitalia, ovipositor, and internal genitalia of females. To

facilitate future phylogenetic analysis, transformation series were postulated for

each character, using the Metriini (genus Metrius Eschscholtz) as out-group. This

provided a linear series for the genera, from most like to least like Metrius:

Entomoantyx; Pachyteles (s. lat.); Physea; Ozaena; and Platycerozaena. Ozaena and

Platycerozaena are postulated to be sister groups, but relationships to one another

of the remaining genera have not been postulated. The following species are

characterized, in terms of adult features and geographical distribution;

Entomoantyx cyanipennis (Chaudoir); Physea hirta LeConte, 1853 and P. latipes

Schaum, 1864; Pachyteles (Goniotropis) parca LeConte, P. kuntzeni Banninger,

and P. elongatus (Chaudoir); Pachyteles (s. str.) gyllenhali Dejean, P. enischnus, new

species, and P. mexicanus (Chaudoir, 1848); and Ozaena lemoulti Banninger. A key

is provided to these species and to the Middle American genera. Based on general

zoogeographic theory for Middle America, the following hypothesis is developed

to explain the distribution patterns of the Middle American genera and the species

32

Ball and McCleve

whose ranges either enter the United States or are near the U .S. -Mexican border:

Middle America was entered at various times during the Tertiary by the ancestor

0/ Entomoantyx and members of each of the genera; i.e., the only genus to evolve as

such in Middle America was Entomoantyx. Most of the extant species

differentiated in Late Tertiary time as east-west vicar iants, as a result of the

influence of climatic change and mountain building on the ranges of the ancestral

populations. Two species, P gyllenhali and O. lemoulti, arrived in the northern

areas comparatively recently, each becoming widespread during Pleistocene time.

TABLE OF CONTENTS

Introduction . 32

Material and Methods . 35

Historical Aspects . 37

Tribal Synonymy . 37

Classification and Relationships . 37

Comparative Morphology . 4 1

Structural and Biochemical Features . 41

Summary . 76

Taxonomic Treatment . 77

Tribe Ozaenini . 77

Entomoantyx , new genus . 82

Physea Brulle . 8 4

Pachyteles Perty . 8 8

Ozaena Olivier . 98

Platycerozaena Banninger . 1 0 1

Zoogeography . 1 02

The Tribes . 102

The Genera . 103

Ozaenine Species of Southwestern United States and Vicinity . 104

Concluding Statement . 106

Acknowledgements . 107

References Cited . 107

Index to Names of Taxa . 1 1 5

INTRODUCTION

The roots of this study extend back in time to the early 1950's, when the first

author found in the collections of the U. S. National Museum of Natural History a

specimen of Ozaena collected in a Plant Quarantine Station at Nogales, Arizona. As

far as was known then, this record represented a substantial range extension for the

genus, and it seemed not unlikely that the specimen was an accidental import from

the American tropics, to the south. Max Banninger, at that time the foremost

authority on Ozaenini, was consulted, and he reported the identity of the specimen

as Ozaena elevata ( cf Banninger, 1956: 400), a species otherwise known only from

South America. The specimen was labelled by him as Ozaena elevata var.

In 1978, the second author collected a specimen of Ozaena at Pena Blanca,

Pajarito Mountains, southern Arizona — a locality near Nogales, but far enough

away to suggest that the species represented was indeed native to the area.

Comparison of this specimen with the one taken at Nogales showed that the two

were sufficiently similar to be regarded as conspecific. But, what species did they

represent? Having entered into discussion about ozaenines in the United States, we

Quaest. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

33

Figs. 1 — 4. Photographs of habitus, dorsal aspect, of: 1, Metrius contractus Eschscholtz; U.S.A.

California, Siskiyou County, Pickewish Campground (UASM); Standardized Body Length = 10.9

mm. 2, Physea latipes Schaum; Mexico, Venedio (CASC): SBL = 11.1 mm. 3, Pachy teles

(Goniotropis) parca LeConte; U.S.A., Arizona, Santa Cruz County. Santa Rita Mts., Madera

Canyon; SBL = 10.9 mm. 4, P. ( Goniotropis ) kuntzeni (Biinninger); U.S.A. Arizona, Santa Cruz

County, Pena Blanca (UASM); SBL = 15.2 mm.

Quae st. Ent ., 1990, 26(1)

34

Ball and McCleve

Figs. 5 — 8. Photographs of habitus, dorsal aspect, of: 5, P. ( sensu stricto) gyllenhali (Dejean);

U.S.A., Arizona, Graham County, Aravaipa Canyon, 17.7 km. N. Klondyke (UASM); SBL = 4.4

mm. 6, P. (sensu stricto) enischnus, new species; Mexico, Jalisco, nr. Ixtapa, gallery forest SBL =

7.0 mm. 7, Ozaena lemoulti Banninger; U.S.A., Arizona, Pajarito Mts., Pena Blanca; SBL = 18.0

mm. (SMCC); 8, Platycerozaena brevicornis (Bates); French Guiana Monte de Kaw, Piste de

Kaw, Km. 3 (UASM); SBL =11.4 mm.

Quaest. Ent ., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

35

asked this question of one another some seven years ago. Encouraged by Terry L.

Erwin, we decided to answer it, and at the same time to put into a more general

context the ozaenines inhabiting the southwestern United States.

In the course of our preliminary investigation, we found striking differences

among the genus-level taxa represented in the United States, and these discoveries

led us into an investigation of the Middle American ozaenine genera. Gradually, the

emphasis of the study changed from a species-level faunal study to a generic

treatment, with data about these species appended.

This paper is intended to achieve two goals: first, to establish the basis for a

phylogenetic analysis of the ozaenine genera of the world; and second, to

summarize the information that can be gleaned from knowledge of ozaenine species

at the northern limits of the Tribe in the New World. The only questions we attempt

to answer are two: first, proximity of relationships of ozaenines and the genus

Metrius Eschscholtz, made necessary by the potential out-group status of the

latter; and two, the original question about the identity of the specimens of Ozaena

that led to this study. Darlington's (1950) demonstration of close relationships

between ozaenines and paussines is so well established as to be unchallengeable.

The principal matter here is to establish at the generic or genus-group level the

propinquity of ancestry of the two tribe-level groups.

MATERIAL AND METHODS

Material

We studied about 400 specimens of Metriini and Ozaenini, and

representatives of several paussine genera. The material is in the collections noted

below. Each collection is designated in the text by a coden; these are listed below,

in alphabetical order, in association with the names and addresses of the

institutions represented. Names of curators are indicated in parentheses.

AMNH- Department of Entomology, American Museum of Natural History,

Central Park West at 79th Street, New York, New York, 10024, U.S.A.

(L. H. Herman).

ASUT- Department of Zoology, Life Sciences Center, Arizona State University,

Tempe, Arizona 85281, U.S.A. (F. Hasbrouck).

BMNH- Department of Entomology, British Museum (Natural History), Cromwell

Road, London, SW7 5BD, United Kingdom. (N.E. Stork).

CASC- Department of Entomology, California Academy of Sciences, Golden Gate

Park, San Francisco, California 94118, U.S.A. (D.H. Kavanaugh)..

CISC- California Insect Survey, Division of Entomology, University of

California, Berkeley, California, 94720, U.S.A. (J. Chemsak)..

CNCI- Canadian National Collection of Insects, Biosystematics Research

Centre, Agriculture Canada, Ottawa, Ontario, K1A 0C6. (J. M. Campbell).

CUIC- Department of Entomology, Comstock Hall, Cornell University, Ithaca,

New York, 14850, U.S.A. (J.K. Liebherr).

EGRC- Edward G. Riley Collection, Department of Entomology, Texas A & M

University, College Station, Texas 77843, U. S. A.

ETHZ- Entomologische Institut, Eidgenossische Technische Hochschule-

Zentrum, Universitatstrasse 2, CH-8006, Zurich, Switzerland (W.

Sauter).

FSCA- Florida State Collection of Arthropods, Division of Plant Industry, 1911

34th Street, S.W., P.O. Box 1269, Gainesville, Florida, 32602, U.S.A.

(R.E. Woodruff).

Quaest. Ent., 1990, 26(1)

36

Ball and McCleve

MCZC- Department of Entomology, Museum of Comparative Zoology, Harvard

University, Cambridge, Massachusetts, 02138, U.S.A. (S.R. Shaw; D.R.

Maddison).

MCZC Fall- H.C. Fall Collection, MCZC, address as above..

MNHP- Entomologie, Museum National d'Histoire Naturelle, Paris 75005,

France. (H. Perrin)..

OSUC- Ohio State University Collection of Insects and Spiders, 1735 Neil

Avenue, Columbus, Ohio, 43210, U.S.A. (C.A. Triplehom).

SMCC- Scott McCleve, 2210 13th Street, Douglas, Arizona, 85607, U.S.A.

TAIU- Department of Biology Collections, Texas A&I University, Kingsville,

Texas, 78363, U.S.A. (James A. Gillaspy).

TAMU- Department of Entomology, Texas A&M University, College Station,

Texas, 77843, U.S.A. (Horace R. Burke).

UASM- Strickland Entomological Museum, Department of Entomology,

University of Alberta, Edmonton, Alberta, Canada, T6G 2E3.

USNM- United States National Entomological Collection, Department of

Entomology, United States National Museum of Natural History,

Washington, D.C., 20560, U.S.A. (T.L. Erwin).

ZMHB- Museum fur Naturkunde der Humboldt Universitat zu Berlin, Bereich

Zoologisches Museum, Invalidenstrasse 43, DDR- 1040 Berlin (F. Hieke).

Methods.

Methods were standard, involving visual comparison of structural features,

using magnifying equipment from an 8X hand lens to stereobinocular microscopes

to a Cambridge Scanning Electron Microscope. Data were recorded in print, as line

drawings, and as photographs.

Measurements. — These were made with a Wild stereobinocular microscope

Model M-5, at 25X magnification. They were taken to determine size and

proportions.

Length is expressed as Standardized Body Length, determined as the sum of

length of: head (from mandibular condyle to posterior margin of compound eye);

pronotum (along mid-line); and elytra (along the suture, from apex of scutellum to

apex of elytra). Width is maximum transverse width of elytra. Measurements

presented in the species descriptions are of the smallest and largest males and

females, as determined by visual inspection of the material at hand. The values

obtained are thus rough approximations of total range in body size.

Ranking. — Used are subfamily, tribe, genus, subgenus and species. For the

higher ranks, we accepted those proposed by Banninger (1927), Kryzhanovskij

(1976), and Erwin (1979a). For genus and subgenus, we sought major gaps in

continuity of variation of structural features. The resulting taxonomic treatment is

thus conservative.

Relationships. — Statements about this topic are based on inferred relative

propinquity of descent, as determined by shared derived features. As out-group,

we used the monobasic tribe Metriini. Because of our deliberately limited data

base, we did not attempt a formal phylogenetic analysis of the taxa.

Species are regarded as evolving units reproductively isolated from other

such units, the gaps being judged by discontinuity in structural features.

Quae st. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

37

HISTORICAL ASPECTS

Tribal-level Synonymy

Tribe OZAENINI

"La cinquieme division" (in part) Latreille, 1817: 189.

Bipartis (in part) Latreille and Dejean, 1822: 79. — Latreille, 1829: 386.

Scaritides (in part) Dejean, 1825: 356. — Dejean and Boisduval, 1829: 230.

Brachinides (in part) Audouin and Brulle, 1834: 240. — Chenu, 1851: 87.

Ozenides Lacordaire, 1854: 154, 155. — Chaudoir, 1868: 43.

Ozaenidae Hope, 1838: 107.— LeConte, 1861: 5, 14.— Jeannel, 1941: 89.— 1946: 46 .—

Regenfuss, 1975: 283.

Ozaeninae Bates, 1881: 24. — Winkler, 1924: 83. — Crowson, 1955: 6. — Basilewsky, 1962:

291.— Nagel, 1979a: 9, 10, 11.— 1979b: 15.— Crowson, 1981: 502, 504.

Ozaenini Schaum, 1860: 773. — LeConte, 1861: 14. — Horn, 1881: 124, 128. — LeConte and

Horn, 1883: 23.— Sloane, 1920: 118.— Leng, 1920: 49.— Sloane, 1923: 242.— Csiki,

1927: 425. — Banninger, 1927: 177. — Andrewes, 1929: 162. — van Emden, 1942: 24. —

Blackwelder, 1944: 23.— Darlington, 1950: 49.— Ball, 1960: 94.— Bell, 1967: 105.—

Lindroth, 1969: XXII.— Hlavac, 1971: 57.— Kryzhanovskij, 1976: 82, 87.— Eisner et al.,

1977: 1385.— Reichardt, 1977: 377.— Ball, 1979: 91, 95, 100.— Ward, 1979: 185, 186,

188, 190.— Moore, 1979: 198.— Goulet, 1979: 205.— Thompson, 1979: 212, 226, 231,

232.— Erwin, 1979b: 481, 557, 583, 591.- Erwin and Sims, 1984: 374.— Ball, 1985: 24.—

Erwin, 1985: 451, 467.— Stork, 1985: 1113.

Ozaenina Iakobson, 1906: 316. — Bousquet, 1986: 378.

Mystropomini Horn, 1881: 116-117.— Sloane, 1923: 246.

Mystropominae Dupuis, 1911: 2.

Mystropomitae Jeannel, 1946: 47.

Paussidae (in part) Crowson, 1955: 6. — Deuve, 1988: 176.

Paussitae (in part) Erwin and Sims, 1984: 374. — Erwin, 1985: 467.

Paussinae (in part) Moore, in Moore et al., 1987: 26.

Classification and Relationships

Tribal level. — Latreille (1817: 189) included Ozaena in his fifth division of

the Carabiques, which included also "Les Morions", "Les Aristes", "Les

Harpales", "Les Feroniens"," Les Licinines", "Les Badistes", and "Les Panagees".

Latreille and Dejean (1822), Dejean (1825), and Dejean and Boisduval (1829)

included Ozaena in the Bipartis, along with the scaritines and genus Morion

Latreille. Dejean (1825: 355) stated implicitly, however, that Ozaena was an

aberrant element, and later authors (Audouin and Brulle, 1834, and Chenu, 1851)

included the ozaenines with the brachinines and various lebiomorphs. In part, this

association was based on the crepitating behavior of adults ozaenines and

brachinines, though the diagnostic feature given was habital — adults with rather

thick bodies.

Hope (1838) first recognized ozaenines as a distinct assemblage, though he

included with them Nomius Castelnau, Melisodera Westwood, and Catapiesis

Brulle. Hope neglected to give reasons for erecting the Ozaenidae. Lacordaire

(1854) followed suit, including in the group eight ozaenine genera and Nomius,

characterized in part on a reduced mesostemum so that the mid-coxae are in contact

with one another. He noted also the posterolateral elytral flanges, and that they did

not occur in Nomius.

Schaum (1860) did not treat the Ozaenini in detail, though he (/.c., p. 773)

located the group (as defined by Lacordaire) in the carabine assemblage ( i.e ., adults

with mid-coxal cavities open) and in a sub-group including Omophronini,

Quae st. Ent., 1990, 26(1)

38

Ball and McCleve

Elaphrini, Carabini, Loricerini, Promecognathini, and Mormolycini. He did not

place them in the sub-group that followed that included the Scaritini, Siagonini, and

Hiletini, nor with the brachinines and other truncatipennian groups. This was

indeed a radical departure in classification at that time, and quite appropriate.

Chaudoir (1868) provided a synopsis of the Ozaenini, bringing together

information about all of the taxa previously described, and describing new genera

and species. For the group diagnosis, he gave special emphasis to the reduced

mesostemum that Lacordaire had recorded, and noted as well, following Schaum

(1860) that the mesothoracic suture extended to the mid-coxae. He stated also his

belief that the ozaenines should be placed between the brachinines and helluonines,

where they had been placed by previous authors. Perhaps Chaudoir's most

important contribution in this work was to identify the distinctiveness of Ozaena

dentipes Olivier (type species of Ozaena ), separating it from the other species that

had been described in the genus, and transferring the latter to other taxa —

principally to P achy teles Perty.

Bates (1881: 24) placed the ozaenines about as Schaum had done, between

loricerines and scaritines, in their own subfamily. In the same year, Horn arrayed the

ozaenines in two tribes: Mystropomini and Ozaenini. The basis for this division

was an error: Horn, while correctly recording that the middle coxal cavities were

open in Mystropomus , . mistakenly stated that they were closed in the remaining

genera of ozaenines. He placed the Ozaenini (minus Mystropomus ) in the subfamily

Harpalinae, near the Panagaeini. Mystropomus was left in the Carabinae, or first

major division of the Carabidae. Although Bates ( l.c .) had pointed out the error

(Horn sent him his MS before it was published), he did so in such a gentlemanly

manner that the point seemed to have been lost. Thus, LeConte and Horn (1883) and

Leng (1920) followed Horn's arrangement. Dupuis (1911) also recognized the

mystropomines as a group separate from the Ozaenini, at least implicitly. Because

he treated the Metriinae in the same publication, evidently he accepted Horn's

opinion about a close relationship between the latter and the mystropomines.

Sloane (1923: 246) noted Horn's mistaken interpretation of the thoracic

structure of the Ozaenini ( sensu Horn), and re-combined the latter with

Mystropomus. He placed the re-constituted Ozaenini in his "Carabidae Clausae",

along with Metriini, Migadopini, Scaritini, Siagonini, Enceladini,

Promecognathini, Elaphrini, Loricerini, and Omophronini. Csiki (1927) and

Andrewes (1929) followed Sloane's sequencing.

Banninger (1927) revised the Ozaenini, providing a much more comprehensive

treatment than Chaudoir's. He gave a clear tribal diagnosis and detailed

description, based on adult external features of the known genera. He confirmed

that the affinities of the Ozaenini were with the carabines (broad sense), and

particularly with the Cicindisini, Nototylini, and Metriini. Among diagnostic

features for the Ozaenini that he emphasized were the elytral flanges.

Kolbe (1927) hypothesized that ozaenines and paussids were closely related.

Van Emden (1942) characterized the larvae of the Ozaenini, emphasizing the

unusual modifications of the urogomphi and posterior abdominal segments.

Jeannel (1941) included in his new taxon Isochaeta (based on the apical

position of both fore tibial spurs): trachypachines, gehringiines, metriines,

ozaenines, and paussines. However, he did not pursue the matter of relationships of

any of these groups to one another. Nonetheless, one can see from the sequence of

taxa that probably he considered the last three to be related to one another.

Darlington (1950: 48) re-asserted the basis for hypothesizing close

relationship between paussids and ozaenines, noting that Kolbe was the first so to

insist. Nonetheless, he retained the ozaenines as a separate tribe, and in effect

Quaest. Ent ., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

39

treated the paussids as an informal taxon to include the Protopaussini and Paussini.

Darlington did not comment about classification of the ozaenine genera, but he

developed a detailed classification for the paussine genera.

Basilewsky (1962: 291) transferred the Protopaussini from the subfamily

Paussinae to the Ozaeninae, on the basis of the plesiotypic unreduced antennal

pedicel. He did not take into account the apotypic features shared by

protopaussines and paussines.

Transformation series in a variety of character systems support adequately the

hypothesis of monophyly of paussines and ozaenines. Crowson (1955) accepted

this, but rather than treating this complex as a group within the family Carabidae, he

chose to recognize the paussine complex as a separate family, including therein the

Ozaenini.

While there is little doubt that the ozaenine-paussine complex is

monophyletic, there is some doubt that the Ozaenini are monophyletic relative to

the paussines, i.e., some extant ozaenine genus might be the sister group of the

Protopaussini + Paussini. So, the more distal part of the reconstructed phylogeny

of this complex is not resolved at the tribal level. What about the more basal part: is

there an extant sister group for the ozaenine-paussine complex?

Luna de Carvalho (1959) pointed out the marked similarity between the

genitalia of metriines and ozaenine-paussine males. Bell (1967) united in his new

taxon Septisternia the tribes Metriini, Ozaenini and Paussini (implicitly

Protopaussini + Paussini), implying thereby that Metrius was the sister group of

the ozaenine-paussine complex. This hypothesis was accepted by Regenfuss

(1975) and by various subsequent authors.

On the basis of superficial examination of larval features, Goulet (1979: 205)

suggested that "metriines are simply ozaenines", and this opinion of striking

similarity was borne out by Bousquet's (1986) detailed study of the larval

characteristics of Metrius.

Similarity between adults of Metrius and those of the ozaenine-paussine

complex in complex defensive secretions and their delivery systems provide more

evidence of close relationships of these taxa (Eisner and Aneshansley, 1981).

Thus, metriines and the ozaenine-paussine complex can be treated as a

monophyletic assemblage. Without going into detail here, we assert that the

Metriini is the sister group of the hypothetical ancestor of the ozaenine-paussine

complex.

Erwin and Sims (1984) and Erwin (1985) hypothesized a close relationship

among the supertribes Metriitae (including Metriini, only), Paussitae (including

Ozaenini + Paussini and four other tribes) and Brachinitae. These supertribes were

included in the subfamily Paussinae. As indicated above, metriites belong with the

paussite assemblage. Thus only two main lineages are represented in the Paussinae:

Paussitae and Brachinitae. Are these two really sister groups? In addition to Erwin

and Sims (/.c.), Eisner et al. (1977), Bousquet (1986) and Deuve (1988) have stated

so. Others (Forsyth, 1972; Crowson, 1981, p. 503, Bell, 1983, p. 595, and Moore

et al., 1987, p. 26 and 48) indicate either implicitly or explicitly their belief that the

similarity between brachinites and paussites, in complex defensive secretions and

complex delivery systems — the best evidence available for close relationship —

are the result of convergence. We favor the latter opinion.

Similarities between brachinites and Psydriformes ( sensu Erwin, 1985) are in

apotypic features of several systems that seem to be functionally independent

(thoracic structure, organization of the antennal cleaner of the fore tibia, structure of

the male genitalia and ovipositor). It seems to us that the best explanation for this

array of similarities is inheritance from a common ancestry. Thus, we hypothesize

Quae st. Ent., 1990, 26(1)

40

Ball and McCleve

that the brachinites are either members of the Psydriformes, or at least are the sister

group of this group of carabids that, overall, is more highly derived than is the

ozaenine-paussine complex.

Incredible as it seems, then, the elaborate defensive system of brachinites and

paussites must have evolved independently if the similarities between brachinites

and psydriforms are indicative of close relationship — as we hypothesize. Thus,

there is a clear conflict of evidence. This conflict must be resolved, in terms of

additional evidence yet to be found that will tip the balance one way or the other.

Forbes (1926: 59) in his monumental publication about wing folding, pointed out

the similarity between Brachinus and Passus adults in their distinctive folding

pattern of the hind wings. He believed that this similarity did, in fact, indicate

relationship (personal communication). It remains to be determined, however, if

other brachinines and paussines have this same form of wing folding, and it remains

to be determined, as well, if the pattern is plesiotypic or apotypic in the Carabidae.

This lead is worth pursuing.

The question of ranking of the metriine + ozaenine + paussine complex

remains to be answered. Erwin and Sims ( l.c .) and Erwin (l.c.) recognize two

supertribes, as noted above. Bousquet (1986: 378) proposes recognition of one

tribe, the Paussini, to include two subtribes — Ozaenina and Paussina (Metriini +

Protopaussini + Paussini of authors). We favor treating this complex as a

subfamily, with three tribes, only: Metriini, Ozaenini, and Paussini. (We exclude

thereby Nototylini, and Cicindisini that were included by previous authors). If we

did not use the rank of subfamily in the Carabidae, we would follow Bousquet in

his system of ranking.

The history of classification of the Ozaenini has been one of surprising

discoveries and recurrent themes. The most surprising discovery is that of close

relationship between metriines and ozaenines+paussines. It was presaged by Horn

(1881: 117), when he pointed out the similarities in structural features between

Mystropomus . and Metrius. However, this lead was not followed for many years.

The recurrent theme is the relationship between ozaenines and scaritines on the one

hand, and between paussines (present sense) and brachinines, on the other. The

linear arrangement by Erwin (1985: 467) is very similar in part to that presented in

the past, with ozaenines either in (Dejean, 1825) or near (Schaum, 1860) a complex

including the scaritines, and either in (Audouin and Brulle, 1834) or near (Chaudoir,

1868) the brachinines. We conclude that our predecessors of the last century did

rather well, though they had simpler equipment and lacked the knowledge of carabid

diversity that we have now.

Generic level. — The only explicit attempt to classify the genera of Ozaenini

was by Jeannel (1946: 46-48). Treating this assemblage as a family, he recognized

three subfamilies: the monogeneric Australian Mystropomitae and Neotropica/

Physeitae, and the Neotropical-Afrotropical-Oriental Ozaenitae to include all of the

remaining genera. For the Ozaenitae, he recognized three tribes: the Oriental

Eustrini, including Eustra Schmidt-Goebel and Dhanya Andrewes; the monogeneric

Neotropical Pachytelini for the genus Pachyteles Perty; and the Neotropical-

Afrotropical-Oriental Ozaenini for the remaining genera.

For the Afrotropical-Madagascan fauna, Jeannel erected the subgenus

Afrozaena, and included it, Sphaerostylus Chaudoir, and Pseudozaena Castelnau

as subgenera of the genus Pseudozaena. Basilewsky (1962: 291-293) accepted

implicitly Jeannel's classification, but added the Protopaussini to the Ozaenidae

(treated by Basilewsky as a subfamily). However, he disagreed with Jeannel's

treatment of Pseudozaena, ranking each of the three subgenera as a genus, and re-

Quaest. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

41

ranking Itamus Schmidt-Goebel as a genus separate from Pseudozaena ( sensu

stricto ).

For the classification of genera, we agree that Mystropomus should be placed

in a higher taxon of its own (as a subtribe), at the base of the tribe. The evidence

that we have seen suggests that Ozaena and its close relative, Platycerozaena

Banninger, are not closely related to any of the other ozaenine genera, and we think

that Pachyteles may be a plesiotypic sister group to at least some of the Old

World genera. Physea and its close relative Physeomorpha Ogueta,are abundantly

distinctive in features of adults, some of which indicate marked primitiveness.

Overall though, we think that this group is not far removed from the pachyteline

assemblage.

In the light of these observations, we cannot offer much support for Jeannel's

classification of the ozaenine genera, though we have nothing to put in its place.

See also Stork (1985: 1120). We think it best to avoid a formal classification of

these genera at this time, though we realize that preparation of such a classification

must be a high priority for future workers on this tribe.

COMPARATIVE MORPHOLOGY

Structural and Biochemical Features

In this section, we describe and compare in an evolutionary context,

microsculpture, various setal patterns, structures, and defensive secretions useful in

characterizing the genera of ozaenine carabids.The major purpose of this section is

to explain the details of character complexes which either have not been used

extensively in classifying ozaenines, or have been used only superficially. As a

working hypothesis, we accept the tribe Metriini (genus Metrius Eschscholtz) as

out-group for the tribes Ozaenini and Paussini {sensu lato, including Protopaussini

and Paussini sensu Darlington, 1950, or Paussidae Jeannel, 1946), which we

believe are sister groups.

Microsculpture. — Adults of Metrius exhibit an isodiametric mesh pattern

over almost the entire body surface. For the ozaenines, dorsal sculpture is

isodiametric, with microlines either distinct or reduced and hard to see, or lost

(Figs. 107A-C). For the ventral surface, mesh pattern varies from isodiametric to

transverse, with most sclerites exhibiting a transverse pattern. No marked or

taxonomically very useful trends were identified.

Setae on dorsal surfaces of head and pronotum. — Metrius adults have a

generalized pattern of setal number: clypeus and vertex, one pair each; pronotum,

two pairs of marginal setae. Ozaenines are more setose generally, or have fewer

setae. Entomoantyx .i.Entomoantyx, new genus; adults have a pair of clypeal

setae, a pair of supraorbitals, and several in a transverse row across the vertex. The

lateral margins of the pronotum have three to five pairs of setae. Physea .i. Physea

Brulle; and Pachyteles .i.Pachyteles Perty;adults have about 12 clypeal setae, one

pair of supraorbitals, and several pairs of setae posterad the compound eyes. The

lateral pronotal setae are several pairs, as in Entomoantyx.

In contrast, adults of Ozaena and Platycerozaena lack clypeal, supraorbital and

lateral pronotal setae.

The transformation series would seem to be:

Quae st. Ent., 1990, 26(1)

42

Ball and McCleve

Figs. 9 — 16. 9 — 15, Left antennomeres, 7 — 11, of: 9, Metrius contractus Esch.; 10, Physea hirta

LeC.; 11, Entomoantyx cyanipennis (Chd.); 12, Pachyteles nr . striola Perty; 13, Pachyteles

kuntzeni (Bann.); 14, Ozaena lemoulti Bann.; 15, Platycerozaena panamensis (Bates). 16, Left

antennomere 11, sense organs, of Metrius contractus Esch. Scale bars = 200 Jim, Figs. 9-15; 20

|im, Fig. 16. Legend: sb- sensillum basiconicunr, sc- sensillum coeloconicum\ and st- sensillum

trichodeum.

Quae st. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

43

Figs. 17 — 23. Left antennomere 11, of: 17, Metrius contractus Esch.; 18, Physea hirta LeC.; 19,

Entomoantyx cyanipennis (Chd.); 20, Pachyteles nr. striola Perty; 21, Pachyteles kuntzeni

(Bann.); 22, Ozaena lemoulti Bann.; 23, Platycerozaena panamensis (Bates). Scale bars = 200 fim,

Chinect Fnt IQQH

44

Ball and McCleve

Figs. 24 — 30. Left antennomere 11, apical aspect, of: 24, Metrius contractus Esch.; 25, Physea

hirta LeC.; 26, Entomoantyx cyanipennis (Chd.); 27, P achy teles nr. striola Perty; 28,

Pachyteles kuntzeni (Bann.); 29, Ozaena lemoulti Bann. . — A, lower mag.; B, higher mag.,

showing sense organs; 30, Platycerozaena panamensis (Bates). Scale bars = 100 pm, Figs. 24 —

29A, and 30; 50pm, Fig. 29B. Legend: sb- sens ilium hasiconicum; st- sensillum trichodeum.

Middle American Genera of the Tribe Ozaenini

45

Antennae. — Variation occurs in overall length, form of antennomeres, and

distribution and frequency of types of sensilla. Overall length can be appreciated

from examination of Figs. 1 to 8. These and Figs. 9 to 30 illustrate also differences

in form and proportions of antennomeres. Note particularly the moniliform articles

of Ozaena adults (Figs. 14 and 22), quadrate form of antennomeres 4-10

characteristic of Pachyteles (Figs. 12, 13, 20, and 21), and transverse antennomeres

4-10 characteristic of Platycer ozaena (Fig. 15).

Antennomere 1 1 among Ozaenini differs from the more basal antennomeres in

being longer (Figs. 10 and 11), and either as broad as (Figs. 12 and 13) or broader

than (Figs. 14 and 22) the latter. Of special note are: slender curved antennomere 11

of Physea (Fig. 18), the basally constricted antennomere 11 of Ozaena (Fig. 22)

and the very broad antennomere 1 1 of Platycer ozaena (Figs. 15 and 23). Compared

to Metrius (Fig. 17), antennomere 11 is narrow in Physea and Entomoantyx,

slightly wider in Pachyteles (Fig. 20), and markedly wider in Ozaena and

Platycerozaena.

Antennomere 11 exhibits differences in cross section also, which are best

appreciated from an apical aspect (Figs. 24-30). For Metrius (Fig. 24),

antennomere 1 1 is terete, without a sharply delimited apical ridge. For Physea and

Entomoantyx (Figs. 25 and 26), antennomere 1 1 is nearly circular, and without a

sharply differentiated apical ridge. For Pachyteles (Figs. 27 and 28) antennomere

1 1 is terete, but with a moderately sharply defined apical ridge. For Ozaena and

Platycerozaena (Figs. 29A and 30), antennomere 11 is terete, with apical ridge

sharply delimited as a carina: straight in Ozaena Fig. 29A), and sinuate in

Platycerozaena (Fig. 30). The orientation of this carina is dorso-ventral.

Sensillar types of ozaenine antennomeres were identified through the

publication of Zacharuck (1985: 25-26). Three types were recognized: trichodea

(relatively long setiform hairs, st in Fig. 16); basiconica (relatively short and thick

setiform hairs, sb in Fig. 16); and coeloconica (cones set in the floor of shallow

depressions or pits in the cuticle, open to the outside through a small round hole, sc

in Fig. 16 ). Only sensilla trichodea and basiconica are considered here.

Antennomeres 1 (scape) to 4 have few sensilla trichodea, and the intervening

surfaces are relatively smooth. These antennomeres are not considered further. For

antennomeres 5-11, the surfaces bearing sensilla are more or less shagreened, the

roughened appearance being the result of the raised cuticular rims around the bases

of the sensilla.

Four types of distribution and frequency of sensilla are recognized.

Type a: sensilla trichodea numerous on antennomeres 5-11, each article with

anterior and posterior faces with reduced setation at the middle, and in the triangle

at the base of antennomere (Figs.9, 10, 12 and 13); sensilla basiconica numerous

and rather long, ca. one-fifth length of sensilla trichodea, and confined to dorsal and

ventral areas of antennomeres ( cf Fig. 16). Taxa whose adults exhibit type a

antennomeres are: Metrius, Physea, and Pachyteles.

Type b: for antennomeres 5-10, as in type a, above; for antennomere 11,

toward apex, sensilla basiconica few and ordered in more or less parallel rows (Fig.

26); exhibited by Entomoantyx adults.

Type c (Fig. 29A-B): for antennomeres 5-10, sensilla trichodea and

basiconica relatively few, basiconica very short, glabrous or nearly glabrous areas

extensive; ventrally, or ventro-laterally, with small groups of sensilla basiconica;

for antennomere 11, sensilla trichodea sparse dorso- ventrally, preapically, and a

row each side of apical carina; about one-third of apex occupied by a dense

concentration of sensilla basiconica. Type c is exhibited by Ozaena adults.

Quaest. Ent., 1990, 26(1)

46

Ball and McCleve

Type d (Fig. 23): for antennomeres 5-10, sensilla trichodea sparse and only

about half length of those of types a-c; sensilla basiconica ventrally in dense

groups, each side of a carina; antennomere 1 1 with relatively sparse, short sensilla

trichodea apically, dorsally and ventrally, around triangular central area; sensilla

basiconica very dense apically and along dorsal and ventral surfaces. Adults of

Platycerozaena exhibit this sensillar type.

Based on extent of departure from the antennae of Metrius, these data

suggest the following linear transformation series:

Metrius - >Physea - >P achy teles - >Ozaena

Entomoantyx Platycerozaena

Labrum. — This sclerite varies in form and setation. In form, it is transverse,

i.e., wider than long, with the anterior margin truncate or nearly so (Figs. 31-33, 35

and 36), or broadly concave (Fig. 34), or elongate, with anterior margin notched

(Fig. 37). Number of long preapical setae varies from 0 ( Platycerozaena , Fig. 37)

to 12 to 16 (. Metrius , Fig. 31), with 7 to 12 intermediate ( Physea , Fig. 32;

Entomoantyx , Fig. 33; Pachyteles, Fig. 34; and Ozaena , Fig. 36). The highest

number group is postulated as plesiotypic, since it is characteristic of the

outgroup. Short setae (sensilla basiconica) are also evident on the dorsal surfaces

of the labra of Ozaena and Platycerozaena. The data suggest the following

branched transformation series:

Physea

Entomoantyx

T

Metrius- >Pachyteles >Ozaena ~>Platyceroz.

Mandibles. — These are illustrated in Figs. 38 to 43, and characterized for

each genus in Table 1. Mandibles of Carabidae have been described in several

publications (cf. Shpeley and Ball, 1978; Forsythe, 1982; and Evans and Forsythe,

1985), and a detailed and more or less consistent nomenclature developed. The

anterior incisor area, of varied width, terminates in the apical incisor tooth. The

occlusal margin of the incisor is the terebral ridge or margin (tm, Figs. 38A-B). A

terebral tooth (tt) is near the posterior part of the terebral margin. Below the tooth

is the retinaculum (r, Fig. 38C), more or less prominent, with an anterior (art) and

posterior (prt) retinacular tooth, joined by a retinacular ridge (Figs. 42A-B, 43E-

F). The retinaculum is terminated by a groove, the premolar incision. Posterior to

this incision is the molar area, divided or not by a molar incision: if divided, the

anterior part of the molar area is the premolar (Figs. 38A-B, pm), the posterior part,

the molar (Figs. 38A-B, m). Ventrally, various ridges occur in different taxa. In

metriines and ozaenines, a ventral retinacular ridge (Fig. 39C, vrr) and a premolar

ridge (Fig. 38C, pr) are recognized. A ventral groove (Fig. 38D, vg) of varied

length bears a row of setae, the latter of varied length and density. The thicker

lateral surface of each mandible contains a triangular depression, the scrobe. In

Metrius , the scrobe bears a single long seta (Fig. 38A, ss); in ozaenines, it bears a

single long seta (Fig. 40A) or a varied number of shorter setae (Figs. 39A and

41 A). The dorsal surface is variously sparsely covered with setae shorter than

those in the scrobe.

Quaest. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

47

Figs. 31 — 37. Labrum, dorsal aspect, of: 31, Metrius contractus Esch.; 32, Physea hirta LeC.; 33,

Entomoantyx cyanipennis (Chd.); 34, Pachyteles nr. striola Perty; 35, Pachyteles kuntzeni

(Bann.); 36, Ozaena lemoulti Bann.; 37, Platycerozaena panamensis (Bates). Scale bars = 200 |im.

48

Ball and McCleve

In all their features, the mandibles of Metrius adults are much like those of the

still more primitive genus Trachypachus. Thus, we accept these features as

plesiotypic. Mandibles of ozaenines are apotypic in the following features:

terebral ridge short (not extended on to retinaculum); retinaculum reduced; and molar

area not divided by an incision. The mandibles characteristic of all the genera

examined, except Platycerozaena , exhibit additionally the following apotypic

features: posterior retinacular tooth moderate to large, extended posteriorly; and

ventral groove long. Considering the generally derived features of Platycerozaena

and its close relationship to Ozaena, we believe that the seemingly plesiotypic

conditions of the former genus (small posterior retinacular tooth and short ventral

groove) are secondary losses, and are thus apotypic.

Within the Central American Ozaenini, we regard as most plesiotypic the

mandibles of Entomoantyx : scrobe with single long seta, surface otherwise

glabrous; and anterior retinacular tooth of the left mandible prominent. Most

derived are the mandibles of Physea : falcate in form, and with the retinacular ridge

reduced by loss of the anterior retinacular tooth. The mandibles of Platycerozaena

also seem markedly apotypic, with broad, partially punctate dorsal surfaces, and

somewhat reduced system of teeth and short ventral grooves. The mandibles of

Ozaena , much like those of Pachyteles, are more derived in that the dorsal surfaces

are punctate, as in Platycerozaena. We believe that the data presented are best

summarized by the following branched transformation series:

Physea

' T

Metrius - >Entomo. - >Pachytel. - > Ozaena - >Platycer.

Nothing is known about feeding habits of adult Metrius and ozaenines,

though the structural features of the mandibles and maxillae (see below) suggest that

these beetles are "mixed feeders, ingesting food as fluid, semi-fluid, mush and

fragments, or a mixture of all of them" (Evans and Forsythe, 1985: 122).

Maxillae. — - Among ozaenines, these structures exhibit limited variation

except in form of terminal palpomeres. For the Central American ozaenines,

however, the more interesting variation is in details of the lacinia. Most taxa have

a moderately long, sharp, slightly curved apical tooth and a moderately dense brush

of marginal setae. Characteristic of Physea (Fig. 45) is a lacinia like that of

Metrius (Fig. 44), with a long markedly curved apical tooth. The lacinia

characteristic of Ozaena (Figs. 47 and 48) has a short, chisel-like, apical tooth, and

a very dense brush of setae. Laciniae with fewer setae and longer teeth are more

likely to be used as rakes to draw food particles into the mouth, whereas laciniae

with more setae are more likely to be involved in a system of pre-oral digestion,

with the dense setae serving to hold the digestive fluid used to liquefy partially the

prey before ingesting it (Evans and Forsythe, 1985: 123). Certainly the most

derived maxillae are those of Ozaena. The data seem to indicate the following linear

transformation series:

Entomoantyx

Metrius - > Physea - >P achy teles - >Ozaena

Platycerozaena

Middle American Genera of the Tribe Ozaenini

49

Figs. 38 — 41. Mandibles, A, C, left mandible, dorsal and ventral aspects, respectively; B, D, right

mandible, dorsal and ventral aspects, respectively: 38, Metrius contractus Esch; 39, Physea hirta

LeC.; 40, Entomoantyx cyanipennis (Chd.); 41, Pachyteles nr. striola Perty. Scale bars = 200

pm, Legend: art- anterior retinacular tooth; m- molar; pm- premolar; pr- premolar ridge; prt-

posterior retinacular tooth; r- retinaculum; tt- terebral tooth; vg- ventral groove; vrr- ventral

retinacular ridge.

[Quae st. Ent ., 1990, 26( 1 )

Figs. 42—48. 42 — 43, mandibles, A, C, E, left mandible, dorsal, ventral, and occlusal aspects,

respectively; B, D, F, right mandible, dorsal, ventral, and occlusal aspects, respectively. 42,

Ozaena lemoulti Bann.; 43, Platycerozaena panamensis (Bates). 44, Left maxilla, ventral aspect, of

Metrius contractus Esch. 45, Right maxilla, ventral aspect (reverse printing, for ease of

comparison), of Physea hirta LeC. 46, Left maxilla, ventral aspect, of Pachyteles kuntzeni

(Bann.). 47, Left maxilla of Ozaena lemoulti Bann.: A, complete structure; B, lacinia, apex. 48,

Left maxilla of Platycerozaena panamensis (Bates). Scale bars = 400 pm, Figs. 42-47A, and 48;

100 pm, 47B. Legend: art- anterior retinacular tooth; m- molar; pm- premolar; pr- posterior

retinacular ridge; prt- posterior retinacular tooth; r- retinaculum; tt- terebral tooth; vg- ventral

groove.

Quaes t. Ent., 1990, 26(1)

TABLE 1. Characteristics of Mandibles of Metrius Eschscholtz, and of Ozaenine Genera of North and Middle America

Middle American Genera of the Tribe Ozaenini

51

— «

2 >

— o

c O

u ^

> O

•o

<u

-o

C T3

4>

T3

O

e

•o

U

"O

c ~o

•o

4>

TJ

c -o

T3

U

T5

• C/3

u IS

at t-

T3

<U

<n

E

C c

T3 OS

S 2

>> T3 0£

>> -o OS

S 5 2

£ K C

C/3 <l> O

Urn

o

9i

< w

z o

0/

<D

<D

■o

o

E

H

c

<3

0

5

o

s:

Quae st. Ent., 1990, 26(1)

Platycero- many + wide moderate slightly curved thin moderate small not short

zaena setae not extended divided

on to Retinacul.

52

Ball and McCleve

We believe that this structural transformation series parallels a functional

transformation series, involving a shift from ingestion of particulate to ingestion

of more liquid food.

Labium. — Features of note involve the mentum, and terminal labial palpomeres

(Figs. 49-55). On the menta of Metrius, Entomoantyx, and P. ( Goniotropis ) (Figs.

49, 51 and 53), one or two pairs of long tactile setae are located paramedially

posterad the tooth. Such setae are absent from the menta of Physea, Ozaena, and

Platycerozaena (Figs. 50, 54 and 55).

Lateral lobes of the mentum are moderately long in Metrius , Physea ,

Entomoantyx , and Pachyteles, and markedly long in Ozaena and Platycerozaena. In

addition, in the last-named genus, the lateral lobes are markedly narrowed and

pointed apically.

Epilobes (Fig. 49, el) are broader medially in Metrius, Physea, and

Pachyteles. They are broader apically in Entomoantyx, and in Ozaena and

Platycerozaena they are narrow.

The mental tooth is moderately long in Metrius, Entomoantyx, Physea, and

Pachyteles', somewhat reduced in Ozaena, and markedly reduced in Platycerozaena.

The apex of the tooth is notched in Metrius contractus Esch., and not notched in an

undescribed species of Metrius (Y. Bousquet, personal communication) or in the

New World ozaenine genera — though it is notched in the Australian genus

Mystropomus. .

The glossal sclerites (Fig. 49, gs) are bisetose in members of Metrius,

Entomoantyx, Physea, and Pachyteles. These setae are lacking from the glossal

sclerites of Ozaena and Platycerozaena, evidently a reflection of the general

reduction in setae characteristic of these genera.

Terminal palpomeres vary somewhat within the New World ozaenine genera.

In most, they are like those of Pachyteles (Fig. 52), thus broader than those of

Metrius (Fig. 49). The terminal palpomeres of Physea, however, are parallel-sided

and more elongate (Fig. 50).

We believe that the following branched transformation series summarizes

adequately the pattern of the various forms and details of the labium:

Entomoantyx

T

Metrius* - Pachyteles - >Ozaena - > Platycerozaena

_ Physea _

Thoracic structures. — Metrius adults exhibit an hypertrophied intercoxal

process of the prosternum, which is so large that it covers the mesostemum. The

mesosternal and metasternal intercoxal processes are firmly articulated to one

another, as in most carabids. Ozaenines seem to be rather more loosely articulated

than are adults of most carabid groups. The intercoxal process of the prostemum is

of normal size. The intercoxal processes of the meso- and metastema are rather

loosely articulated in adults of Entomoantyx, P . ( Goniotropis ), Ozaena, and

Platycerozaena. Among adults of Physea and P. {Pachyteles), the processes are

reduced so that the middle coxae are in contact with one another medially.

The following linear transformation series summarizes this system:

Quaest. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

53

Stork (1985: 1115 and Fig. 12) reports in adults of Dhanya prothoracic pits,

one on each side, laterad the fore coxal cavities. He notes also the presence of such

structures in adults of Mystropomus, but that they are positioned differently, and

thus probably not homologues of the pits in Dhanya. He notes a possible

involvement of these pits in a possible association with ants.

The metathorax of Metrius adults is small, the result of loss of the flight

function and ultimately of the metathoracic wings and associated muscles. This

group of losses and reductions is apotypic for Metrius.

Metathoracic wings. — The Middle American genera are characterized by

wings with the wedge cell lacking, but otherwise with complete venation, and with

the oblongum cell large, and quadrangular (c/. Ward, 1979: 183, Figs. 1-3 and 7). In

contrast, members of the Oriental genus Dhanya exhibit markedly reduced venation,

including a triangular, stalked oblongum cell (Stork, 1985: 1129, Fig. 43). In this

latter feature, members of this genus resemble those of the Paussina (Darlington,

1950).

Legs. — Features of particular note include projections of the fore femora,

antennal cleaner of the fore tibiae, and sexual dimorphism of the fore tarsomeres.

The fore femora without projections are those of adults of Metrius (Fig. 56),

Entomoantyx (Fig. 58 ), and some South American species of Pachyteles (subgenus

Tropopsis ). Adults of Physea (Fig. 57) and of Platycerozaena (Fig. 62) exhibit a

prominent broad swelling ventrally. Adults of Pachyteles (Figs. 59 and 60) have

prominent narrow projections, whereas those of Ozaena (Figs. 61A-B) have

shorter, setose projections.

The transformation series indicated by these data is the following:

Antennal cleaner. — This complex comb organ (Darlington, 1950: 60), includes

some form of groove or notch on the front tibia, and associated setae and spines,

projections, and/or spurs. Hlavac (1971) provides a detailed analysis for the

Carabidae. Figures 63 to 69 illustrate antennal cleaners for Metrius and the

ozaenines. Table 2 provides details for the taxa of interest.

Structures are as follows: a more or less extensive groove in the mesal surface

of the fore tibia, the cleaning channel (ch); a more or less expanded portion of the

tibia adjacent to the channel (mex); a setal band (sb) extended across the channel

and parallel to the posterior edge of the channel; several large sinuate clip setae

(els), origin posteriorly at the medial expansion; and anterior row of setae (asr)

along the anterior edge of the channel; a zone of confluence distally, where the setal

band and anterior setal row almost meet, and in some taxa seem to disappear in a

dense patch of other setae (Fig. 69B).

Quae st. Ent.. 1990, 26(1)

54

Ball and McCleve

Figs. 49 — 57. 49 — 55, Labium, ventral aspect, of: 49, Metrius contractus Esch; 50, Physea hirta

LeC.; 51, Entomoantyx cyanipennis (Chd.); 52, Pachyteles nr. striola Perty; 53, Pachyteles

parca LeC.; 54, Ozaena lemoulti Bann.; 55, Platycerozaena panamensis (Bates). 56-57, left fore

femur, anterior aspect, of: 56, Metrius contractus Esch; 57, Physea hirta LeC. Scale bars = 200

pm. Legend: el- epilobe of mentum; gs- glossal sclerite of mentum.

Quae st. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

55

Figs. 58 — 64. 58 — 62, Left fore femur, 58, and 60-62, anterior aspect, 59, posterior aspect, of:

58, Entomoantyx cyanipennis (Chd.); 59, Pachyteles nr. striola Perty; 60, Pachyteles kuntzeni

(Bann.); 61, Ozaena lemoulti Bann., A- entire sclerite, B- spine; 62, Platycerozaena panamensis

(Bates). 63-64, Left fore tibia. A- anterior aspect, B- inner aspect, of: 63. Metrius contractus

Esch: 64, Physea hirta LeC. Scale bars = 200 |im. Legend: asp- anterior tibial spur; asr- anterior

setal row; ch- channel; els- clip setae; mex- median expansion; psp- posterior tibial spur; sb- setal

band.

Quae st. Ent ., 1990, 26(1)

56

Ball and McCleve

Figs. 65 — 69. 65 — 66A-B, and 67 — 69, front tibiae, 65 — 66 and 68-69, left; 67, right (printed in

reverse, for ease of comparison); A, anterior aspect, B, inner aspect, of: 65, Entomoantyx

cyanipennis (Chd.); 66, Pachyteles nr. striola Perty; 67, Pachyteles kuntzeni (Bann.); 68,

Ozaena lemoulti Bann.; 69, Platycerozaena panamensis (Bates). 66C, Left fore tarsomeres 1-3,

ventral aspect of male Pachyteles nr. striola Perty. Scale bars = 200 |lm, Legend: asp- anterior

tibial spur; asr- anterior setal row; ch- channel; els- clip setae; mex- median expansion; psp-

posterior tibial spur; sb- setal band.

Quaest. Ent., 1990, 26(1)

TABLE 2. Characteristics of Antennal Cleaning Organ of Metrius Eschscholtz, and of Ozaenine Genera of

North and Middle America

Middle American Genera of the Tribe Ozaenini

57

Quae st. Ent 1990, 26(1)

Platycerozaena C small, term. slight 0 setae small long, sinuation

at MEx sparse slight

58

Ball and McCleve

Hlavac (1971: 56) places Metrius and the Australian ozaenine genus

Mystropomus . in grade B, characterized as follows: tibial spurs not part of

cleaning mechanism; setal band long, with distinct vertical section and confluent

zone, length of setal band/length of tibia 26-58 per cent, in most taxa less than 40

per cent; confluent zone short, 15 to 35 per cent length of band; median expansion

evident, in most taxa not shifted far anterad; channel shallow, developed far basad

of clip setae or not; fore tibia not compressed antero-posteriorly. Metrius and

Mystropomus are classified as "advanced Grade B", presumably because of the

extended channels that they exhibit. However, this condition would seem to be

plesiotypic among ozaenines.

The remaining ozaenines are classified by Hlavac as members of Grade C: setal

band long (length of setal band/length of tibia 33-69 per cent), divided into a large

distal region or confluent zone, and a proximal cleaning arc; distal region from 33

to 69 per cent length of setal band; median expansion markedly developed

anteriorly; channel deep, short, not extended above clip setae; anterior and

posterior setal rows, if present, not in from of cleaning aggregations.

The various forms of cleaning organs seem to form the following branched

transformation series:

Following development of the more complex cleaner (C from B), further

development occurs with hypertrophy of the median expansion (Figs. 66A-B, and

67A-B). In contrast, the cleaning organ is reduced in the line Ozaena +

Platycerozaena- Physea (Figs. 68A-B, 69A-B, and 64A-B), with decrease in size

of channel, loss of clip setae and loss or reduction of the anterior setal row. This

reduction probably coincides with modification of other parts of the fore tibiae

(markedly broadened in Physea ), or in modifications of the antennae (reduction in

setae in Ozaena and Platycerozaena; thickening of the antennomeres in Ozaena ,

shortening of the antennae in Platycerozaena ). In any event, it seems unlikely that

the reduced organs can function as antennal cleaners. Darlington (1950: 65) claims

that such reduction is a precursor to total loss of the antennal cleaner exhibited by

paussine adults, many of which have markedly expanded tibiae, and markedly

expanded antennomeres that lack standard tactile and chemosensory setae. We can

appreciate that the antennal modifications of Ozaena and Platycerozaena render the

cleaning organ superfluous, but the antennomeres of Physea seem to have a normal

complement of sense organs. If other carabids with normal antennae need to groom

them, how do Physea adults manage? Do they have some other mechanism, or is

their grooming behavior so modified that they can use effectively the remnants of

their cleaning organs?

Genital segments of males. — These are abdominal segments VIII and IX/X. The

latter segment is either composite, or one of either IX or X ( cf Bils, 1976).

Although Bils' paper treats females, we assume that the tergum that bears the

explosion chambers of the defensive system is the same in both sexes, and Bils

labels this "T IX/X" (/.c., Figs. 9 and 1 1, EK).

Tergum VIII is of about the same form in all taxa examined. We do not comment

further about it. Sternum VIII and the "ring sclerite" (sclerites of segment IX/X) do

exhibit some interesting variation.

Quae st. Ent., 1990, 26(1)

TABLE 3. Characteristics of Male Genitalia of Metrius Eschscholtz, and of the Ozaenine Genera of

North and Middle America

Middle American Genera of the Tribe Ozaenini

00

5

CO

3 w

O 3

•a 2

< <

•g a g

P o 2

guc

5

3 3

c e D

3 2

CQ <

00

c

<D

►J

X

<U

D.

<

. 00

CX c

3 O

• o Vh

<u a o

•Ills

13 0-1 i—i

~0 cd X ai

Oh

> p

,<u

O

I

e «

<U X)

E J2

. o

. JL, a)

C o XI

g .b .2?

03 O 6,

.a w

£ ‘£

<u

.s5 ex

c« £

c 2

a «3

5

(N -2 «

* 73 &

2 E o

C aT 73 •-

S £ <D U

E 2 c J

2 cd O O

a c a a

§

2 B § ° h « 2

cd b o u

.ScS»t

.2? J ■&•§ ^

T3 -5 Cd X o5

« Vh

*- 2 c

cd d O

ex % x

C/3 cd 1/1

cd

o

& u

* S 2

<2 a 8

2 ^

<u O , Cd

g-!3

Es o.y

O O'c j3

o o S

R

<3

<3

b s

*2 a

CA r«

cd tjq

- <u o

cd t

BB-o

oo. « -s

cd c .2

03 C 4-1

1)

X

J-2

§

3 , cd

O O £

s 1 -

<n

r- .3

• T3

2 1

2 -o •=>

•o I 2 s

g E c b

O 3; o *R C

u, O u. O <u>

x c a a >

<u

jS

2 I

R

3

Q>

b S

<2 a

03 r«

cd tq

cd

3

o-

cu

X

59

Quae st. Ent 1990, 26(1)

(sensu stricto) moderate to Entomoant. to length trichia Entomoant. sclerite; blade-like;

marked median lobe few lightly scl. setation

tube various,

(continued on next page)

60

Ball and McCleve

60

2

o3 w

O 03

a 2

< <

JL8

13 =1 «

E o fa

§U<

o3 o3

c/5 fli

03 C

PQ <

60

C

<U

fa

13

<u

3

C

c

o

o

fa

fa

OQ

<

H

fa

l3

§ Z

< fa

z o

- .s 'S

1 £ C

ill

3 1 rrt l—

o -2 u

U ,, T3 60

Soil

3 % O

=- <u

Oh C/5

S £ ■= 'C

a § e «

O- C g O

c3 O *: tfl

<u

> (4-1

•5 ° ■ A

c fa p ^

« -§ s w

x o

<D O

o .y

S ■

2 2

o O £

u

-O

JS

i

O

vO .3

• 13

« g

O fa

c x o

« ° u >>*0

C #s 'tli

•3 73 ^ 13 13

g cd y g G

0 0 3 0 3

«-H >- P »-l O

OhX fa X fa

2

3 13

1 1

g §■

-fa fa

fa o

tj

a O

a, w

3 %%

in

mi

a g's ^

CJ "O

^ 13 13

C 3 03 ti O

E O O C

fa X X > ^

<U c

> o

la _

3 , 03

I go

•fs (J o

£ S 'B

<D

X

J-S

§

o

0^ -3

• 13

8 1

« >> fa

o 13 «

« fa ^

o’ c 55

i_ <u x

Cu > o

fa 60

!*

Quae st. Ent., 1990, 26(1)

Platycerozaena curvature slightly ca. .90 L. without as in as in as in spatulate;

very projected median micro- Ozaena Ozaena Ozaena setose on much

slight ventrally lobe trichia of medial margin;

longer than left.

Middle American Genera of the Tribe Ozaenini 6 1

Figs. 70 — 73. 70, left elytron of Entomoantyx cyanipennis (Chd.): A, apical half, dorsal

aspect; B, flange of Coanda, lateral aspect. 71-73, left stylomeres of females of: 71, Physea hirta

LeC., A- ventral aspect, B- ventral aspect, apical portion, C- medial aspect; 72, Metrius contractus

Esch., A, ventral aspect, B- caudo-ventral aspect, C- apical portion, ventral aspect; 73,

Entomoantyx cyanipennis (Chd.), A- ventral aspect, B- apical portion, ventral aspect. Scale bars =

200pm, Figs. 70A, 71A,C, 72A; 40mm, 70B, 7 IB, 72B-C, 73A-B. Legend: fc- flange of Coanda;

ns- nematiform setae; psf- preapical sensory furrow; psp- preapical sensory pegs.

Quae st. Ent ., 1990, 26(1)

62

Ball and McCleve

Figs. 74 — 78. Left stylomeres of females of: 74, Pachyteles gyllenhali (Dej.), A- medial aspect,

B- caudal aspect; 75, Pachyteles enischnus, new species. A- medial aspect, B- ventral aspect; 76,

Pachyteles mexicanus Chd., A- medial aspect, B- ventral aspect; 77, Pachyteles kuntzeni

(Bann.), A- ventral aspect, B- apical portion, ventral aspect, C- caudal aspect; 78, Pachyteles

parca LeC., A- medial aspect, B-ventral aspect, C- caudal aspect. Scale bars = 100 |J.m, Legend:

an- apical notch; ns- nematiform seta; ts, trichoid seta.

Quaest. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

63

Figs. 79 — 81. Left stylomeres of females of: 79, Pachyteles elongatus (Chd.), A- medial aspect,

B- ventral aspect, C- caudal aspect; 80, Ozaena lemoulti Bann., A- ventral aspect, B-apical portion,

ventral aspect, C- caudal aspect; 81, Platycerozaena panamensis (Bates), A- medial aspect. B-

apical portion, medial aspect, C- caudal aspect. Scale bars = 200 mm, 79A-B, 80A-B, and 81 A;

100 mm, 79C, 80C, and 81B-C.

Quae st. Ent., 1990, 26(1)

64

Ball and McCleve

Figs. 82 — 87. Male genitalia of: 82. Metrius contractus Esch., A- median lobe, left lateral aspect,

internal sac everted, B- same, right lateral aspect, apical portion, C-D, left and right parameres,

respectively, dorsal aspect; 83, Entomoantyx cyanipennis (Chd.), A- median lobe, left lateral

aspect, internal sac inverted, B- same, right lateral aspect, C-D, left and right parameres,

respectively, dorsal aspect; 84, Physea hirta LeC., A- median lobe, left lateral aspect, internal sac

everted, B- same, apical portion, right lateral aspect, C-D, left and right parameres, respectively,

dorsal aspect; 85, Pachyteles parca LeC., A- median lobe, left lateral aspect, internal sac everted,

B- same, apical portion, right lateral aspect, C- same, apical portion, ventral aspect, D-E, left and

right parameres respectively, dorsal aspect; 86, Pachyteles kuntzeni (Bann.), A- median lobe,

left lateral aspect, internal sac everted, B- same, apical portion, right lateral aspect, C- same, apical

portion, ventral aspect; D-E, left and right parameres, respectively, dorsal aspect; 87, Pachyteles

elongatus (Chd)., A- median lobe, left lateral aspect, internal sac everted, B- same, right lateral

aspect; C-D, left and right parameres, respectively, dorsal aspect. Scale bars = 0.5 |im, Legend: 1-

3, major regions of internal sac; a- median lobe, basal articulation point for parameres; ab, apical

brush of internal sac; b- apical portion of median lobe; c- carinula of apical portion of median lobe;

d- digital projection of terminal sclerite of internal sac; r- rod of apical portion of ejaculatory duct;

t- terminal sclerite of internal sac.

Quaest. Ent.< 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

65

Quaest. Ent 1990, 26(1)

66

Ball and McCleve

Figs. 88 — 96. 88 — 91, Male genitalia of: 88, Pachyteles gyllenhali (Dej.), A- median lobe, left

lateral aspect, internal sac everted, B, same, apical portion, right lateral aspect, C-D, left and right

parameres, respectively, dorsal aspect; 89, Pachyteles mexicanus Chd., A-median lobe, left

lateral aspect, internal sac everted, B- same, apical portion, right lateral aspect; C-D, left and right

parameres respectively, dorsal aspect; 90, Ozaena lemoulti Bann., A- median lobe, left lateral

aspect, internal sac everted; B, same, apical portion, right lateral aspect, C-D, left and right

parameres, respectively, dorsal aspect; 91, Platycerozaena panamensis (Bates), A- median lobe,

left lateral aspect, internal sac everted; B, same, apical portion; C and D, left and right parameres,

respectively, dorsal aspect. 92-96, Ring sclerites of segments IX/X of: 92, Metrius contractus

Esch., dorsal aspect; 93, Entomoantyx cyanipennis (Chd.), dorsal aspect; 94, Physea hirta LeC.,

ventral aspect; 95, Pachyteles marginicollis (Sober), dorsal aspect; 96, Platycerozaena magna

(Bates). Scale bars = 0.5 pm, Legend: ab- apical brush of internal sac; ec- explosion chamber of

pygidial gland system; hs- hemisternite of segment IX/X; t- terminal sclerite of internal sac.

Quae st. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

67

Quae st. Ent., 1990, 26(1)

68

Ball and McCleve

Figs. 97 — 106. 97, Male genitalia of Pachyteles enischnus , new species: A, median lobe, left

lateral aspect, with internal sac everted: B-C, left and right parameres. respectively, ventral aspect.

98-100A and 101 — 102, Sternum VIII. ventral aspect, of: 98, Metrius contractus Esch.; 99,

Entomoantyx cyanipennis (Chd.); 100A, Physea latipes Schaum; 101, Pachyteles parca LeC.:

102, Pachyteles mexicanus Chd. 100B, Tergum IX/X, dorsal aspect, of Physea latipes Schaum.

103-106, female reproductive tracts, lateral aspects, of: 103, Metrius contractus Esch; 104,

Entomoantyx cyanipennis (Chd.); 105, Physea latipes Schaum; 106, Pachyteles parca LeC.

Scale bars = 0.5 |im. Legend: be- bursa copulatrix; bs- bursal sclerite; co- common oviduct; ec-

explosion chamber; sgd- spermathecal gland duct; sp- spermatheca; spd- spermathecal duct.

Quaest. Ent ., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

69

In Metrius males, the posterior margin of sternum VIII is broadly notched. The

two anterior projections join at their apices to form a ring. We interpret these

projections as apodemes. Among males of the Ozaenini, sternum VIII consists of a

pair of small lateral sclerites connected to one another by membrane, and without

apodemes.

The ring sclerite (Figs. 92-96) surrounds the genitalia which, during

copulation, protrude between tergum and sternum. The tergum is narrow, and

laterally on each side bears the explosion chambers (Fig. 92, ec). The sternum

comprises a pair of apodemes joined anteriorly to form the ring; to the right one is

connected a hemisternite (Fig. 92, hs), which is more or less extensive. This

hemistemite is larger in Metrius males, being smaller among the ozaenines.

The sternal apodemes of Metrius males, at their junction, form a narrow angle.

This angle is greater among the ozaenines, the junction being broadly rounded in

Entomoantyx males (Fig. 93). The ring is markedly asymmetrical in Metrius and

Physea males (Fig. 94), slightly to the right in Ozaena and Platycerozaena (Fig.

96), and about symmetrical in Entomoantyx and Pachyteles males. These data

suggest the following linear transformation series:

Metrius - » Physea - > Ozaena - > Pachyteles - >Entomo.

Platycerozaena

Although the ring sclerite of Physea is in form much like that of Metrius, in

fact. the former ring is a good deal shorter than the latter, and is thus more derived

than is apparent at first.

Male genitalia. — This system comprises the sclerotized median lobe which has

enfolded in it the actual organ of intromission, the internal sac or endophallus, and

attached externally to the base is a pair of plate-like or digit-like parameres. Stork

(1985: 1120 and Figs. 32-37) provides excellent data about the male genitalia of

the species of Dhanya.

Among Metrius and ozaenine males, the median lobe is a compressed tube,

with an open base, i.e., without a distinct basal bulb (c/. Figs. 82A-91A and 97A).

In lateral aspect, the articulation point for the parameres (Fig. 82A, a) marks the

ventral base of the shaft. The shaft is curved ventrad, more abruptly so apically in

Metrius males, and narrowed in males of the ozaenine genera. The apical portion

(Fig. 82 A, b) located ventrad, is marked in Metrius mr.les by a sharp constriction in

the median lobe, and comprises a thin lobe with apex directed posterad, and bent

sharply away from the ventral curve. Among ozaenine males, the apical portion

varies from non-existent (Figs. 90A-B) to prominent and subtruncate (Figs. 86A-

B), or round and spatulate (Figs. 87A-B). See Table 3 for details.

The internal sac is complicated in form and more so in its armature of

microtrichial fields and microspines derived from microtrichia. An additional piece

of armature associated with the internal sac is the rod (Fig. 82A, r) of the

ejaculatory duct — a trough-like structure on the dorsal surface of the duct, and

which is varied in length. It is designated posterior rod by Stork (1985: 1120),

and is very long in males of Dhanya ( l.c ., Figs. 32-37).

Details of armature are best appreciated with the internal sac everted and, if

possible, fully inflated (Figs. 82A-B, 84A-B, etc.), (for ozaenines, it is difficult to

extend fully the sac; for the few males of Entomoantyx that we had available, it

proved to be impossible). Three areas or fields are recognized: a dorso-basal lobe

of microtrichia (1); a median band or collar of microtrichia or microspines (2); and

Quae st. Ent., 1990, 26(1)

70

Ball and McCleve

an apical lobe that bears the ostium or gonopore and various sclerites and

microtrichial fields (3). Details for the genera are provided in Table 3. Stork ( l.c

Figs. 32-37) figures similar structures for the species of Dhanya.

The parameres are varied in form, setation, and relative size. Illustrations

(Figs. 82-91 and 97) were made from the dorsal aspect. Details are in Table 3.

Compared to the genitalia of Metrius males, those of the Ozaenini exhibit median

lobes with less differentiated apices, longer sclerotized rods of the internal sac,

internal sacs with the apical parts more elaborate, and right parameres of most taxa

broader, and either asetose or with shorter setae. A possible transformation series

is the following:

Physea

t

Metrius - >Entomoant. - >P. (Pachytel.) - tP. ( Gonio .)

Ozaena, Platycerozaena

This series emphasizes the close association of Entomoantyx with Metrius in form

and setation of the right paramere, and internal sac with a concentration of

microtrichia on the dorso-basal area. Farthest from Metrius is subgenus

Goniotropis, with its elaborate apical part of the internal sac. Males of subgenus

Pachyteles seem to occupy a central point in the Ozaenini, with a structural plan of

genitalia sufficiently complex to be ancestral to the other types. Although

elaboration seems to be the main thrust of evolution of the genitalia, reduction

seems to characterize Physea : shorter rod, reduced parameres, and a rather simple

internal sac with the only projection being a large apical lobe. Also, the

microtrichial field in the basal area seems to have been reduced in Physea ,

Pachyteles , Ozaena , and Platycerozaena.

Genital segments of females. — Abdominal terga VIII and I X/X (Fig. 100B) are

essentially the same in Metrius and the ozaenines and are not noted further. Sternum

VIII consists of a pair of hemistemites, each of which in Metrius females has an

asetose broad median part and a broad, anteriorly directed apodeme. Among

ozaenines, there is appreciable intergeneric variation. The median posterior parts

are less extensive in most taxa than in Metrius (Fig. 98), and the apodemes are

narrower. Physea females have the median parts shorter, with more membrane

between (Fig. 100A), and Physea and P. ( Goniotropis ) females (Fig. 101) have

setae variously distributed. Females of Entomoantyx have reduced hemistemites

with the median margins markedly sinuate (Fig. 99). No transformation series is

offered because the complexity of these structures cannot be summarized so

simply.

Ovipositor sclerites. — These consist of a pair of slender valvifers articulated

with tergum I X/X , and articulated to each valvifer a single-articled stylomere

(Figs. 71-81), of varied length and setation, but each with a preapical sensory pit

with one or more nematiform setae. The stylomeres are articulated in such a way

that they are exserted straight posterad, without first being partially rotated from a

flattened position in the body cavity. Presumably, this is a primitive feature of the

paussine stock.

Stylomeres of Metrius females (Figs. 72A-B) are moderately long, broad at

base, and broadly rounded at the apex; the surface, especially dorsally, has

numerous basiconic sensilla; and the sensory furrow is preapical, with a pair of long

Quaest. Ent ., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

71

nematiform setae and several basiconic sensilla (Fig. 72C) that are longer than usual

for Carabidae. The stylomeres of Entomoantyx females (Figs. 73A-B) are similar to

those of Metrius.

The stylomeres of Physea females (Figs. 71A-B) are much more elongate and

slender, the surface with few sensilla, but including two preapical nematiform setae.

The stylomeres of Pachyteles females are various in form and proportions.

Those of P. enischnus n. sp. (Figs. 75A-B) are most like the stylomeres of Metrius :

mandible-like, falcate, with broad base and rounded apex, the nematiform setae

preapical; however, the surface is covered with thick, short trichoid setae. The

stylomeres of P. gyllenhali differ from those of P. enischnus n. sp. by the bifid

apices (Fig. 74B). The stylomeres of P. mexicanus (Fig. 76A-B) are much more

slender, digitiform, cylindrical (at least preapically), the surface with numerous

slender trichoid setae in addition to those that are short and thick, and the sensory

furrow and nematiform setae are apical rather than preapical.

The stylomeres of P. parca (Figs. 78A-C) and P. kuntzeni (Figs. 77A-C) are

markedly similar to one another, and are much like those of P. mexicanus :

cylindrical, with nematiform setae nearly apical; but they are shorter (palpiform) and

with a much denser vestiture of setae. In contrast, the stylomeres of Pachyteles

elongatus are long and very slender (Figs. 79A-C), like those of Physea, except that

the nematiform setae are apical and the surface is moderately densely setose.

The stylomeres of Ozaena (Figs. 80A-C) are short, palpiform, cylindrical, with

the nematiform setae apical, and with numerous trichoid setae apically and

preapically. Those of Platycer ozaena (Figs. 81A-C) are similar to the stylomeres of

Ozaena, but the form is terete rather than cylindrical, and the apical and preapical

setae are longer.

The following linear transformation series summarizes these data:

Metr. - >Entomo. - >P. (s. str .) - >P. ( Gon .)

- 1 -

I - > Ozaena Platycer. Physea

Along the main horizontal axis, there is an overall decrease in stylomere length,

a shift of the sensory furrow toward the apex, and an increase in setation. Physea

females are exceptional, with their long glabrous stylomeres, and Pachyteles

elongatus varies in a similar way.

Bursal sclerite and spermatheca. — At the posterior end of the bursa copulatrix

(be) of female Metrius (Fig. 103), a posterior chamber is extended dorsad the

common oviduct (co), connected to the dorsal wall of the bursa is a spermatheca

(not shown, but like sp in Fig. 104) with a short sinuous duct (spd), and an

extensive spermathecal gland (not shown) connected by a short spermathecal gland

duct (sgd). Table 4 provides details for the ozaenine genera studied. Based on

absence or presence of the bursal sclerite and on form of the latter, the following

branched transformation series is postulated:

Quaes t. Ent., 1990, 26(1)

TABLE 4. Characteristics of the Reproductive Tract of Females of Metrius and of Ozaenine Genera of

North and Middle America

72

Ball and McCleve

5 a

Oh J?

co Q

C3

O

<U

£ C

CO o

is

CO hS

73 ■ c

£ «

OQ co

.‘i

*

t o

5 -c

C C/5

C/5

fc o

O

-2 o

Quaest. Ent., 1990, 26(1)

Platycerozaena very small large dorsal wall narrow, short

of bursa short

TABLE 5. Defensive Secretions Recovered from Adult Metrius and Adults of Three Genera of Ozaenini.

From Eisner, et al 1977; and Roach, et. al ., 1979.

Middle American Genera of the Tribe Ozaenini

-o

c

3

O

— , a

B B

o o

H U

B d

S a

* 8

C T3

O

i N

— CD

>-> <u c

JS x> o

CN — O"

ts o

D N

£ C

.3 <u

ro Tf

(N —

-3 O

>> N

X c y

*2 D C

„ x, o

■? ti

<N — O"

O

N

C D

D C

X) o

^ ■§

X

X X

<3

3

M

<3

O

.a

D

3

CQ

<o

5

O

.O

Oo

<3

t:

D

Dh

.Q

75

2!

<o

£

On

U

D

J

a

!■

pq

73

Quaest. Ent., 1990, 26(1)

sp?; 4-, Platycerozaena panamensis Bates.

74

Ball and McCleve

Defensive secretions and their delivery. — Eisner et al. (1977), Roach et al.

(1979), and Eisner and Aneshansley (1981) analyze in some detail the defensive

secretions and their delivery for ozaenines. Forsyth (1972) describes the structure

of the pygidial glands that produce the secretions, and Deuve (1988: 167, Fig. 10)

contributes an analysis of the position of the openings of the pygidial glands.

The secretions are various benzoquinones and the hydrocarbon n-pentadecane.

Table 5 shows the names of secretions that have been recovered from adults of

Metrius and representatives of three ozaenine genera considered here. The limited

data show a graded series with numbers of compounds increasing from two (in

Metrius ) to three and five in ozaenines. Lack of n-pentadecane from the armory of

Physea would seem to be a loss.

The mixtures are hot and are delivered with explosive force, in a cloud of

corrosive vapor. Details of the process are explained by Eisner and Aneshansley

( loc . cit.), though it was known for 150 years that these beetles were

"bombardiers".

Ozaenines and paussines are able to direct the defensive jets forward by

means of the flanges of Coanda on the elytra (Stork, 1985: 1115), as explained by

Eisner and Aneshansley (loc. cit.).

Another part of the delivery system includes umbilical setae that are expanded

(Figs. 107C-D), and to which droplets of the defensive secretions adhere when

they are fired forward. Also, the beetles brush the secretions over the body, using

the legs (Eisner and Aneshansley, l.c.). Characteristic of Ozaena and Platycerozaena

adults are modified setae (basiconic Type 2; cf. Nagel, 1979b: 27). These are

illustrated in Figs. 107A-B. They are on the lateral margins of the pronotum, as well

as on the elytra. Their flattened, ridged surfaces would seem to be ideal for

increasing the evaporative surface area for the defensive secretions, and thus

enhancing the effectiveness of the latter. Unfortunately, Eisner and his co-workers

did not mention these setae.

Forsyth, and Eisner and Aneshansley report that the pygidial glands of adult

Metrius and the ozaenines are similar in structure, i.e., two chambered. We have not

studied them in detail, though we have seen the explosion chambers located laterally

on Tergum IX/X (cf. Figs. 92-96 and 100B). This, combined with the numerous

other similarities in structural features, is strong evidence for linking these taxa

phylogenetically.

In summary, the defensive system shows remarkable complexity both in

structure and function between Metrius and the Ozaenini + Paussini, linking these

taxa in a single higher-level taxon. The ancestor of the group evidently evolved a

system for development and release of simple hot benzoquinones. The delivery

system was improved by evolution of the flanges of Coanda, and the chemical

system became more complex with development of additional corrosive

components. With data for only four of the six genera, the following branched

transformation series is incomplete:

Sexual dimorphism: adhesive setae of fore tarsomeres. — Males of Metrius

have fore tarsomeres 1 and 2 or 1 , 2 and 3 (Y. Bousquet, personal communication)

Quaest. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

75

Fig. 107. Specialized setae on the elytron of Ozaena lemoulti Bann.: A-B, sensilla basiconica,

type 2; C, umbilical setae; D- sensilla basiconica, type 2. Scale bars = 50 pm.

markedly expanded, and on the ventral surface of each is a pad of adhesive articulo-

setae (Stork, 1980: 287; cf. his Figs. 15B-E; referred to as "spongy pubescence"

by previous authors). Among ozaenines, the vestiture-bearing tarsomeres are less

expanded. Males of Physea , almost all of Pachyteles (s. str .) (Fig. 66C) and some

of subgenus Goniotropis have adhesive setae on only fore tarsomeres 1 and 2.

Males of some species of Goniotropis, and of Entomoantyx have vestiture on

tarsomeres 1-3. Males of Pachyteles gyllenhali *. Ozaena and Platycerozaena (and

some species of Dhanya — see Stork, 1985: 1115) lack adhesive vestiture and none

of the fore tarsomeres are expanded. The data are summarized in the following

branched transformation series:

Trends are in two directions: increase in number of tarsomeres with vestiture;

and complete loss of adhesive vestiture.

Quae st. Ent., 1990, 26(1)

76

Ball and McCleve

Summary

In the absence of evidence that the New World ozaenine genera are

monophyletic relative to their Old World counterparts, only a limited evolutionary

analysis seems practical. This analysis consists of statements about phylogenetic

distance from the putative out-group, Metrius, and about striking features of

evolutionary divergence in particular character systems.

In terms of overall divergence, as measured by a summation of the relative

distance from Metrius in each of the postulated transformation series, the sequence

of genera is as follows, from most like to least like: Entomoantyx — Pachyteles —

Physea — Ozaena-Platycerozaena. Probably this sequence does not represent a

system of sister-group relationships, though Ozaena and Platycerozaena must be

sister groups. Physea is strikingly aberrant in details of mandibles, male and female

genitalia and ovipositor, as well as body form. The few defensive secretions lend

credence to the proposition that this genus, though aberrant, is from a basically

primitive stock.

A reviewer of a previous draft of the manuscript on which this paper is based,

conducted a numerical cladistic analysis of the data, using the program HENNIG

86, and reported that a single most parsimonious cladogram resulted, expressed

linearly as a series of inclusions: Metrius + {Entomoantyx + {Pachyteles + {Physea

+ {Ozaena + Platycerozaena)))). The consistency index is 0.81 or 26/32,

indicating that there are only six extra steps in the cladogram of the six genera.

The sequence of genera is exactly the same as in the linear arrangement

presented above. Nonetheless, as explained above, we do not believe that

evolutionary significance should be accorded to the branching pattern, i.e., that

Physea is really sister-group of Ozaena + Platycerozaena.

We note two striking developments in the New World Ozaenini. Both may

involve close association with ants. One of these developments is an escape from

the constraints of life under bark, and is represented by the genus Physea. Body

form is markedly modified {i.e., broadened, as in Fig. 2; cf. Figs. 3 and 4). Other

divergent features of Physea are noted above. We do not know their functional

significance in relation to life in an environment with fewer physical constraints, or

if some of these modifications evolved in response to life with ants. An explanate

body plus flattened appendages may have to do with provision of a dorsal shield

under which the appendages can be concealed as a safeguard against attacks by

ants. The elongate antennae, which we believe are secondarily elongated, are

anomalous, especially considering the reduced antennal cleaner. One would think

that short antennae would be more easily protected from the ravages of ants.

A second striking development involves the genera Ozaena and

Platycerozaena. The more evolved chemical defensive system includes possibly

setae that are specialized for enhanced effectiveness in dispersion of the defensive

secretions, which in turn are more numerous than in other New World ozaenines

(demonstrated for Platycerozaena ; inferred for Ozaena). This more complex system

suggests evolution to cope with more efficient enemies — as for Physea , possibly

ants. Loss of the normal tactile setae from the dorsal surface of head, pronotum and

elytra also suggests association with ants — by analogy with lack of setae in

paussines. Other striking modifications of these two ozaenine genera involve:

antennae, with general reduction in sensilla trichodea, modification in form, and

antennomere 1 1 with its concentration of sensilla basiconica; various

modifications of labrum, maxillae, and labium, suggesting changes in food or

feeding mechanisms; and the peculiarly modified palpiform stylomeres of the

Quaest. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

77

females, which suggest a sensory function rather than one of simple deposition of

eggs.

We believe that the reduced antennal cleaner of the fore tibia has developed

twice: once in the ancestral stock of Physea + Physeomorpha, and once in the

ancestral stock of Ozaena + Platycerozaena. This feature, plus loss of tactile setae

from the Ozaena-Platycerozaena lineage, presages similar losses from the paussine

lineage, as pointed out by Darlington (1950) for Physea. He noted, as well, the

myrmecophilous habits of Physea and the paussines. Another similarity to the

paussine lineage is the peculiar sensilla basiconica of adults of Ozaena and

Platycerozaena. Conceivably, either the Physea or Ozaena lineage could be the

sister group of the paussines, though the similarities might reflect simply parallel

developments, based on an underlying general similarity.

The general ecologically-based scenario that we think the data reflect is one

of: evolution in the tropics of form, structure and habits for life under bark

( Entomoantyx and Pachyteles)\ differentiation in the sub-cortical habitat

(P achy teles)’, secondary loss or modification of these features, associated with

development of a myrmecophilic existence {Physea and Physeomorpha). We do not

have information about habitat for the Ozaena-Platycerozaena lineage, but we

judge from body form of adults that they are sub-cortical, and we believe that they

are adapted for life with ants, too.

A different scenario, based on lack of association of Physea adults with sub¬

cortical habits, places its lineage at the base of the Ozaenini.

Finally, we believe that Metrius is a relict genus, surviving in the temperate

zone, living in an ancestral-type habitat of leaf litter, beyond the periphery of the

range of its more evolved ozaenine relatives. In many body features, adults are

primitive, but not in body form, absence of wings, or in development of the large

intercoxal process of the prosternum. Thus, this basically primitive group has

evolved its own special features.

TAXONOMIC TREATMENT

Provided in this section are: description of the Tribe Ozaenini based on

features of adults of New World taxa; descriptions of the genera known from

Middle America; and descriptions of selected species. These species are either

known from the United States, or their northern limit of geographical range is near

the U. S. -Mexican border, or their structural features indicate aspects that we found

instructive in appreciating the divergence of the New World Ozaenini.

For the species, we report limited quantitative data: range of body size

(Standardized Body Length and maximum width). The only generally useful ratio

that we found and report here is width of elytra/ Standardized Body Length

(W/SBL). Two groups are recognized in the material that we examined: species

with low values (0.31-0.35), whose adults have long slender bodies; and species

with higher values (0.37-0.46), whose adults have broader, stouter bodies.

Tribe OZAENINI

Description. — Adults small to average for Carabidae (SBL ca. 3-18 mm.). Body form

sub-cylindrical to terete and explanate.

Color. Body uniform black to testaceous, or various combinations of these somber colors,

or aeneous-green; elytra somber-colored, or aeneous, like rest of body, or bright metallic blue-

green.

Quae st. Ent., 1990, 26(1)

78

Ball and McCleve

Microsculpture. Most members with mesh pattern of dorsal sclerites (including elytra)

isodiametric, mesh pattern of lateral and ventral sclerites predominantly transverse, though

surface not grated (Allen and Ball, 1980: 487); some members with mesh pattern of body

sclerites and elytra uniformly isodiametric; microlines distinctly developed, or partially effaced

and thus difficult to see at magnification of 50X or less, or absent (Fig. 107 A).

Luster. Dorsum (including elytra) dull (most members) or glossy, venter more glossy, or

dull as dorsum.

Vestiture. Surfaces of body and elytra either with or without pile of short trichoid (hair-like)

setae, pilosity dense or sparse; members of most taxa with ventral and lateral sclerites sparsely

pilose. Articles of legs sparsely or densely pilose. Members of some taxa with setae on head and

prothorax, and/or serially arranged setae on various intervals of elytra. Wider, flat, costate setae

(Fig. 107A and B) on dorsal surfaces of pronotum and elytra of members of some taxa. Fore

tarsomeres 1-3 of males (Fig. 66C) with or without adhesive vestiture. Mandibles, cardines,

submentum and mentum with or without pilosity.

Fixed (or standard) setae ( i.e ., those characteristic of most carabid taxa) and spines.

Clypeus with one to three pairs of setae laterally, or asetose. Head with one pair of supraorbital

setae (members of some taxa with supraorbital setae indistinguishable from other long trichoid

setae of head), or asetose. Antennae: trichoid setae as follows — antennomere 1 with one to

several; antennomeres 2 and 3 with ring of few near apex, antennomere 4 with one or several

rings of few, preapically; antennomeres 5-11 with many to few, variously arranged, or

antennomere 11 without; other setae — shorter, thicker basiconic sensilla on antennomeres 5-11,

associated with rugose matt areas variously, from numerous and widely distributed on lateral

surfaces to very few concentrated in small oval areas on ventro-lateral surfaces, or toward apex of

antennomere 11 (Figs. 29A&B, 30). Labrum pre-apically with seven to 12 forward-directed

setae, or asetose ( Platycerozaena members). Mandibles each with or without single long seta in

scrobe (Fig. 40A) (members of Physea with numerous long setae, one of which may be the

normal seta of the scrobe). Maxillary setae: cardo asetose or with one to three; stipes, two;

palpomere, several. Labial setae: submentum, two to five or six; mentum, two to five or six; glossal

sclerite, two to four; palpomeres 2 and 3 plurisetose (Figs. 44-47); palpomere 1 glabrous or with

few setae. Pronotum with more than two pairs of lateral marginal setae, or without marginal setae.

Prosternum with apex glabrous or sparsely setae. Elytra with umbilical setae (each elytron with

about 30), without parascutellar setae, with or without (members of few taxa) discal setae. Tibiae

each with several rows of spines. Tarsomeres 1-5 ventro-laterally with one or two rows of setae on

each margin, ventral surfaces of tarsomeres 1-4 with patches of longer setae. Standard setae of

abdominal sterna not distinguishable from vestiture.

Head. Form approximately quadrate. Clypeus transverse, surface plane, anterior margin

slightly concave. Frons and vertex uniformly slightly vaulted, or frons with pair of indistinctly

delimited impressions near fronto-clypeal suture. Supra-antennal areas ridged laterally, flat or

reflexed. Occipital area broad, not constricted, or with shallow groove dorsally. Temporal areas

either small or swollen laterally, each as narrow lobe extended on posterior surface of eye.

Antennal fossae close to anterior margin of eye, sub-antennal area either plane or depressed.

Gula average. Eyes elliptical, with long axis either parallel or perpendicular to long axis of head.

Antennae. Varied in relative length, from shorter than combined length of head and

pronotum, to about one third of body length; filiform to sub-clavate; antennomere 1 (scape)

large; antennomere 2 (pedicel) shorter than scape, small; antennomeres 3 and 4 elongate or

short and moniliform; antennomeres 5-10 sub-cylindrical, longer than wide (Fig. 10); or more or

less compressed, extended ventro-laterally, broad surfaces anterior and posterior with antenna

extended at right angles to longitudinal body axis, either longer than wide (Fig. 11), quadrate

(Fig. 13), or wider than long (Fig. 15); or moniliform, short, thick and cylindrical (Fig. 14);

antennomere 1 1 (Figs. 25-30) more or less enlarged, either longer or wider than antennomeres

5-10, shape various — pre-apically sub-cylindrical (Fig. 25) or more or less transverse (Figs.

29A-30), apex blunt or narrowly keeled, keel straight (Fig. 29A) or sinuate (Fig. 30).

Mouthparts. Labrum transverse, narrow (Fig. 32), or broader (Figs. 33, 35, and 36), or

almost as long as wide (Fig. 37); anterior margin straight or broadly concave, or narrowly notched

(Fig. 37). Mandibles: trigonal, with broad bases, symmetrical in general outline, scrobes broad,

distinctly marked, retinaculum very large (Figs. 40A-D) or small (Figs. 39A-D); terebral

margins prominent, posteriorly extended or not on dorsal surface of retinaculum (Fig. 42A);

terebral tooth evident on both left and right mandibles (Figs. 39A,B- 43A,B); retinaculum

Quaest. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

79

prominent or not, posterior tooth more or less evident, directed posterad, anterior tooth

prominent or not; molar area variously toothed; ventral groove more (Fig. 40C-D) or less (Fig.

39C-D) extensively densely setose. Maxillae: average for Carabidae; lacinia with apical tooth

normal (long and slender, Fig. 46), or shorter and broader, chisel-like (Fig. 47B); galeomeres 1

and 2 of various proportions; palpomeres various, 4 relatively slender (Fig. 45) to broad (Figs.

47A and 48). Labium: submentum narrow, transverse; mentum transverse, lateral lobes of various

lengths, more or less tapered apically; epilobes prominent; mental tooth large or small, apically

not notched (in New World taxa); ligula uniformly sclerotized, paraglossae adnate to glossal

sclerite, latter with blunt apex. Palpomeres 2 and 3 subequal, longer than 1; palpomere 3 slender

or broad, apical margin broad, subtruncate.

Prothorax. Pronotum transverse or longer than wide, short compared to length of elytra;

base and apex similar in width, but base distinctly narrower than apex; disc slightly convex,

median longitudinal and anterior transverse impressions shallow, posterior transverse impression

shallow to deep; lateral grooves distinct, postero-lateral impressions shallow to deep; sides plane

or reflexed, lateral margins broad or narrow, evenly curved throughout length or more or less

sinuate posteriorly; antero-lateral and postero-lateral angles various, from broadly obtuse to

narrowly acute. Proepipleura broad, either sharply extended laterally and horizontal to

pleuronotal margin, or extended dorsad in same plane as pleuron. Pleural sclerites average for

Carabidae. Prosternum with intercoxal process broad or slender, with apex near posterior

margins of coxae.

Pterothorax. Mesepisternum and mesosternum separated by suture. Mesepimeron

extended to margin of mid-coxa, broad, posterior margin sinuate, medially overlapping

metepisternum. Mesosternum narrow, intercoxal process in contact or not with process of

metasternum. Metathorax of average proportions, or distinctly shortened. Metepisternum long

and slender, with lateral margin much longer than width of basal margin; or short and broad, with

lateral and basal margins subequal. Metepimeron distinct, wide in relation to apical width of

metepisternum. Metasternum average for Carabidae.

. Elytra. Elongate, parallel-sided, or expanded and inflated, with widest point near middle,

and tapered both anteriorly and posteriorly. Disc flattened or somewhat inflated, tapered

gradually apically. Basal ridge short, extended no farther than plane of base of interneur 4, and

anterad discal plane. Humeri each denticulate or rounded. Apical margin oblique. Flange of

Coanda (Fig. 70B) preapico-lateral. Intervals moderately convex to flat, interneurs shallow,

punctate or not, obsolescent or absent. Epipleura average for Carabidae.

Metathoracic wings. Fully developed in adults of most taxa, variously reduced in few.

Oblongum cell large, quadrangular; wedge cell absent.

Legs. Coxae, trochanters, middle and posterior femora and tibiae average for carabids, or

tibiae markedly compressed (Figs. 64A-B). Anterior and posterior coxae separated from one

another by intercoxal processes; middle coxae separated or not by intercoxal processes. Front

femur more or less cylindrical or flattened, with or without antero-ventral projections (Figs. 58-

61A,B) or swellings (Fig. 57). Front tibia isochaete (both spurs terminal), expanded apically or

flattened and generally explanate. Antennal cleaner with channel large (Figs. 66A-B) or smaller;

medial expansion prominent (Figs. 67A-B) or not (Figs. 68A-B); clip setae (Fig. 65) present

or absent; anterior setal row with setae large (Figs. 67A-B) or small (Figs. 64A-B); setal band

not sinuate (Fig. 68B) to markedly so (Figs. 67A-B). Tarsomeres thickened, with dorsal surfaces

slightly depressed, 5 longer than any of 1-4, tarsal claws smooth, not denticulate.

Abdomen. Sterna II- VII and tergum VIII average for carabids, without distinctive features.

Males with sternum VIII with pair of small hemisternites, without apodemes, joined medially by

membrane. Ring sclerite (Segment IX-X, Figs. 92-96) of males with tergum narrow, laterally each

side with sclerotized explosion chambers ( ec ) of pygidial gland system; sternal apodemes joined

in form of symmetrical or markedly asymmetrical ring; right hemisternite various in form,

bilobed or not. Females with sternum VIII (Figs. 98-102) with pair of more or less reduced

hemisternites, joined medially by membrane; hemisternites setose or asetose, each with short

apodeme projected anteriorly. Tergum IX/X narrow.

Male genitalia. Median lobe compressed, basal opening wide; apical portion various, from

simple short point to more or less elaborate ventrally-directed projection (Fig. 87A) or cleft

(Fig. 89B); apical orifice dorsal. Ejaculatory duct in median lobe surrounded by long, rod-like

sclerite ( rod, r. Figs. 83A-91A, and 97A). Internal sac short, infolded but only partially

inverted, with three areas: basal asetose membranous area; medial collar, with microtrichia; and

Quaest. Ent., 1990, 26(1)

80

Ball and McCleve

apical portion variously lobed and with or without sclerites (Figs. 83-91, and 97). Parameres

extensive, left broad basally but tapered apically, apex with or without few setae. Right paramere

long, apex extended nearly to apex of median lobe, medial margin fringed with setae, more or

less extensively (Figs. 83-91, and 97).

Ovipositor. Ventral surface ventrally directed in repose. Valvifer slender, moderately long.

Stylomere single sclerite, slender, of various length, variously setose and spinose, apex pointed

(Fig. 71A), more broadly rounded (Fig. 81A), subtruncate (Fig. 73A), or cleft (Fig. 74B ).

Sensory furrow preapical (Figs. 71, 74A-B) or apical (Figs. 76A-B) with long nematoid setae and

furrow peg setae, margins without ensiform setae.

Way of life. — Little is known about this topic (Ball, 1960: 94). Adults of

various species of Pachyteles ( sensu lato) have been collected from under bark of

fallen tree trunks in lowland wet tropical forests and from under bark of fallen

cottonwood poplars along waterways in semi-desert areas. Adults of Pachyteles

kuntzeni have been collected at night, on dead oak trees and stumps. Adults of

Physea hirta LeConte have been collected in association with leaf-cutter ants of the

genus Castelnau. Most ozaenine adults, however, have been collected at ultra¬

violet light at night, showing that they are active noctumally.

Adults and larvae of the non-myrmecophilous species probably are general

predators under bark, an inference based on their general structural features.

Forsythe (1982 and 1987) recognizes three types of feeding in carabids: fluid

feeding; fragmentary feeding; and mixed feeding. Mixed feeders are of two types:

either predominantly zoophagous or predominantly phytophagous. Each of these

types has characteristic mandibles and maxillae. Details of mandibular and maxillary

structure of adult ozaenines indicates that they are mixed feeders, predominantly

zoophagous. This means that individuals take particulate animal matter, but secrete

digestive juices extra-orally, on the prey. Thus, presumably, a certain amount of

digestion is extra-oral.

Geographical distribution. — In the New World, the range of the Ozaenini

extends from southern South America (Argentina in the east and Chile in the west)

northward through Central America to southwestern United States (southern Texas

in the east, and southern Arizona in the west).

Included taxa. — Six New World genera are included in the Ozaenini:

Entomoantyx , new genus; Physea Brulle; Physeomorpha Ogueta; Pachyteles Perty;

Ozaena Olivier; and Platycerozaena Banninger. Of these, Physeomorpha is confined

to South America, and Entomoantyx to Middle America. The remaining genera are

in both South and Middle America, and Physea , Pachyteles and Ozaena reach their

northern limits in southern-most southwestern United States.

Key to Genera of Metriini and Ozaenini of Middle and North

America, and Species of Southwestern United States and Northern

Mexico

1

r

2

Fore tibia with both spurs terminal (Fig. 65B). Fore coxal

cavities closed by medial extension of proepimera.

Pterothorax with middle coxal cavities disjunct (mesepimeron

extended to middle coxa) . 2

Combination of character states otherwise .

. other tribes of Carabidae.

( 1 ) Elytron without flange of Coanda (cf. Fig. 70A). Mentum with

tooth notched apically (Fig. 49) or not. Metathorax with

metepistemum short, length of lateral margin subequal to width

at basal margin. Mandible with single seta in scrobe .

Quaest. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

81

2'

3

3'

4

4'

5

5'

6 .

6’

7

T

8

9

9'

10

. Tribe Metriini, Metrius Eschscholtz. (not treated further).

Elytron with flange of Coanda (Fig. 70A), body form various.

Tooth of mentum not notched apically (Figs. 50-55).

Mandibles with one or many setae in scrobe .

. Tribe Ozaenini . 3

(2') Mandible with single long seta in scrobe (Fig. 40A). Antennae

filiform, antennomeres 5-11 densely setose, 1-4 sparsely so.

Elytra metallic blue-green, sharply contrasted with rufous head

and prothorax . Entomoantyx cyanipennis (Chaud.), p. 84

Mandible with several (five or more) setae in scrobe (Figs.

39A, 41 A). Antennae various in form. Dorsal surface more or

less uniform in color, black, piceous, or testaceous . 4

(3') Tibiae markedly compressed and broad (Fig. 64A). Fore femur

with prominent broad swelling near base (Fig. 57). Head with

broad depression (antennal fossa) beneath anterior margin of

eye . Physea Brulle . 5

Tibiae sub-cylindrical, not compressed and broad. Fore femur

with broad swelling or spine (Figs. 58-62). Head without

antennal fossa . 6

(4) Elytra with uniform vestiture of short setae, in addition to long

tactile setae . Physea hirta LeConte, p. 87

Elytra without vestiture of short setae, only several rows of

long, tactile setae . Physea latipes Schaum, p. 87

(4') Anterior tibia with prominent clip setae associated with

antennal cleaning channel (Fig. 65B), latter extended beyond

median expansion. Head, pronotum and elytra with normal long

setae, with or without more or less dense vestiture of short

trichoid setae. Mandibles with dorsal surfaces impunctate .

. P achy teles Perty . 7

Fore tibia without clip setae, channel of antennal cleaner

terminated at plane of median expansion (Fig. 68A). Head,

pronotum and elytra without normal long setae, glabrous, or

with vestiture of short, thickened setae (Fig. 107A). Mandibles

with dorsal surfaces more or less densely punctate (Figs. 42A,

B-43A, B) . 12

(6) Middle coxae in contact with one another, not separated by

intercoxal extensions of meso- and metastema . 8

Middle coxae separated by extensions of meso- and

metastema . 1 0

(7) Pronotum and dorsal surface of elytra with vestiture of short

setae . Pachyteles gyllenhali (Dejean), p. 94

Pronotum and elytra with sparse, longer setae, those of elytra in

rows on intervals . 9.

(8') Pronotum with anterior and posterior lateral angles acute .

. Pachyteles mexicanus Chaudoir, p. 97

Pronotum with anterior angles rounded, posterior angles about

rectangular . Pachyteles enischnus, new species, p. 96

(7') Pronotum narrow, only slightly wider than, or about as wide as,

head . Pachyteles elongatus (Chaudoir), p. 93

Quaest. Ent., 1990, 26(1)

82

Ball and McCleve

10'

1 1

11’

12

12’

Pronotum broader, distinctly wider than head . 1 1

(10') Smaller (SBL 9- 1 2 mm.), pronotum narrower (Fig. 3) .

. Pachyteles parca LeConte, p. 90

Larger (SBL 15-18 mm.), pronotum broader (Fig. 4) .

. Pachyteles kuntzeni (Banninger), p. 92

( 6') Antennae long, about one third length of body; antennomere 1 1

with shagreened area confined to apical third (Fig. 22), apical

ridge straight (Fig. 29A) . Ozaena lemoulti Banninger, p. 100

Antennae short, claviform, antennomeres 5-11 markedly broad and

flat; antennomere 1 1 with shagreened area apical and extended

along dorsal and ventral margins (Fig. 23); apical ridge sinuate

(Fig. 30) . Platycerozaena Banninger, p. 101

Entomoantyx, new genus

Frontispiece, and Figs. 11, 19, 26, 33, 40A-D, 51, 58, 65A-B, 70A-B, 73A-B, 83A-

D, 93, 99, and 104.

TYPE SPECIES: Ozaena cyanipennis Chaudoir, 1852: 40. Here designated.

Ozaena ; Chaudoir, 1852: 40. — 1854: 307.

Pachyteles ; Chaudoir, 1868: 66.— Bates, 1881: 27.— Csiki, 1927: 427.

Tropopsis ; Banninger, 1927: 207. — Blackwelder, 1944: 23. — Reichardt, 1977: 377. — Erwin,

1979B: 557.

Notes about synonymy. — The type species of Entomoantyx was removed

from Pachyteles and included in Tropopsis Sober by Banninger on the basis of

plesiotypic features (middle coxae separated; fore femur without ventral spine;

antennal cleaner of fore tibia less modified). But, in apotypic features of antennal

form and development of the antennal cleaner, T. marginipennis Sober (type species

of Tropopsis) is more like Pachyteles than like E. cyanipennis. Furthermore, E.

cyanipennis is characterized by a unique derived feature (form of mental epilobes),

reduced number of labral setae (seven or eight), and more plesiotypic features than

Pachyteles or Tropopsis in retaining a scrobal seta, a hardly modified antennal

cleaner, and probably structure of the male genitalia (presence of a dense dorso-

basal patch of microtrichia, and styliform right paramere). We believe, thus, that E.

cyanipennis is not closely related to either Tropopsis or Pachyteles. For these

reasons, we remove E. cyanipennis from Tropopsis. Because E. cyanipennis does

not have a named genus for assignment, we have proposed a new one. Because the

differences between Tropopsis and Pachyteles are slight, we combine these taxa,

with the former name being a junior synonym. See below for details.

Derivation of generic name. — From the Greek entomon, cut, and antyx,

margin, a name that alludes to the scalloped lateral margins of the pronotum of adult

E. cyanipennis.

Recognition. — Adults are recognized easily among Middle American

ozaenines by the bright bluish-green elytra that contrast strikingly with the rufous

head and pronotum. The lateral margins of the pronotum (Frontispiece) are

scalloped. Body size is small. The antennae, though rather short, have antennomeres

5-11 longer than wide (Fig. 11). Males have fore tarsomeres 1-3 with adhesive

setae ventrally, median lobe of genitalia with a broad truncate apex (Figs. 83A-B),

and internal sac with a sagittate dense patch of microtrichia dorso-basally (x,Fig.

83B), and left paramere digitate and densely setose apically. Females are

Quaest. Ent ., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

83

characterized by rather short, broad stylomeres (Figs. 73A-B), with subtruncate

apex and sensory groove markedly preapical.

Description. — Habitus as in Frontispiece. Size small, SBL ca. 3. 5-4. 5 mm., width 1.5-2

mm.

Color. See "Recognition" section, above.

Microsculpture and luster. Pterothoracic pleura and sterna with mesh pattern transverse;

otherwise, as described for tribe. Surface generally slightly shining.

Vestiture. Tarsomeres with dorsal surfaces sparsely setose; fore tarsomeres 1-3 of males

ventrally with adhesive vestiture. Thoracic and abdominal sterna with sparse covering of rather

long setae. Otherwise, as described for tribe.

Chaetotaxy. Clypeus with two pairs of setae. Vertex of head with one pair of supraorbital,

one pair of paramedial, and one pair of paralateral setae, latter two groups in row across vertex.

Antennae: antennomere 1 with several setae; antennomeres 2-4 with apical row of long setae;

antennomeres 5-11 with dense covering of setae, except nearly glabrous areas on anterior and

posterior faces. Mouthparts: labrum with seven to eight long setae near anterior margin;

mandibles each with one long seta in scrobe (Figs. 40A-B); maxillary stipites each with two lateral

setae; submentum, mentum, and glossal sclerite each with one pair of setae. Pronotum with three

pairs of lateral setae: one pair anteriorly, one pair near middle, and one pair anterior to postero¬

lateral angles; one pair of paramedial setae near anterior margin. Each elytron with about five

setae in intervals 3 and 5, umbilical setae ca. 15. Legs, number of setae (fore, middle and hind):

coxae, O-numerous-numerous; trochanters, 1-1-1; femora, numerous, numerous, numerous.

Head. Frons without impressions. Eyes moderately prominent (Frontispiece); temporal

area each side small, not swollen. Supraantennal area each side with sharp ridge.

Antennae (Figs. 11, 19 and 26). Filiform, antennomeres 5-10 slightly longer than wide,

antennomere 11 distinctly so (Fig. 11). Antennomere 11 (Fig. 19) with apex about symmetrical,

moderately broadly rounded; circular in cross section, apical margin without distinct keel (Fig.

26).

. Mouthparts. Labrum (Fig. 33) transverse. Mandibles (Figs. 40A-D): see Table 1 for details.

Maxillae: average for Ozaenini, as in Figs. 45 and 46. Labium (Fig. 51): mentum with prominent

tooth, lateral lobes broadly rounded apically, epilobes each with sharp tooth near apex;

palpomere 3 triangular, with apex subtruncate.

Prothorax. Pronotum (Frontispiece) transverse; lateral grooves moderately broad; disc

convex; linear impressions (anterior transverse and median longitudinal) clearly impressed;

lateral margins scalloped; postero-lateral impressions deep. Propleura and prosternum as

described for tribe.

Pterothorax. As described for tribe, and middle coxal cavities separate, with meso- and

meta- intercoxal processes in contact with one another. Metepisternum not overlapped by

extension of mesepimeron.

Elytra. Intervals broad, indistinctly elevated, interneurs shallow, indistinct. Basal ridge

extended to about base of interneur 5.

Legs. In most features, as described for Ozaenini. Fore femur terete in cross section,

without ventral projections. Antennal cleaner of fore tibia (Figs. 65A and B): grade C (see under

"Structural and Biochemical Features"); details in Table 2.

Abdomen. Segments II- VII with tergum and sternum unmodified, or description of

segments VIII and IX/X, see under "Structural and biochemical features, genital segments".

Male genitalia (Figs. 83A-D) and ovipositor Figs. 73A-B). See these items under

"Structural and biochemical features", and Table 3 for details of male genitalia.

Ovipositor (Figs. 72A-C). Stylomeres of moderate length, broad at base. For details, see

this topic under "Structural and biochemical features".

Bursa copulatrix and spermatheca. See Fig. 104 and for details, Table 4.

Way of life. — Members of this genus live in lowland tropical forests. Adults

have been collected under bark of fallen tree trunks, and at U-V light, at night. The

distinctive color pattern of red and blue suggests aposomaty, and this may be

indicative of a way of life that differs from other New World ozaenines — for

example, more time spent by adults in situations where they are exposed to

predators that hunt using visual stimuli.

Quae st. Ent., 1990, 26(1)

84

Ball and McCleve

Geographical distribution. — This genus is known only from the Gulf-

Caribbean Versant of Middle America: from Nicaragua northward to the state of

Veracruz, Mexico.

Relationships. — This genus is either a very primitive member of the

Pachyteles assemblage of Ozaenini, or possibly even the sister group of the other

assemblages of ozaenine genera, excluding the Australian Mystropomus Chaudoir

and Oriental Anentmetus Andrewes.

Included species. — Entomoantyx includes only the species E. cyanipennis

(Chaudoir).

Entomoantyx cyanipennis (Chaudoir)

Ozaena cyanipennis Chaudoir, 1852: 40. TYPE MATERIAL: three males, each labelled "Ex

Musaeo Chaudoir [red print] in front of following box label: "cyanipennis Chaud. Mexique

57 Salle"; Chaudoir-Oberthur Coll., Box 132 bis (MNHP). LECTOTYPE (here selected),

first specimen in series. — Species synonymy same as for genus, above.

Ozaena cyanoptera Thomson, 1856: 330. TYPE MATERIAL: HOLOTYPE female, labelled "Ex

Musaeo Chaudoir" [red print], in front of following box label: "cyanipennis Chaud Bui Mus

1852 p. 40 cyanoptera Thomson Ann Soc Ent 1856 p. 330 Toxpam".

Pachyteles cyanipennis Chaudoir, 1868: 66. — Bates, 1881: 27. — Csiki, 1927: 428.

Tropopsis cyanipennis ; Banninger, 1927: 207. — Blackwelder, 1944: 23. — Reichardt, 1977: 377.

Pachyteles cyanoptera; Chaudoir, 1868: 66.

Notes about synonymy. — Chaudoir (1868: 66) recognized the taxonomic

identity of the types of Ozaena cyanipennis and O. cyanoptera , and established the

synonymy, accordingly.

Recognition. — Among New World ozaenines, adults of E. cyanipennis are

unique in color pattern: rufous head and prothorax, and metallic blue elytra. Adults

are small in size.

Habitus as in frontispiece. Standardized Body Length (male) 4.4 mm; females

3.8-4. 1 mm. Width, male 1.92 mm., females 1.6- 1.7 mm; W/SBL male 0.44, females

0.42-0.43. Other features as noted for genus Entomoantyx , above.

Way of life and geographical distribution. — As above, under Entomoantyx.

Material examined. — In addition to the types, we have seen 14 specimens

from Mexico, as follows.

Chiapas. 27 km. SW Simojovel, VII. 17. 1962; J.M. Campbell (CNCI). Veracruz.

Cordova, VI.29.1966; J.S. Buckett, M.R.& R.C. Gordon (CISC). 33 km. NE Catemaco, Los Tuxtlas

Biological Station, 160 m, VII. 1983; S.& J. Peck (UASM). Coyame, at Lake Catemaco, VII. 1-

15.1963; D.R. Whitehead (UASM). Lake Catemaco, V.24-25. 1969; H.F. Howden (UASM). Dos

Amates, VI. 16-17. 1969; D. Bright & J.M. Campbell (CNCI). Sontecomapan, VI. 10. 1969; H.F.

Howden (UASM).

Physea Brulle, 1834

Physea 1834: 473. TYPE SPECIES: Trachelizus rufa Brulle [= Ozaena testudinea Klug, 1834:

80]; by monotypy.— Chaudoir, 1854: 289, 310.— 1868: 72.— Lacordaire, 1854: 160.—

Bates, 1881: 27.— Csiki, 1927: 431.— Banninger, 1927: 212.— Van Emden, 1942: 25.—

Blackwelder, 1944: 23. — Jeannel, 1946: 47. — Darlington, 1950: 50, 51, 65. — Ball, 1960:

94.— Reichardt, 1977: 376.— Thompson, 1979: 214.— Erwin, 1979: 557.

Trachelizus Brulle, 1834: 258. TYPE SPECIES: T. rufa Brulle, 1834: 259; by monotypy.—

Sober, 1836: 598.— Chevrolat, in d'Orbigny, 1848: 626.— Chaudoir, 1854: 310.

Trachelyzus Chenu, 1851: 89.

Notes about synonymy. — The generic name Trachelizus appeared first in the

Dejean catalogue (1834: 243), credited to Chevrolat, for a genus of brentids. Brulle

(in Audo.uin and Brulle, 1834: 258) used this name, credited to Sober, for a genus of

Quaest. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

85

ozaenine carabids, with T. rufus Sober as the only included species, and thus type

of this genus. Subsequently, in an addendum to the same volume (1834: 473),

Physea Brulle was proposed as a replacement name for Trachelizus, on the basis

that the latter name was, in effect, a junior homonym of Trachelizus Dejean. Also

implied in the text was that Ozaena testudinea Klug, 1834 was a senior synonym of

T. rufus Brulle. In any event, the name of the type species of Physea must be

Trachelizus rufus, though the correct name of the species is P. testudinea Klug.

Derivation of generic name. — From Greek, meaning ampoule or bulb-shaped

lamp; a vial for holy oil; evidently in allusion to the body form of an adult, which

resembles a rather broad bottle constricted toward the top. with a narrow opening,

corked by the head. The word Trachelizus is from Greek, meaning to wring or

twist, presumably in allusion to the seemingly marked constriction between head

and pronotum.

Recognition. — Adults are rufous or rufo-brunneous in color, with partially

darkened appendages. The elytra are inflated and broad in relation to length (Fig.

2). Antennomeres 5-10 (Fig. 10) are filiform, distinctly longer than wide. The legs

are flattened, especially the tibiae (Figs. 64A and B), and the tarsomeres are

relatively slender. These features render Physea the most easily recognized genus

in the Ozaenini.

Description. — Habitus as in Fig. 2. Size moderate, SBL ca. 10-12 mm., maximum width

4. 9-5. 2 mm.

Color. Body uniform rufous or rufo-brunneous; appendages same color, or mandibles,

antennomeres and tibiae darkened.

Microsculpture and luster. As for Ozaenini, with mesh pattern transverse on mesepimera,

isodiametric on mesepisterna, and isodiametric to slightly transverse on metepisterna. Dorsal

surface matte, ventral surface matte or pterothoracic and abdominal sterna slightly iridescent.

Vestiture. Dorsal surface setose or glabrous, mandibles with scrobes setose basally. Ventral

surface generally setose, or at least abdominal sternum IV with paramedial patches of setae, and

row of setae near posterior margin of each of sterna IV to VII. Male fore tarsomeres 1 and 2 with

adhesive vestiture ventrally.

Chaetotaxy. Clypeus with about 12 setae. Vertex with one pair of supraorbital setae and

transverse row of about four shorter setae. Antennae: antennomere 1 (scape) with several setae; 2

and 3 each with preapical ring of setae; antennomere 4 with scattered setae; antennomeres 5-11

generally setose except anterior and posterior glabrous areas. Mouthparts: labrum (Fig. 32) with

row of about 10 setae near anterior margin; mandibles without fixed setae; maxillary stipes

laterally with two setae; submentum, mentum, and glossal sclerite each with single pair of setae;

mentum without paramedial setae; labial palpomere 2 trisetose. Pronotum with marginal setae

numerous anteriorly and posteriorly, asetose medially. Each elytron with several rows of long

setae on disc. Legs (number of setae, fore, middle, and hind): coxae, O-numerous-numerous;

trochanters, several-several-one or two; femora, numerous-numerous-numerous.

Head. Frontal impressions not indicated. Eyes in lobate setose clefts, large. Supraantennal

area reflexed strikingly anteriorly, in form of broad plate each side (Fig. 2).

Antennae (Figs. 10, 18 and 25). Antennomeres 5-11 (Fig. 10) filiform, distinctly longer

than wide. Antennomere 11 (Fig. 18) with apical margin markedly asymmetrical, apex narrowly

rounded, without distinct ridge (Fig. 25).

Mouthparts. Labrum transverse (Fig. 32). Mandibles (Figs. 39A-D) falcate, most of occlusal

margin smooth, or details, see Table 1. Maxillae average for Ozaenini, lacinia (c/. Fig. 46)

terminated in long sharp tooth; palpomeres slender, elongate, apex of maxillary palpomere 4

subtruncate. Labium (Fig. 50): mentum with broadly rounded lateral lobes, each epilobe widened

preapically; tooth prominent; labial palpomere 3 elongate, slender, apex truncate.

Prothorax. Pronotum (Fig. 2) markedly transverse, sides markedly reflexed, lateral

margins smooth; lateral grooves broad; impressions shallow. Prosternum with intercoxal process

narrow.

Quae st. Ent., 1990, 26(1)

86

Ball and McCleve

Pterothorax. As described for Ozaenini, and intercoxal processes of meso- and

metasternum reduced, middle coxal cavities and middle coxae in contact medially. Metepistemum

overlapped by lobe of mesepimeron.

Elytra. Markedly expanded, convex. Surface smooth, flat, no indication of intervals and

interneurs. Basal ridge extended only to about base of interneur 6.

Legs. Middle coxae globose, more so than usual. Femora (Fig. 57) and tibiae (Figs. 64A

and B) markedly compressed, especially tibiae. Femora with ventral surfaces grooved, fore femur

(Fig. 57) with swelling ventrally, near base. Antennal cleaner of fore tibia (Figs. 64A and B)

small. For details, see Table 2.

Abdomen. Sclerites of segments II-VII unmodified, or description of sclerites of segments

VIII and IX-X, see under "Structural and biochemical features, genital segments". See also Figs.

94 and 100 A and B.

Male genitalia (Figs. 84A-D). For details, see this topic under "Structural and biochemical

features" and Table 3.

Ovipositor (Figs. 71A-C). ^tylomeres long and slender, blade-like, or details, see this

topic under "Structural and biochemical features".

Bursa copulatrix and spermatheca (Fig. 105). For details, see this topic under "Structural

and biochemical features”, and Table 4.

Defensive secretions. — Three compounds, only. See Table 5 for details.

Way of life. — Members of this genus seem to be associated with leaf-cutter

ants of the genus Atta. See under species treatments, below, for additional details.

Geographical distribution. — This genus is confined to the New World,

ranging on the mainland from Argentina to southwestern United States.

Relationships. — In body and leg form and structure of the antennal cleaner,

adults of Physea are much like those of the monobasic genus (1965a: 1 13). They

differ in form of antennomeres, however, those of Physeomorpha being very short

and transverse. In spite of this difference, we believe that these genera are closely

related, and may be congeneric.

Jeannel (1946: 47) placed Physea in a monobasic subfamily, because of its

obvious distinctiveness within the Ozaenini. Enhancing the distinctiveness of

body form and leg form are the distinctive mandibles, male genitalia, elongate

stylomeres of the ovipositor, bursa copulatrix, and bursal sclerite, and

myrmecophilous way of life. In other derived features (pterothoracic structure,

absence of the scrobal seta from the mandibles, trisetose labial palpomeres and a

complex set of defensive chemicals), adults of Physea are like other New World

ozaenines.

The filiform antennae seem to be a remarkably plesiotypic feature, for they are

more slender than the antennae of adult Metrius , the primitive sister-group of the

Ozaenini. This feature should not be over-emphasized, for two reasons. First, the

putative sister-genus of Physea exhibits more typical ozaenine antennae. Second,

the antennal cleaner is suggestive of a taxon whose members once had antennal

articles that were too extensive to be cleaned effectively by such a structure, and

accordingly the latter was reduced. Subsequently, then, the antennomeres became

slender, once more. Thus, their seeming primitiveness is secondary, and therefore,

these structures are apotypic.

At this time, we are not in position to offer an hypothesis that provides a

sister group for the lineage Physea + Physeomorpha.

Included species. — Six species are members of this genus, including two, P.

hirta LeConte and P. latipes Schaum, that live in Mexico (with the former also

occurring in Texas), and that are treated below.

Quae st. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

87

Physea hirta LeConte

Figs. 10, 18, 25, 32, 39A-D, 45, 50, 57, 64A-B, 71A-C, 84A-D, 94, and Map 1.

Physea hirta LeConte, 1853: 393. TYPE MATERIAL: HOLOTYPE male, labelled "Type 5488"

[red paper]; "Physea hirta Lee" [handwritten]; (MCZ). TYPE LOCALITY (from original

description): Mexico, near U.S. border (Haldeman). — Chaudoir, 1854: 312. — 1868: 72. —

Bates, 1881: 27.— Csiki, 1927: 431. — Banninger, 1927: 212. — Blackwelder, 1944: 23. —

Leng, 1920: 49. — Johnson, 1978: 67.

Recognition. — In habitus (Fig. 2), adults are similar only to those of P. latipes

Schaum Fig. 2). The two species are distinguished by setation (see key) and

pronotal form: elevated lateral portions much broader in P. hirta , and anterior

margin sharply concave.

Habitus as in Fig. 2. Characteristics of Physea. Standardized Body Length,

males 10.7-11.1 mm., females 10.5-11.7 mm; W/SBL males 0.44-0.45, females

0.44.

Way of life. — Adults were collected in Chiapas and Veracruz in and around

the midden heaps of Atta nests, in daylight hours. Van Emden (1936), in a

description of the larva of P. setosa Chaudoir, notes that both larvae and adults of

this species live in Atta nests. Specimens have been collected at night, also. The

late Jorge Hendrichs, of Mexico City, advised the senior author that he had

collected specimens of what was probably this species at night, in the vicinity of

columns of Atta workers.

Geographical distribution :*■ — This species is known from the Gulf and Pacific

Versants of Middle America, from Guatemala to southeastern Texas.

Material examined. — From Mexico and the United States, we have seen 20

specimens from the following localities.

UNITED STATES OF AMERICA. Texas. Comal Co. (USNM). Kennedy Co.

27°10'N, 97°40'W, VIII.28.1976; J.E. Gillaspy (TAIU).

MEXICO. Chiapas. E. slope. Sierra de la Colmena, 16° 24'18"N, 91 24T6"W, Arroyo

Santa Maria, 213 m., nr. Atta nest, VI. 5-10. 1972; G.E. & K.E. Ball, P.A. Meyer (UASM). Same,

VI. 1-10. 1972 (UASM). 27 km. SE Teopisca, Rte. 24, VI.3-4.1969; H.. Howden (UASM).

Oaxaca. Rte. 131, 82 km. S. Juchatengo, oak forest VII. 16-17. 1972; P.A. Meyer, G.E. Ball

(UASM). Hwy. 125, 13 km. N. Hwy. 200, nr. Pinotepa Nacional, U-V light, 195 m., VII. 19. 1986;

S. McCleve, P. Jump (UASM).San Luis Potosi. El Salto, VIII. 8-9. 1968; J.W. McReynolds

(CASC). Same, U-V light, VII.7.1966; R.E. Woodruff (UASM). Tamazunchale, VII.13.1956; D.H.

Janzen (ClSC).Veracruz. Fortin de las Flores, 1010 m., VIII. 1.1964; H.V. Daly (CISC). Same,

U-V light, VII. 7-12. 1974; J.A. Chemsak, J. Powell (CISC). Rio Metlac Cn., NW. Fortin de las

Flores, U-V light, VII. 10. 1974; J.A. Chemsak, E.& J. Linsley, & J. Powell (CISC). Canyon, SW.

Rio Metlac. nr. Fortin, 975-1036 m., ex refuse deposit Atta mexicana, VIII. 13-18, 1971, A.

Newton (MCZC).

Physea latipes Schaum

Figs. 2, 100A-B, 105, and Map 1

Physea latipes Schaum, 1864: 117. TYPE MATERIAL: not seen. TYPE AREA: "Mexico" (from

original description). Chaudoir, 1868: 74 . — Bates, 1881: 27 (as junior synonym of P. hirta

LeC.). — Csiki, 1927: 431 (as junior synonym of P. hirta LeC). — Banninger, 1927: 212. —

Blackwelder, 1944: 23.

Recognition. — See key and treatment of this topic for P. hirta.

Habitus as in Fig. 2. Standardized Body Length, males 10.3-1 1.4 mm., females

1 1.0-1 1.4 mm; W/SBL, males 0.44-0.46, females 0.45-0.46.

Female sternum VIII and tergum IX-X as in Figs. 100A and B, respectively.

Bursa copulatrix as in Fig. 105.

Ouaest. Ent., 1990, 26(1)

88

Ball and McCleve

Geographical distribution. — We have seen material from the Pacific Versant

of Mexico, only. However, we expect that Physea latipes ranges into southern

Arizona, as do many other carabids that occur in the vicinity of Mazatlan, Sinaloa.

This prediction is supported by occurrence in southern Arizona of the host of other

Physea species, Atta mexicana (Smith, 1951). .

Material examined. — We have seen 34 adults from the following localities.

MEXICO. Colima. Manzanillo, VII. 18. 1953; C.& P. Vaurie (AMNH). Guerrero. Iguala,

IX; Barrett (CASC). Jalisco. Ajijic, U-V light, VII. 25. 1964; W. L. Nutting (UASM). Estacion

Biologia Chamela, VII. 8-16. 1985; J. Chemsak et at. (CISC). 61 km. SW Guadalajara, 1310 m.,

VII. 24. 1952; F.W. & F.G. Werner (UASM). Hwy. 200, 33.5 km. S. Puerto Vallarta, 724 m„ U-V

light, VII. 21. 1986; S. McCleve, P. Jump (UASM). Sinaloa. Mazatlan, IX. 15. 1918 (CASC).

"Venedio" [=Venedillo], VII.10- VIII.27.1918 (CASC).

Pachyteles Perty

Pachyteles Perty, 1830: 3. TYPE SPECIES: Pachyteles striola Perty, 1830: 4; fixed by Hope,

1838: 99; subsequent designation. — Chevrolat, in d'Orbigny, 1847: 392. — Lacordaire,

1854: 157.— Chaudoir, 1868: 51.— Bates, 1881: 26.— Horn, 1881: 129.— LeConte and

Horn, 1883: 24.— Leng, 1920: 49.— Csiki, 1927: 427.— Banninger, 1927: 208.— van

Emden, 1942: 25. — Blackwelder, 1944: 23. — Ball, 1960: 94. — Erwin et al., 1977: 4.3. —

Reichardt, 1977: 376.— Eisner et al., 1977: 385.— Ward, 1979: 185.— Thompson, 1979:

232. — Erwin, 1979a: 359. 1979b: 557.- Erwin and Sims, 1984: 374, 427.

Goniotropis Gray, 1832: 274. TYPE SPECIES: Goniotropis braziliensis Gray, 1832: 274; by

monotypy. — Duponchel, in d'Orbigny, 1845: 274. — Lacordaire, 1854: 157. — Chaudoir,

1868: 51. — Bates, 1881: 25 . — Csiki, 1927: 427 (as junior synonym of Pachyteles) . —

Banninger, 1927: 202. — Blackwelder, 1944: 23. — Ball, 1960: 94. — Erwin et al., 1977:

4.3.— Reichardt, 1977: 377.— Moore, 1979: 194.— Erwin, 1979: 557.— Eisner and

Aneshansley, 1981: 83.

Tropopsis Sober, 1849: 179. TYPE SPECIES: Tropopsis marginicollis Sober, 1849: 181 (here

designated, the first of two species named by Sober). — Lacordaire, 1854: 159. — Chaudoir,

1868: 67 (as a section of Pachyteles). — Csiki, 1927: 427. — Banninger, 1927: 207. —

Blackwelder, 1944: 23.— Erwin, et al., 1977: 4.3.— Reichardt, 1977: 377.

Scythropasus Chaudoir, 1852: 289. TYPE SPECIES: Scythropasus elongatus Chaudoir, 1852:

289 (by monotypy). — 1868: 48. — Bates, 1881: 24. — Csiki, 1927: 427. — Banninger,

1927: 207.— Erwin, et al., 1977: 4.3.

Notes about synonymy. — Pachyteles , Goniotropis and Tropopsis are

treated as congeneric because the differences among them seem rather slight,

compared to differences among other New World genera. Certainly, the group as a

whole is markedly divergent, especially in features of the male genitalia and

ovipositor. However, such differences do not seem to be correlated with other

features. The name Scythropasus Chaudoir was synonymized with Goniotropis by

Banninger. The basis for selecting Pachyteles as the name for the genus is priority.

Derivation of generic name. — According to its author (Perty, 1830: 4), the

word Pachyteles is derived from incrassate antennomere 1 1 , and means thick spear

(Greek, pachy + telum ).

Recognition. — Among Middle American ozaenines, adults of Pachyteles are

recognized by a combination of: antennomeres 5-10 short, each about as wide as

long; fore femur with ventral spines (Figs. 59 and 60); antennal cleaner of fore tibia

with median expansion prominent; and base of mental tooth of labium with a pair of

setae. Adults of the South American subgenus Tropopsis lack the femoral spine, and

one undescribed species has long and slender antennal articles.

Description. — Habitus as in Figs. 3-6, body slender, elongate. Size varied, Standardized

Body Length ca. 3.5-17.5 mm, maximum width 1.3-5. 9 mm.

Color. Various somber shades: flavo-rufous to dark piceous, but not black; appendages of

most specimens rather paler than body color.

Middle American Genera of the Tribe Ozaenini

89

Microsculpture and luster. As for Ozaenini, and pterothoracic pleura and sterna with mesh

pattern isodiametric or transverse.

Vestiture. Dorsal surface various, from almost glabrous to densely setose, especially elytra.

Fore tarsomeres 1 and 2 or 1-3 of males with adhesive vestiture ventrally, or fore tarsomeres

glabrous.

Chaetotaxy. Clypeus with three pairs of setae. Vertex of head with several pairs of

supraorbital setae and several behind eyes, also. Antennae, number of setae: antennomere 1,

one to several; 2-3, apical ring; antennomere 4, ca. apical half setose; antennomeres 5-11 densely

setose, except anterior and posterior glabrous triangular areas. Mouthparts: labrum (Figs. 34 and

35) with 10 or more (ca. 16) setae near anterior margin; maxillary stipes with two or three setae;

submentum and mentum with one or more pairs of setae, each; glossal sclerite apically with one

pair setae; labial palpomere 2 with numerous setae, but three preapical setae longer than rest;

palpomere 3 also setose. Pronotum with marginal setae numerous (ca. 10). Each elytron with

several rows of discal setae of about 10 in each of intervals 1, 3, 5, and 7; umbilical setae ca. 25.

Legs (number of setae, fore, middle, hind): coxae, O-numerous-numerous; trochanters, several-

several-several, each with one long seta, others short; femora, numerous-numerous-numerous.

Head. Frontal impressions shallow, broad, but recognizable. Eyes (Figs. 3-6) subtruncate

posteriorly, moderately prominent; temporal lobes small. Supraantennal area not reflexed,

though ridges generally sharp.

Antennae (Figs. 12, 13, 20, 21, 27 and 28). Antennomeres 5-10 almost quadrate, 11

clearly longer than wide (Figs. 12 and 13), flattened antero-posteriorly, terete in cross section

(cf. Figs. 27 and 28). Antennomere 11 (Figs. 20 and 21) with apex broadly rounded, terminated

in straight keel (Figs. 27 and 28).

Mouthparts. Labrum transverse (Figs. 34 and 35). Mandibles (Figs. 41A-D) falcate, occlusal

margins with prominent teeth. For details, see Table 1. Maxillae average for Ozaenini, as in Fig.

46. Labium (Figs. 52 and 53): mentum with lateral lobes more or less pointed apically, each

epilobe widened preapically; tooth prominent; labial palpomere 3 narrowly securiform, apex

subtrijncate.

Prothorax. Pronotum (Figs. 3-6) transverse, distinctly wider than long, to distinctly

longitudinal, slender and slightly longer than wide. Lateral grooves moderately wide. Lateral

margins smooth to crenulate, sides posteriorly sinuate or not; antero- and posterolateral angles

projected or not. Disc moderately convex, impressions distinct but shallow.

Pterothorax. As described for Ozaenini and anterior margin of metepisternum near coxa

overlapped by posterior lobe of mesepimeron.

Elytra. Surface various: distinctly striate, indistinctly so, or smooth; intervals, if evident,

moderately to slightly convex, but broad and not carinulate. Basal ridge extended to about plane

of base of intervals 4 or 5.

Legs. In most features, as described for Ozaenini, no parts remarkably compressed. Fore

femur (Figs. 59 and 60) with dentiform projection ventrally near base (subgenera Pachyteles and

Goniotropis ), or without such projection (subgenus Tropopsis). Antennal cleaner of fore tibia

various. See Figs. 66A and B, and 67A and B. See Table 2 for details.

Abdomen. Sclerites of segments II- VII unmodified, or sclerites of segments VIII and IX/X,

see Figs. 95, 101, and 102. or details, see under "Structural and biochemical features, genital

segments".

Male genitalia (Figs. 85-89, and 97). For details, see Table 3.

Ovipositor (Figs. 74-79). Stylomeres various, from long and slender with narrow apex to

short, rather broad, and with apex bifurcate.

Bursa copulatrix and spermatheca (Fig. 106). For details, see Table 4 .

Defensive secretions. — Four compounds. See Table 5 for details.

Way of life. — Adults of Pachyteles are probably insect predators, most of

them living under bark of fallen tree trunks, and probably flying at night. At the

northwestern periphery of the range, a few individuals have been collected in

agricultural fields, but it is not clear from the labels whether they were collected at

light, at night, or if they were found on the ground during the day. Adults of other

species have been collected at night, in association with dead oaks.

Quae st. Ent., 1990, 26(1)

90

Ball and McCleve

Geographical distribution. — The range of this genus extends through the

forested parts of the Neotropical Region from southern Chile to Mexico, and into

southwestern United States, in the Nearctic Region.

Relationships. — In external features, adults of Pachyteles resemble most

closely those of the Afrotropical genus Afrozaena Jeannel. This apparent

similarity, however, is not borne out by detailed study of structural features.

Pachyteles is without close extant relatives in the New World, also..

Included taxa. — We recognize three subgenera, two of which [Goniotropis

and Pachyteles {sensu stricto )] occur in southwestern United States, and are treated

further below. Included in subgenus Goniotropis is Pachyteles elongatus

(Chaudoir), the type species of Scythropasus Chaudoir. The subgenus Tropopsis

is confined to South America, and is not considered further.

Subgenus Goniotropis Gray

Derivation of subgeneric name. — In the original description of the type

species, G. braziliensis Gray, the author notes (1832: 274) that the anterior femora

and tibiae are strongly toothed and that each elytron terminates in an apical hook.

The name is derived from Greek gonio , meaning angle, and tropis, meaning keel. We

believe the word refers to the flanges of Coanda, which are keel-like, and are

located posteriorly on the outer angles on each elytron.

Recognition. — Adults of Goniotropis have the middle coxal cavities closed

medially, with the intercoxal process of meso- and metastema in contact, and the

antennal cleaner of the fore tibia (Figs. 67A and B) with a prominent projection

extended from the medial expansion. A more detailed characterization is not

required here.

Way of life. — Adults of a few species of Goniotropis have been collected

under the bark of fallen tree trunks, and one was found in a bromeliad growing on

the trunk of a standing pine tree, about 5 meters above the ground. Most specimens

have been taken at U-V light, at night.

Geographical distribution. — The range of Goniotropis includes the northern

half of South America, all of Middle America, and southern Arizona in southwestern

United States.

Relationships. — We hypothesize Goniotropis to be sister group of subgenus

Pachyteles, with their common ancestor being the sister group of subgenus

Tropopsis. This hypothesis is based on transformation series for the intercoxal

processes, armature of the fore tibia and structure of the antennal cleaner.

Included species. — Banninger (1927: 203-204) included 14 species in this

group. Two species occur in Arizona, in southwestern United States. We describe

these because they are in the study area, and as well, the tropical Mexican P.

elongatus (Chaudoir) because it is the type species of Scythropasus, a junior

synonym of Goniotropis.

Pachyteles parca LeConte

Figs. 3, 53, 78A-C, 85A-E, 101, 106, and Map 1

Pachyteles parca LeConte, 1884: 2. HOLOTYPE female, in LeConte-Horn Collection, labelled:

"Ari."; "Type 5487" [red paper]; "Pachyteles parca LeC" [handwritten] (MCZC). TYPE

AREA: Arizona, U.S.A.— Horn, 1894: 308.— Leng, 1920: 49.— Csiki, 1927: 430.

Goniotropis parca-, Banninger, 1927: 204. — Ball, 1960: 94. — Erwin, et al., 1977: 4.3.

-

Quaest. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

91

Pachyteles beyeri Notman, 1919: 225. HOLOTYPE male, labelled: "San Felipe Low Cal";

" Pachyteles beyeri TYPE" [handwritten, on blue paper] (Staten Island Museum, New

York). NEW SYNONYMY.— Csiki, 1927: 204.

Goniotropis beyeri Banninger, 1927: 204. — Ball, 1960: 94. — Erwin, et alii., 1977: 4.3.

Notes about synonymy. — In form of median lobe and details of the internal

sac, the male genitalia of the type of P. beyeri match those of males of P. parca,

collected in Madera Canyon, Santa Rita Mountains, Arizona. In other features,

specimens of the two nominal species are identical, also. We have no doubt that

they are conspecific.

Recognition. — Adults of this species are slender, with transverse pronotum

with sinuate lateral margins (Fig. 3), and of moderate size (Standardized Body

Length less than 12 mm). In general form, they look like small specimens of P.

kuntzeni (Banninger):males differ from those of the latter species in form of the

apical portion of the median lobe (Figs. 85A-B; cf Figs. 86A-B; also, cf Figs. 87 A-

B).

Description. — Habitus as in Fig. 3, with character states of Goniotropis and: SBL males,

9.9-11.4 mm., females 8.8-11.2 mm.; W/SBL males 0.31-0.32, females 0.33.

Male genitalia (Figs. 85A-D). Median lobe in lateral aspect with apical portion with carinula

(£.) on left side, apex nearly truncate; internal sac with small apical brush (ab) and digitus (d.),

terminal lobe small. Left paramere (Fig. 85D) with patch of setae preapically; right paramere

(Fig. 85E) with about half medial margin setose.

Ovipositor. Stylomeres each as in P. kuntzeni (cf. Figs. 77A-C), cylindrical, straight, apex

broad and circular, not tapered to point, sensory furrow nearly apical; surface with irregular rows

of thick basiconic and slender trichoid sensilla, dense cluster of these near apex.

Bursa copulatrix. As in Fig. 106.

Way of life. — All known specimens have been collected at night, at light,

principally at ultra-violet light, in the vicinity of oak-pine forests. Months of

collection are from June to September.

Relationships. — Based on marked similarity in structural features and on

evidently parapatric distribution pattern we hypothesize that this species and P.

kuntzeni are sister taxa.

Geographical distribution (Map 1). — This species is known from

northwestern Mexico (Baja California and northern Sonora) western Durango, and

southern Arizona, in southwestern United States. The Durango specimen, a male,

was determined by Banninger, in 1926, as "Goniotropis sp?"

Material examined. — In addition to the types, we have seen 29 specimens

from the following localities.

UNITED STATES OF AMERICA. Arizona. Cochise County. — Chiricahua Mts., Portal,

VII. 15. 1968; D.J. & J.N. Knull (OSUC). Guadalupe Canyon, at light, VII. 1.1975; S. McCleve, M.

Topham (SMCC). Same, at light, VII.3 1.1975; S. McCleve (SMCC). Huachuca Mts., VIII. 18.1936;

J.N. Knull (OSUC). Same, Miller Canyon, 1524 m„ VIII. 17. 1974; E. R. Hoebeke (CUIC).

Peloncillo Mts., 53 km. E. Douglas, at light, VII. 17. 1973; S. McCleve (SMCC). Graham County. —

Galiuro Mts., Aravaipa Canyon, 17.7 km. N. Klondyke, U-V light, VII. 24. 1976; G.E. Ball, J.M.

Campbell, P.M. Hammond (UASM). Same, on Turkey Creek, VI. 22. 1976; S. McCleve (SMCC).

Pima County. — Santa Catalina Mts., Molino Basin, VII. 31. 1974; D.M. Bright (CNCI). Same,

VIII. 8. 1969 (FSCA). Santa Cruz County. — Pajarito Mts., Pena Blanca Canyon, 1191 m., U-V

light, VII. 2. 1980; S. McCleve (SMCC). Same, VII,13,1968 (FSCA). Same, VII.16.1964; R.H.

Arnett, Jr. & E. R. Van Tassell (SCA). Same, VII. 13. 1970; K. Stephan (SCA). Same, VIII.7.1959;

R.H. Arnett, Jr. (FSCA). Same, U-V light, VIII. 1 1.1968; G.E. Ball family & R.B. Madge (UASM).

Same, U-V light, VII.28.1978; S. McCleve (SMCC). 4.3km. above Pena Blanca, VIII. 12. 1983; E.

Riley (EGRC). Santa Rita Mts., Madera Canyon, VIII. 8. 1977; .T. Hovore (SCA). Same, 1490 m.,

VIII. 23. 1959; J.G. Franclemont (CUIC). Same, VII.20.1959 (CUIC).Same, VIII. 1 1-24.1963; G.E.

and K.E. Ball (UASM).

Quaest. Ent., 1990, 26(1)

92

Ball and McCleve

MEXICO. Baja California Sur. 3 km. E. La Burrera, 515 m., IX. 2. 1977 (CASC). San

Jose del Cabo [Horn, 1894: 308. Sierra El Chinche [Horn, 1894: 308]. Durango. Canelas; J.

Flohr [ZMHB], Sonora. 55 km. SW Moctezuma, 1066 m„ VI. 10. 1982; S. McCleve (SMCC).

Rte. 16, 32.3 km. E. Rio Yaqui, U-V light, VII.26-27.1987; S. McCleve (UASM). Sierra Alamos,

2.7 km. S. 1.9 km. W. Alamos, U-V light; S. McCleve (UASM).

Pachyteles kuntzeni (Banninger) NEW COMBINATION

Figs. 4, 13, 21, 28, 35, 46, 60, 67A-B, 77A-C, 86A-E, and Map 1

Goniotropis kuntzeni Banninger, 1927: 204. HOLOTYPE female, labelled: Canelas, Durango

Mexico Flohr (Banninger Coll., ETHZ) TYPE LOCALITY: Mexico, Durango, Canelas. —

Blackwelder, 1944: 23. — Erwin, et at., 1977: 4.3. — Erwin and Halpem, 1978: 360.

Notes about type material. — We have not seen the holotype, but we have seen

a male paratype in ZMHB from the type locality, and our material (from Sonora and

southern Arizona) matches the features, including those of the male genitalia, of that

specimen. Accordingly, we are confident that our identification is correct.

Recognition. — Adults of this species are large (SBL 16.0- 17.6 mm), the

largest known of Pachyteles , with transverse pronotum with sinuate lateral margins

(Fig. 4). They are much like large specimens of P. parca. Males of the two species

are similar in details of the genitalia, but in males of P. kuntzeni , the apex of the

median lobe (Fig. 86A) is obliquely truncate, or additional details, see the

description, below. In shape of the stylomeres of the ovipositor, females of P .

kuntzeni and P. parca are similar, but those of P. kuntzeni have more setae (Figs.

77A-C; cf Figs. 78A-C).

Description. — Habitus as in Fig. 4. With character states of subgenus Goniotropis and

Standardized Body Length males 16.0-16.9 mm., females 15.7-17.2 mm; W/SBL males 0.33-

0.35, females 0.32-0.33.

Male genitalia (Figs. 86A-E). Median lobe in left lateral aspect (Fig. 86A) with apical

portion prominent, rather broad, carinulate on left side, apex obliquely truncate. Internal sac with

large apical brush (ab), pendent terminal lobe (t) with large digitus (d). Left paramere (Fig.

86D) with few setae preapically along medial margin. Right paramere (Fig. 86E) with extensive

brush of setae for most of length of medial margin.

Ovipositor (Figs. 77A-C). Stylomeres of moderate length, cylindrical, with broad circular

apex, not tapered to point, and sensory groove preapical. Vestiture moderately dense, of longer

sensilla trichodea and thicker sensilla basiconica.

Bursa copulatrix. About same as that of P. parca (above; cf Fig. 106).

Way of life. — Specimens have been collected in association with dead oak

trees, at night.

Geographical distribution (Map 1). — This species is known only from

northwestern Mexico (Durango and Sonora) and southernmost United States

(Arizona).

Relationships. — This species is postulated to be the sister taxon of P. parca.

Material examined. — In addition to the male paratype, we have seen 13

specimens, from the following localities.

UNITED STATES OF AMERICA. . Arizona. Cochise County. — Huachuca Mts., Ash

Canyon, 1548m., VIII. 2. 1979; N. McFarland (SMCC). Maricopa County. — Tempe, XI. 28. 1966; T.

Paca (ASUT). Santa Cruz County. — Atascosa Mts., Sycamore Canyon, on oak stump at night,

VII. 12. 1977; S. McCleve (SMCC). Pajarito Mts., Walker Canyon, 1191 m., on dead oak, at night,

VII. 28. 1978. S. McCleve (SMCC). Pajarito Mts., Pena Blanca, 1219 m., U-V light; G.E.&K.E. Ball,

& R.B. Madge (UASM). County not known. — S. Graham Mts., 1524 m., VIII. 20. 1974; K. Stephan

(FSCA).

MEXICO. Sonora. 16.1 km. E. Cananea, VIII. 16. 1949; G.M. Bradt (AMNH).

Quaest. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

93

Pachyteles elongatus (Chaudoir)

Figs. 79A-C, 87A-D

Scythropasus elongatus Chaudoir, 1854: 295. TYPE MATERIAL: HOLOTYPE male, in Chaudoir-

Oberthiir Collection Box 132, labelled "Mexique"; "Ex Musaeo Mniszech"; "Elongatus

Chaud." [handwritten, not in Chaudoir's hand] (MNHP). — 1868: 48. — Bates, 1881: 24. —

Csiki, 1927: 427.

Goniotropis elongatus ; Banninger, 1927: 203. — Blackwelder, 1944: 23.- — Erwin, et al., 1977:

4.3.

Notes about type material. — The specimen indicated above as holotype is in

a series with a female labelled "Temax, N. Yucatan Gaumer" (Bates Coll) and a male

labelled "Yucatan (Bates Coll)", and is not labelled as type. However, Chaudoir

(1868: 48) recorded that the single specimen on which the name was based was

collected in Mexico and was in the Mniszech collection. The specimen labelled as

holotype fits these conditions.

Recognition. — The slender body and pronotum as long as or slightly longer

than wide distinguishes adults of this species from other Mexican members of

Pachyteles ( sensu lato). Males have adhesive vestiture on fore tarsomeres 1-3,

rather than on 1-2 only, as in most other species of Pachyteles , and the apical

portion of the median lobe (Fig. 87A) is distinctive ( cf Figs. 85A and 86A). The

long slender stylomeres of the ovipositor (Figs. 79A-C) are distinctive for females.

Description. — With character states of subgenus Goniotropis and form slender,

pronotum as long or longer than wide. Standardized Body Length males 10.6-11.0 mm., females

9.6-11.6 mm.; W/SBL males 0.30-0.32, females 0.32-0.33.

Male genitalia (Figs. 87A-D). Median lobe in lateral aspect (Fig. 87A) with apical portion

broadly rounded, prominent. Internal sac (Figs. 87A-B) with collar area covered with slender

microtrichia; without apical brush; terminal sclerite truncate, not lobed. Parameres (Figs. 87C-D):

left paramere with narrowed apex, with few setae preapically on medial margin; right paramere

elongate, medial margin extensively setose, most densely so preapically.

Ovipositor (Figs. 79A-C). Stylomeres long and slender, each with two or more nematiform

setae (Fig. 79C), surfaces with few trichoid setae in median area, more densely setose preapically

and apically.

Bursa copulatrix. Not studied.

Way of life. — Specimens known to us were collected at ultra-violet light, at

night, in or in the vicinity of lowland tropical forest.

Geographical distribution. — This species is in the northern part of the

Neotropical region, ranging in Central America from Nicaragua to Tamaulipas on

the Gulf Versant, and to Nayarit on the Pacific Versant.

Relationships. — The distinctive body form, form of apex of median lobe,

armature of the internal sac, and slender stylomeres of the ovipositor indicate that

this species is not very close to P. parca and P. kuntzeni, the only other known

species of Goniotropis in Mexico.

Material examined. — In addition to the type and Yucatan specimens in the

Chaudoir-Oberthiir collection noted above, we have seen 10 specimens from

localities in Mexico.

Chiapas.. Palenque ruins, 122 m., U-V light, VI. 8. 1966; G.E. Ball & D.R. Whitehead

(UASM). Jalisco. 24 km. S. Tomatlan, lowland 2nd growth forest and pasture, 110 m., at U-V

light, VII. 11.1984; S. McCleve & P. Jump (UASM). Nayarit.. 58 km. SW Las Piedras lowland

forest 118 m, at U-V light, VII. 7-8. 1984; S. McCleve & P. Jump (UASM). San Luis Potosi..

Palitla, VIII. 5. 1966; O.S. lint (USNM). Tamaulipas.. ca. 40 km. N. Ciudad Monte Nacimiento, at

light, VII.31.1970; C.W. O'Brien (UASM). Veracruz. Lake Catemaco, U-V light, VII. 10-

18.1963; D.R. Whitehead (UASM). Same, VI.9-25.1969; H. F. Howden (UASM). Same, Coyame,

U-V light, VII. 5. 1967; R.E. Woodruff (UASM). Los Tuxtlas Biological Station, ca. 30 km. E.

Catemaco, ca. 30 m., VI. 29-30. 1983; R.S. Anderson (UASM).

Quaest. Ent., 1990, 26(1)

94

Ball and McCleve

Subgenus Pachyteles ( sensu stricto)

Recognition. — See this topic above, for subgenus Goniotropis. Habitus is

illustrated by Figs. 5 and 6. These forms are rather similar to one another, and

though other members of the subgenus look like them generally, some have

strikingly different pronota, and differ as well in punctation of the dorsal surface

and striation pattern of the elytra.

Way of life. — See the general statement under the genus, and details in the

following species treatments.

Geographical distribution. — The range of Pachyteles ( sensu stricto ) is

coextensive with the range of the genus.

Relationships. — See this topic under Goniotropis and Pachyteles (s. lat.).

Included species .— According to Banninger (1927: 210-212), 48 species are

included in Pachyteles (5. str.), arranged in several groups. We treat only three here,

including one that is new.

Pachyteles gyllenhali (Dejean)

Figs. 5, 74A-B, and 88A-D

Ozaena gyllenhali Dejean, 1825: 436. TYPE MATERIAL: in Chaudoir-Oberthiir Collection, Box

132, in front of the following box label — "Gyllenhali Dej. Antilles? C. Dejean."

HOLOTYPE female, labelled: "Gyllenhali m. in Amer. inf." [green paper]; "Gyllenhali"

[green paper]; "Ex Musaeo Chaudoir" [red print on white paper], — Chaudoir, 1854: 301.

Pachyteles gyllenhali-. Chaudoir, 1868: 55. — Csiki, 1927: 428. — Banninger, 1927: 211. —

Blackwelder, 1944: 23. — Erwin, et alii., 1977: 4.3. — Erwin and Sims, 1984: 427.

Ozaena verticalis Chaudoir, 1848: 104. TYPE MATERIAL: two females — first labelled "Ex

Musaeo Chaudoir" [red print on white paper]; second, "Goudet" [green paper], "Ex

Musaeo Chaudoir" [red print on white paper] — in Chaudoir-Oberthiir Collection, Box 132,

in front of the following box label: "verticalis Chaud. Colombie Duport". LECTOTYPE

(here selected) first female in series (MNHP).— 1854: 301. NEW SYNONYMY.

Pachyteles verticalis-, Chaudoir, 1868: 56. — Bates, 1881: 25. — Csiki, 1927: 428. — Banninger,

1927: 210. — Blackwelder, 1944: 23.

Pachyteles testaceus Horn, 1868: 129. LECTOTYPE female, in LeConte-Horn Coll., labelled:

"Ariz"; TYPE NO. 1029.1 [red paper]; "Pachyteles testaceus Horn" (MCZC). NEW

SYNONYMY. TYPE LOCALITY: Fort Grant, Arizona.— Horn, 1881: 128.— LeConte and

Horn, 1883: 24. — Horn, 1894: 308. — Leng, 1920: 49. — Blackwelder, 1944: 23. — Ball,

1960: 94.— Erwin, et al, 1977: 4.3.

Notes about synonymy. — The type specimens of the three nominal taxa noted

above are very similar in size and external features. Horn (1868: 130) noted the

marked similarity among them, but based on the limited material at his disposal he

concluded that three species were represented. With the more extensive material

available to us, especially of the nominal mainland taxa, we are unable to distinguish

among them.

Recognition. — A combination of small body size (Standardized Body

Length ca. 3.5- 6 mm.), densely setose dorsal surface, and pale (rufo-flavous)

integument distinguishes adults of this species from other North-Middle American

species of Pachyteles. Males lack adhesive setae of the fore tarsomeres, and the

apical portion of the median lobe (Fig. 88A) is small, narrow and pointed. Females

have the stylomere of the ovipositor with a bifurcate apex (Fig. 74B).

Description. — Habitus of adults as in Fig. 5, with features of subgenus Pachyteles. SB L

males 4. 0-5. 4 mm., females 3. 6-4. 6 mm; W/SBL males 0.40, females 0.37-0.39.

Color. Integument pale, especially specimens from Arizona and northwestern Mexico, but

many from elsewhere with irregular fuscous cloud medially on elytra.

Quaest. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

95

Microsculpture. Average for genus.

Vestiture. Dorsal surface of head, pronotum and elytra densely setose. Males without

adhesive vestiture on fore tarsomeres.

Head. Average for Ozaenini.

Pronotum (Fig. 5). Subrotund, transverse; lateral margins sinuate posteriorly. Anterior

angles acute; posterior angles rectangular.

Elytra. Moderately densely punctate, interneurs very shallow to obsolescent, intervals

nearly flat.

Legs. Average for subgenus.

Male genitalia (Figs. 88A-D). Median lobe (Figs. 88A-B) with apical portion small, apex

pointed. Internal sac with basal rod (r; cf. Fig. 82A) long, terminal lobe (t) penis-like, without

preapical lobe, dorsal lobe, or microspines. Left paramere asetose (Fig. 88C); right paramere

with single seta on medial margin.

Ovipositor (Figs. 74A-B). Stylomeres triangular in outline, each with base broad , cupped

medially (Fig. 74A), somewhat flattened, apex bifurcate (Fig. 74B); sensory groove at base of

furcation, preapical in position. Setation rather sparse, principally of sensilla basiconica.

Way of life. — Adults of this species have been collected on dead tree trunks,

under bark, and in saguaro cactus. Many specimens have been collected at night, at

light, and particularly ultra violet light. The wide range of this species, including

islandic localities in both the Caribbean Sea and Pacific ocean, indicates that adults

disperse readily.

Geographical distribution. — The range of this species is extensive, including

the Greater Antillean island of Cuba, the Tres Marias Islands off the Pacific Coast

of Mexico, and on the mainland, from Brazil in South America (Blackwelder, 1944:

23), throughout Middle America to southern Arizona.

Material examined. — We have seen 193 specimens of this species, from the

following localities in Cuba, the United States, and Mexico.

UNITED STATES OF AMERICA. Arizona. Cochise County. — Guadalupe Canyon, U-V

light, VII. 31. 1975; S. McCleve (SMCC). Same, VIII. 2.1977 (SMCC). Gila County.— Globe,

VII. 1927; D.K. Duncan (MCZC). Wheatfield, near Globe, 11.25.1932; . Parker, D.K. Duncan

(MCZC). Same (CASC). Graham County. — Galiuro Mts., Aravaipa Canyon, east end, U-V light,

VII. 24-25. 1974; S. McCleve (SMCC). Same (AMNH). Aravaipa Canyon, U-V light, VIII. 1 2. 1 975;

S. McCleve (SMCC). Same, Turkey Creek, 1.6 km. S. Aravaipa Creek, U-V light, VIII. 1 1.1975; S.

McCleve (SMCC). Same, ca. 16 km. NW Klondyke, 900 m„ U-V light, VII.30-3 1.1975; G. E. Ball,

H.E. Frania (UASM). Same, U-V light, VII. 24. 1976; G.E. Ball, J.M. Campbell, P.M. Hammond

(UASM). Same, U-V light, VIII.24. 1 977; G.E. & K.E. Ball (UASM). Arivaipa, VIII. 29. 1933;

Bryant (CASC). Safford, under bark of^cottonwood, beside creek, 1.10.1938; O. Bryant (CASC).

Maricopa County. — Phoenix, XI. 27. 1919; E. Schiffel (MCZC). Pima County. — Ajo Mts., Alamo

Canyon, in rotting saguaro cactus, VII. 24; H.B. Leech (CASC). Same, VII. 25; H.B. Leech, J.W.

Green (CASC). Arivaca, XI. 30. 1969; K. Stephan (FSCA). Same, Arivaca Creek, VII. 31. 1952;

H.B. Leech, J.W. Green (CASC). Baboquivari Mts., W. side, Baboquivari Canyon, VII. 25-

27.1952; H.B. Leech, J.W. Green (CASC). Redington, XII. 7. 1969; K. Stephan (FSCA). Tucson,

VII. 21. 1917 (CUIC). Same, San Xavier Mission, VII. 29. 1924; E.P. Van Duzee (CASC). Pinal

County. — Florence (Fall, MCZ). Santa Cruz County. — Cobabi Mts., Santa Cruz Village,

X. 12. 1916 (USNM). Pajarito Mts., Pena Blanca Canyon, Lot 511, VII. 26. 1961; R.H. Arnett, E. Van

Tassell (FSCA). Patagonia, VII.36; E.S. Ross (CASC). Same, VIII.9.1940; E.S. Ross (CASC).

Same, VII. 18. 1948; C. & P. Vaurie (AMNH). Same, IX. 28. 1968 (SCA). County not known. —

Galiuro Mts.; Hubbard and Schwarz (FSCA, USNM). Arizona, Charles Palm (AMNH).

CUBA. Cayamas, 1.14; E.A. Schwarz (MCZC).

MEXICO. Baja California. Pelican Island, VII. 5. 1921; J.C. Chamberlin (CASC). Baja

California Norte. 14.5 km. SE Rancho Laguna, VII. 1.1973; Fisher, Westcott (CASC). Baja

California Sur. El Sargento, VII. 29. 1971; H.G. Real, R.E. Main (CASC). Miraflores,

VIII. 7. 1971; H.G. Real, R.E. Main (CASC). 8 km. S. Miraflores, VII. 10. 1938; Michelbacher, Ross

(CASC). 24 km. E. San Jorge, VII.24.1971; H.G. Real, R.E. Main (CASC). San Jose Island,

V. 28. 1921; E.P. Van Duzee (CASC). Santa Rosa (Fall, MCZC). [Sierra] El Chinche, 609 m„

under stones (CASC). Chiapas. Cerro Baul, ridge SE of, 21 km. W Rizo del Oro, 1615 m..

Quae st. Ent., 1990, 26(1)

96

Ball and McCleve

!

cloud forest, IX. 6-8. 1970; C. Mullinex, D.E. Breedlove (CASC). 48 km. NW Ocosingo,

VI. 30. 1977; E.M. Fisher (CASC). Sierra de la Colmena, e. slope, nr. La Caverna, Arroyo Santa

Maria, 16°24'18"N, 91°24'16''W, 213 m„ on ground, VI. 1-10.1972; G.E.& K.E. Ball, P.A.

Meyer (UASM). Sierra de la Colmena, San Cristobal trail, 701-853 m., VI. 8. 1972; P.A. Meyer,

G.E.& K.E. Ball (UASM). 19 km. S. Solosuchiapa, Rte. 195, 640 m„ IV.25.1966; G.E. Ball, D.R.

Whitehead (UASM). Colima. 11.3 km. NE Colima, XII.3.1948; E.S. Ross (CASC). Mt. Colima,

SE slope, XII. 2. 1948; H.B. Leech (CASC). Jalisco, nr. Ixtapa, ca. 30 m., gallery forest, dead

tree; XII.22.1970; G.E.& K.E. Ball (UASM). 15 km. S. Mazamitla, 1676 m„ VII.29-3 1.1952; F.W.

& F.G. Werner (UASM). 20 km. S Tecalitlan, 1615 m„ VIII.3.1967; Ball, T.L. Erwin, R.E. Leech

(UASM). Nayarit. Jesus Maria, VII.6.1955; B. Malkin (CASC). San Bias, VI. 15. 1955; B. Malkin

(CASC). Islas Tres Marias, Madre Maria Island, Arroyo Hondo, V. 17. 1925; H.H. Keifer (CASC).

Sinaloa. 8 km. N. Mazatlan, U-V light; J.A. Chemsak (CISC). Sonora. 8 km. E. Alamos,

VIII. 1 1.1973; K. Stephan (FSCA). 21 km. SE. Alamos, X. 30. 1972; K. Stephan (FSCA). Bahia

Kino, X. 25. 1980; P. Jump (SMCC). Sierra San Luis, Varela Ranch, Canon Bonita, U-V light, and

under bark of cottonwood logs; G.E. Ball & D.R. Maddison (UASM). Tabasco. 96 km. SE.

Villahermosa, U-V light, VI. 6-7. 1972; P.A. Meyer, G.E.& K.E. Ball (UASM). Veracruz.

Atoyac, VI. 24. 1982; M.A. Ivie (OSUC). Cordoba; D.A. Fenyes (CASC). Fortin de las Flores,

VI. 20-30. 1963; D.R. Whitehead (UASM). park canon, 3.2 km. W. Fortin de las Flores, Rte. 150,

VIII. 3-6. 1965; Cornell Univ. Mexico Field Party, 1965 (CUIC). 56 km. SE. Jalapa, XII.26.1963;

C.A.& M.J. Tauber (CUIC). Sierra de las Tuxtlas, Lake Catemaco, Coyame, under bark, VII. 1 -

15.1963; D.R. Whitehead (UASM). Same, under bark, VII. 1 0- 1 8. 1 963; D.R. Whitehead (UASM).

0.5 km. W. Sontecomapan, 305 m., IX.20&26. 1965; G.E. Ball, D.R. Whitehead (UASM). Same,

IX. 18-26. 1965 (UASM). 4 km. W. Sontecomapan, IV.3&10.1966; G.E. Ball, D.R. Whitehead

(UASM). Same, on log, VI. 1-5 & 20.1966 (UASM). Same, U-V light (UASM).

Pachyteles enischnus, new species

Figs. 6, 75A-B, 97A-D, and Map 1

Type material. — HOLOTYPE male, labelled: "MEX. Jalisco nr. Ixtapa gallery forest dead

tree ca. 100' [elevation above sea level] XII. 22. 70"; "Puerto Vallarta MEX trip 1970 G.E.&K.E.

Ball collectors" (USNM). ALLOTYPE female, "15 km.S. Mazamitla Jal. MEX. 5500' pine-oak

forest July 30, 1952 FE & FG Werner" (USNM). Two male PARATYPES (USNM) labelled

same as holotype, and one male PARATYPE (UASM) labelled same as allotype. Eleven

additional PARATYPES, labelled as follows. Male and female, "MEXICO Jalisco 33.8 km. S.

Puerto Vallarta pine-oak forest 750 m., at U-V light 9, 10. VII. 1984 S. McCleve, P. Jump"

(UASM). our males, three females, "MEXICO Nayarit 57.9 km. s.w. Las Piedras lowland forest

118 m„ at U-V light 7-8.VII.1984 S. McCleve, P. Jump" (UASM). Male, "Sin. Mex. 200 ft. 5-3-

49"; "GM Bradt Collector" (AMNH). Female, "5 mi. E. Alamos SONORA MEX VIII. 11. 1973 K.

Stephan & D.S. Chandler" (OSUC).

Derivation of specific epithet. — This is based on the Greek adjective

enischnos , meaning thin, in allusion to the slender adult body form (Fig. 6).

Recognition. — Body size, reduced setation of the dorsal surface, and

pronotum with markedly sinuate lateral margins distinguish this species from other

members of the genus that range into northern Mexico and southwestern United

States.

Description . — Habitus as in Fig. 6. Standardized Body Length of males 7. 2-7. 6 mm.,

females 7. 0-7. 7 mm.; W/SBL males 0.34-0.35, females 0.33-0.35.

Color. Body rufous to rufo-piceous, legs, antennae and palpi slightly paler than body.

Microsculpture and luster. Head dorsum with mesh pattern isodiametric, microlines fine,

nearly effaced on center of vertex. Pronotum with mesh pattern transverse, partly effaced on disc.

Elytra with mesh pattern isodiametric in lateral channels, transverse and partly effaced on disc.

Dorsal surface generally shiny.

Vestiture. Dorsal surface of head and pronotum sparsely setose. Elytra with discal intervals

serially, sparsely setose. Thoracic and abdominal sterna moderately densely setose.

Head (Fig. 6). Eyes average in size and convexity. Frontal impressions broad, irregular in

outline, irregularly punctate. Vertex coarsely, sparsely punctate.

Quae st. Ent., 1990, 26(1)

*

Middle American Genera of the Tribe Ozaenini

97

Pronotum (Fig. 6). Transverse, surface sparsely, irregularly punctate. Anterior and

posterior margins nearly straight. Lateral margins markedly sinuate posteriorly. Postero-lateral

angles about rectangular. Disc slightly convex. Lateral grooves broad anteriorly and posteriorly,

narrow medially, margins not beaded. Postero-lateral impressions broad and irregular,

continuous anteriorly with lateral grooves.

Elytra. In form, average for Pachyteles, humeri broadly rounded, slightly prominent.

Interneurs shallow, rather broad, intervals only slightly convex. Intervals sparsely punctate.

Metathoracic wings. Macropterous, fully developed.

Legs. Average for subgenus Pachyteles.

Male genitalia (Figs. 97A-C). Median lobe (Fig. 97A) with apical portion distinct, extended

ventrad, apex broadly rounded, nearly subtruncate. Internal sac with collar area with dense

covering of slender seta-like microtrichia; apical portion with broad sclerite on left side,

terminated in obtusely pointed lobe; basal part of sac with distinct longitudinally oriented ridges

and several setae near apex of median lobe. Left paramere (Fig. 97B) shorter than right

paramere (Fig. 97C), glabrous. Right paramere digitate, sparsely setose apically and on apical

part of medial margin.

Ovipositor (Figs. 75A-B). Stylomeres falcate, each with apex broadly pointed, not bifurcate.

Sensory groove ventral, remote from apex. Lateral and ventral surfaces with numerous thick

sensilla basiconica (Fig. 75A).

Way of life. — Adults were collected from under bark of dead trees in gallery

forest bordering deciduous tropical forest, near sea level, to pine-oak forest at ca.

1500 m. Most specimens, however, were taken at light, at night, indicating

nocturnal flight activity during the rainy season in northwestern Mexico.

Geographical distribution. — This species is known only from western

Mexico, from Jalisco to Sonora.

Relationships. — Adults of this species are like those of P . filiformis

Chaudoir, in size and body form. The latter species occurs in the east, and farther

southward in Mexico. General similarity plus allopatric ranges suggest that these

two species might be closely related to one another.

Pachyteles mexicanus (Chaudoir)

Figs. 76A-B, 89A-D, and 102.

Ozaena mexicana Chaudoir, 1848: 106. TYPE MATERIAL: two males, two females, in Chaudoir-

Oberthiir Collection, Box 132, each labelled "Ex Musaeo Chaudoir [red print], in front of

the following box label: "mexicana Chaud Mexique". LECTOTYPE (here selected): a male,

first specimen in series (MNHP). — 1854: 306.

Pachyteles mexicanus-, Chaudoir, 1868: 65. — Bates, 1881: 27. — Csiki, 1927: 429. —

Blackwelder, 1944: 23.- — Erwin, et al., 1977: 4.3.

Recognition. — Adults are broad-bodied, with pronotum with anterior angles

acute and projected laterally, and smooth or nearly smooth elytra.

Description. — With features of Pachyteles (s. str.) and body broad and sturdy.

Standardized Body length of males 5. 9-7. 2 mm., females 6. 6-7. 6 mm; W/SBL males 0.39-0.43,

females 0.42-0.46.

Metathoracic wings. Relatively small, with apical portion reduced.

Male genitalia (Figs. 89A-D). Median lobe in right lateral aspect (Fig. 89B) with apex

broadly bifid. Internal sac (Figs. 89A-B) narrow, with slender collar with dense covering of seta-

like microtrichia; apical part (only partly everted) with broadly pointed apex. Left paramere (Fig.

89C) with apex broadly pointed, asetose. Right paramere (Fig. 89D) slender, elongate, preapical

part of medial margin sparsely setose, setae short.

Ovipositor (Figs. 76A-B). Stylomeres moderately elongate, apex blunt, obliquely truncate;

nematiform setae terminal; surface with long slender trichoid sensilla and short broad basiconic

sensilla.

Way of life. — Specimens have been collected under bark of fallen trees in

lowland tropical rain forest, at about 720 m. elevation, and in cloud forest at

Quaest. Ent., 1990, 26(1)

98

Ball and McCleve

1400m., in a pile of leaf litter and oak branches. Specimens were collected at night,

on the surface, also. The reduced wings of adults indicate that flight is not possible,

and this indication is supported by absence of specimens from catches by light

traps.

Geographical distribution. — The range of this species extends from

Nicaragua northward to San Luis Potosi on the Gulf Versant of Mexico.

Relationships. — The distinctive male genitalia and ovipositor, and smooth

glabrous elytra place this species clearly apart from the others treated in this paper.

Material examined. — In addition to the types, we have seen 57 specimens of

this species, from the following localities in Mexico.

Chiapas. 11 km. S. Jitotol, Rte. 195, pine-sweetgum, ca. 1650 m., V. 5. 1977; Mexican

Exp. 1977, J.S. Ashe, H.E. Frania, D. Shpeley (UASM). Sierra de la Colmena, San Cristobal trail,

701-853 m. [Lacandon forest], June 8, 1972; P.A. Meyer, G.E.& K.E. Ball (UASM). Yerba

Buena Hospital, 2.4 km. N. Pueblo Nuevo, 1554-1829 m., June 21-22, 1972; P.A. Meyer, G.E.

Ball (UASM). Oaxaca. 21 km. S. Valle Nacional, 1128 m„ VIII. 71; A. Newton (MCZC). 9.7 km.

S. Valle Nacional, 650 m„ V.18-20.1971; H.. Howden (UASM). Same, VII.20-3 1.1971; 299CS

(UASM). 17.3 km. S. Valle Nacional, Rte. 175, montane trop. for., ca. 1000 m., IV. 26. 1977;

Mexican Exp. 1977, J.S. Ashe, H.E. Frania, D. Shpeley (UASM). Puebla. 2.4 km. N.

Tlaxcalantonga, VII. 3-8. 1971 ; 273 DH (UASM). San Luis Potosi. 29 km. S. Tamazunchale,

XI. 22. 1946; E.S. Ross (CASC). Veracruz. 3.9 km. N. Coscomatepec, 1400 m., VIII. 12. 1987;

Mexico Field Party, 1987: J. K. Liebherr, D.K. Millman (CUIC). 6.4 km. N. Huatusco, 1280 m.,

VII. 2. 1973; A. Newton (MCZC). 7.1 km. N. Huatusco, 1300 m., on clay bank, at night,

VIII. 15. 1987; Mexican Field Party, 1987: J.K. Liebherr, D.K. Millman (CUIC). 7 km. S. Huatusco,

cloud forest, ca. 2164 m., VII. 24. 1977; Mexican Exp., 1977, J.S. Ashe, H.E. Frania, D. Shpeley

(UASM). Jalapa; M. Trujillo (AMNH). Jalapa, May (CASC).

Ozaena Olivier

Ozaena Olivier, 1812: 617. TYPE SPECIES: Ozaena dentipes Olivier, 1812: 620; by

monotypy. — Dejean, 1825: 356, 433. — Dejean and Boisduval, 1829: 186, 231.— Dejean,

1831: 471. — Brulle, in Audouin and Brulle, 1834: 258. — Castelnau-Laporte, 1834: 144. —

Blanchard, in Cuvier, 1842: 127. — Chevrolat, in d'Orbigny, 1847: 376. — Lacordaire, 1854:

156.— Chaudoir, 1854: 289, 297.— 1868: 49.— Csiki, 1927: 427 Banninger, 1927:

193.— 1931: 184.— Blackwelder, 1944: 22.— Banninger, 1949: 132.— Ball, 1960: 95.—

Ogueta, 1965: 75.— Reichardt, 1977: 377.— Erwin, et at, 1977: 4.3.

Ictinus Castelnau-Laporte, 1834: 53. TYPE SPECIES: Ictinus tenebrioides Castelnau-

Laporte, 1834: 53; by monotypy; = O. dentipes Olivier. — 1835: 144. — Hope, 1838: 99. —

Duponchel, E., in d'Orbigny, 1846: 16.

Ozena Chenu, 1851: 87 (misspelling).

Nomenclatural note. — Hope (1838: 99) fixed "Ic. Rogerii Dejean" as type

species of Ictinus, but this name was not among those originally associated with

that generic name.

Derivation of generic name. — From Greek, meaning to smell, in allusion to the

odorous defensive secretions of the adults. Here is a fine but simple example of the

fact that even the museum taxonomists of the early 19th Century were cognizant of

features of living organisms, and were prepared to use such features. The bizarre

notion that museum taxonomists were interested in structural features only should

be put to rest.

Recognition. — Mature adults have black integuments. The antennae are

relatively long for ozaenines (Fig. 14). The labium (Fig. 54) has long lateral lobes

of the mentum, broadly rounded apically. The elytra and lateral margins of the

pronotum have short broad setae with ridged surfaces (Fig. 107A-B).

Description. — Habitus as in Fig. 7, size large. Standardized Body Length ca. 14-20 mm,

maximum width 5. 1-6.1 mm.

Color. Body and appendages black, or very dark piceous.

Quaest. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

99

Microsculpture and luster. As in Tribe Ozaenini, and on mesopleura mesh pattern

transverse, on metapleura isodiametric to transverse. Surface shining in most species, dull in

some.

Vestiture. Scrobes of mandibles with normal setae. Dorsal surface of elytra and lateral

margins of pronotum with short broad setae, surface ridged and flattened, and apex cleft (Fig.

107B). Males without adhesive vestiture on fore tarsomeres. Ventral surface generally setose,

with normal trichoid setae, abdominal sterna densely setose.

Chaetotaxy. Clypeus and vertex of head asetose, temporal areas each side with several

setae. Antennae (Figs. 7, 14, and 29A-B): with antennomeres sparsely setose, setae generally

short; antennomere 1 1 with dense concentration of sensilla basiconica in apical third to half.

Mouthparts: apical margin of labrum with about 12 setae; mandibular scrobes without setae;

maxillary stipes with two setae laterally; labial submentum and mentum with several setae each;

glossal sclerite apically without setae; labial palpomere 2 without long setae. Pronotum: without

fixed trichoid setae. Disc of elytra: without fixed trichoid setae. But note sensilla basiconica, Figs.

107A-C. Legs (fore, middle hind): coxae, O-numerous-numerous; trochanters, generally setose;

femora, asetose.

Head. Frontal impressions elongate, shallow, irregular. Eyes (Fig. 7) prominent,

posteriorly with moderately large temporal lobes. Supraantennal areas not reflexed, but

extended laterally each side as obtuse point.

Antennae. Long, extended posteriorly clearly past elytral humeri (Fig. 7); antennomeres

1-4 cylindrical, 5-11 (Fig. 14) moniliform, with 11 (Figs. 29A-B) distinctly swollen and apically

with distinct straight sharp keel.

Mouthparts. Labrum (Fig. 36) transverse. Mandibles (Figs. 42A-D) short, thick, occlusal

margin toothed. For details, see Table 1. Maxillae average in most respects, but lacinia (Figs.

47A,B) with thick brush of curved setae and terminated in short chisel-like tooth. Palpomere 4

thick, apex obliquely truncate. Labium (Fig. 54) with lateral lobes of mentum large, broadly

rounded apically; tooth short; epilobes narrow, not extended to apex; palpomere 3 thick, short in

O. lemoulti adults, apex truncate.

Prothorax. Pronotum (Fig. 7) short, distinctly wider than long. Sides explanate, elevated or

flat ( O . lemoulti). Disc convex. Impressions distinct. Prosternum with intercoxal process broad,

short.

Pterothorax. As for Tribe Ozaenini, and mid-coxal cavities closed medially by junction of

intercoxal processes of meso-and metasternum. Metepisternum overlapped by posterior lobe of

mesepimeron.

Elytra. Intervals broad, slightly elevated. Intemeurs shallow, punctate, punctures large. Basal

ridge very short, hardly evident. Humeri denticulate.

Metathoracic wings. Fully developed.

Legs. As described for Ozaenini, and fore femora (Fig. 61 A) about cylindrical, thickened

apically, each ventrally toward base with small setose denticulate process (Fig. 6 IB). Fore tibia

with antennal cleaner (Figs. 68A-B) reduced, or details, see Table 2. Tarsi average for Ozaenini.

Abdomen. Sclerites of segments II- VII unmodified, or details of sclerites of segments VIII

and IX/X, see under "Structural and biochemical features- genital segments".

Male genitalia (Figs. 90A-D). or details, see under Structural and biochemical features",

and Table 3.

Ovipositor (Figs. 80A-C). Stylomeres short, moderately densely setose apically, sensory

furrow apical, with single nematiform seta.

Bursa copulatrix and spermatheca. See Table 4 for details.

Way of life. — Evidently, most known specimens of this genus preserved in

museums have been collected at light, in lowland (principally tropical) forest.

Nothing else is known about habitat or activity.

Geographical distribution. — The range of Ozaena extends from Argentina in

South America to southern Arizona in North America. Only one species, O. lemoulti

Banninger, is known to occur in Middle and southern North America.

Relationships. — The sister group of Ozaena seems to be Platycerozaena ,

based on similarities in setal reduction, type of setae, reduction of antennal cleaner,

distribution of setation on the antennae, and form of the stylomeres, bursa

Quae st. Ent., 1990, 26(1)

00

Ball and McCleve

copulatrix, and spermatheca. We have considered seriously the proposition that

these two groups are congeneric. The relationships of the lineage represented by

these two genera is not evident at this time.

Included species. — Ozaena includes ten species, eight of which have been

described.

Ozaena lemoulti Banninger

Figs. 7, 14, 22, 29A-B, 36, 42A-D, 47A-B, 54, 61A-B, 68A-B, 80A-C, 90A-D, and

107A-D.

Ozaena lemoulti Banninger, 1932: 185. TYPE MATERIAL: HOLOTYPE male, labelled

"GUYANE FRANCSE St Jean du Maroni Collection LeMoult"; "Ozaena lemoulti

Banninger” [handwritten] (Banninger Collection, Zurich). TYPE LOCALITY: as indicated

on locality label of holotype. — Blackwelder, 1944: 23. — Banninger, 1949: 133. — Ogueta,

1965b: 87. — Erwin et al, 1977: 4.3. Banninger, 1956: 400.

Ozaena elevata Ball, 1960: 95 (not Banninger, 1956).

Ozaena halffteri Ogueta, 1965b: 83. HOLOTYPE female, labelled "Mexico, estado de Veracruz,

Tlapacoyan, 5.IX.1953, leg. Ticul Alvarez y Gonzalo Halffter" (Ogueta Collection). NEW

SYNONYMY.— Erwin, et al., 1977: 4.3.

Notes about synonymy. — We have not seen the holotype of O. halffteri, but

we have seen specimens from Arizona and from localities extending collectively

through the whole of Middle America and northern South America. A detailed study

shows that the supposed diagnostic features given by Ogueta in his key to species

{l.c., 76-77) exhibit too much variation to support the hypothesis that two species

are represented in the material noted above. Thus, we regard the names O. lemoulti

and O. halffteri as synonyms of one another.

The name Ozaena elevata was published for a specimen of O. lemoulti,

collected at Nogales, Arizona (Ball, 1960: 95), based on Banninger' s determination

as Ozaena elevata var? However, that determination was made before the material

was available to indicate the limits of O. lemoulti and O elevata. We are satisfied

that the present identification of the Nogales specimen is correct. Nonetheless, we

are a bit doubtful if O. lemoulti and O elevata Banninger are specifically distinct.

The material available is not sufficient to permit this second synonymy. See

Banninger, 1956: 400.

Recognition. — The only species of Ozaena in Middle and North America,

adults of this species might be confused only with those of Pachyteles kuntzeni,

which are also large and uniformly dark in color. Form of the antennae (Figs. 14, 22,

and 29A-B) distinguishes readily members of these taxa. The pronotum (Fig. 7) of

adult O. lemoulti is shorter, broader, and with lateral margins evenly curved, not

sinuate posteriorly. The elytra of adult O. lemoulti bear dorsally distinctive ribbed,

apically branched setae (Fig. 107A-B), not exhibited by adults of P. kuntzeni.

Description. — Habitus as in Fig. 7. Standardized Body length males 16.8-18.4 mm.,

females 16.4-18.2 mm.; W/SBL males 0.31, females 0.31-0.33. Other features as described

above, for genus.

Geographical distribution. — The range of this species extends from Ecuador

and Cayenne in northern South America through Middle America to southern

Arizona.

Relationships. — Based on details of body form, of distribution of antennal

sensilla, and of mandibular structure, and allopatric geographical distribution, we

postulate that O. lemoulti and O. elevata, if distinct, are sister species.

Material examined. — From Mexico and the United States, we have seen 1 1

specimens from the following localities. We have seen also 22 additional

specimens from localities in Brazil (Para), Cayenne, and Venezuela, in northern

Quaest. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

101

South America; and from Belize, Costa Rica, Guatemala, and Panama, in Middle

America .

UNITED STATES OF AMERICA. Arizona. Santa Cruz County. — Nogales, VI. 19. 1945, in

house (USNM). Pena Blanca Canyon, Pajarito Mts., 1191 m., VII. 27. 1978; S. McCleve (SMCC).

MEXICO . Chiapas. Palenque ruins, 100 m., U-V light, May 20, 1972; P. Meyer,

G.E.&K.E. Ball (UASM). Coahuila. Saltillo; E. Palmer (MCZC). Morelos. Cuernavaca; V.

Barrett (CASC). San Luis Potosi. El Salto, VIII. 8. 1966; O. S. Flint (USNM). Palitla, at light,

VII. 21. 1970; Schafffner, Murray, Phelps, Hart (TAMU). Tamazunchale, V. 20. 1952; M. Cazier, W.

Gertsch, R. Schrammel (AMNH). Tamaulipas. El Salto Falls, 42 km. W. Antiguo Morelos, 610

m., VII. 1 1-14.1963; Duckworth & Davis (USNM). Veracruz. Fortin de las Flores, 1010 m., at

light, VII. 7-12. 1974; J.A. Chemsak, E.& J. Linsley, and J. Powell (CISC).

Platycerozaena Banninger

Figs. 8, 15, 23, 30, 37, 43A-F, 48, 55, 62, 69A-B, 81A-C, 91A-D, and 96.

Ozaena (in part); Bates, 1874: 23.- — 1881: 25. — Csiki, 1927: 427.

Platycerozaena Banninger, 1927: 197. TYPE SPECIES: Ozaena brevicornis Bates, 1874: 24; by

monotypy. — Blackwelder, 1944: 23. — Ogueta, 1965c: 361. — Reichardt, 1977: 377. —

Erwin, 1979: 557.— Roach, et al., 1979: 18.

Ozaena ( Platycerozaena ) Banninger, 1949: 133.

Nomenclatural note. — Ogueta (1965c: 361) indicated O magna as type

species of Platycerozaena. However, this name was not associated with this

generic name when the latter was proposed.

Derivation of generic name. — From Greek, platyceros, meaning flat hom, to be

interpreted as antenna, combined with Ozaena ; literally, the Ozaena with flat

antennae.

Ranking. — Banninger (1949), without explanation, included Platycerozaena

as a subgenus of Ozaena. We agree with Ogueta (1965c: 361) that the two groups,

though no doubt closely related, are each monophyletic and abundantly distinct

from each other. Accordingly, we choose to rank Platycerozaena as a distinct

genus.

Recognition. — Adults of this genus are readily distinguished by black color,

elongate labrum (Fig. 37), short antennae with markedly transverse antennomeres 5-

10 (Fig. 15), mentum with very small tooth (Fig. 55), and small antenna cleaner

(Figs. 69A-B).

Description. — Habitus as in Fig. 8, moderate in size. Standardized Body length ca. 6.0-

8.0 mm., maximum width 1.8-2. 5 mm, slender in form.

Color. Body and appendages of mature adults black.

Microsculpture and luster. As for Tribe Ozaenini, and pterothoracic pleura and sterna with

transverse mesh pattern. Elytra with microlines fine, mesh pattern transverse. Surface shining to

subiridescent.

Vestiture. Dorsum with sparse covering of short setae, these flattened, expanded, ribbed

and branched on pronotum and elytra (cf. Fig. 107A-B). Ventral surface sparsely setose.

Chaetotaxy. Labrum (Fig. 37), clypeus, vertex and temporal areas of head, lateral margins

of pronotum and elytra without long tactile ("fixed") setae. Antennae: antennomeres 1-10 with

short trichoid setae; 5-10 each with patch of sensilla basiconica on ventral anterior and posterior

surfaces; antennomere 1 1 with sensilla basiconica extensive laterally and preapically; central

triangular area more or less glabrous (Figs. 23 and 30). Mouthparts: without fixed setae (as in

Ozaena). Legs (fore-middle-hind): coxae, numerous-numerous-numerous; trochanters,

numerous-numerous-sparse. Basal fore tarsomeres of males without adhesive setae.

Head. Frontal impressions broad, shallow. Eyes (Fig. 8) prominent, posteriorly with small

temporal lobe each side. Supraantennal area each side not reflexed, ridged, extended laterally as

obtuse point.

Antennae. Short, not extended posteriorly beyond basal margin of pronotum.

Antennomeres 1-4 more or less cylindrical, rather short; antennomeres 5-11 (Fig. 15) slender.

Quaes^En^^ 990^6^

102

Ball and McCleve

markedly compressed; antennomere 1 1 (Figs. 23 and 30) more elongate, broad, apical margin

broadly rounded in lateral aspect, apex narrowly keeled, keel sinuate (Fig. 30).

Mouthparts. Labrum (Fig. 37) elongate. Mandibles (Figs. 43A-F) with occlusal margins

toothed, short, broad (see Table 1 for details). Maxillae (Fig. 48) average for Ozaenini,

palpomere 4 markedly broad, apical margin obliquely truncate. Labium (Fig. 55): mentum with

long slender lateral lobes, each pointed apically; epilobes slender, terminated just short of apex;

palpomere 3 broad, apex truncate.

Prothorax. Pronotum transverse, lateral margins narrow, beaded, sinuate posteriorly or

evenly rounded; impressions distinct. Prosternum with intercoxal process short, rather slender.

Pterothorax. As for Ozaenini, and base of metepisternum narrowly overlapped by base of

mesepimeron.

Elytra. Intervals broad, slightly elevated. Interneurs narrow, punctate. Basal ridge very

short, hardly evident. Humeri denticulate.

Legs. In most features, as described for Ozaenini. Fore femora each clavate, with ventral

margin protruded as broad projection (Fig. 62). Antennal cleaner of fore tibia (Figs. 69A-B)

reduced (see Table 2 for details).

Abdomen. Sclerites of segments II- VII unmodified, or description of sclerites of segments

VIII and IX/X, see under "Structural and biochemical features, genital segments". See also Fig.

96.

Male genitalia. See under "Structural and biochemical features" and Table 3. See also Figs.

91A-D.

Ovipositor (Figs. 81A-C). Stylomeres short, moderately densely setose toward apex,

rather broad in ventral aspect, sensory furrow and nematiform setae not identified.

Bursa copulatrix and spermatheca. See Table 4 for details.

Defensive secretions. — Five compounds. See Table 5 for details.

Way of life. — Nothing has been reported about this topic for Platycerozaena.

The reduced setation suggests some unusual mode of living, in which tactile

sensation is not a premium. Paussines also exhibit reduction in sensilla, and they

live with ants. Perhaps, then, members of Platycerozaena are myrmecophilous, too.

Geographical distribution. — This genus is known only from central Brazil

northward to Nicaragua, in Lower Middle America.

Relationships. — This genus seems to be the sister group of Ozaena Olivier.

For details, see under the latter genus.

Included species. — According to Ogueta (1965c: 362-363), Platycerozaena

includes four species.

ZOOGEOGRAPHY

Because the boundaries of this study are artificial both phylogenetically and

geographically, it lacks the unity required to develop a coherent generalizing

evolutionary hypothesis. Accordingly, we attempt only to relate to general patterns

the taxonomic bits and pieces that we have treated.

The Tribes

Metriines are confined to dry temperate forests of the west coast of United

States, beyond the periphery of the range of the Ozaenini. The latter group is pan-

tropical, and in the New World ranges from the margins of the Sonoran desert in

southwestern United States to the Chilean rain forests in South America. The group

is centered in the lowland tropics, with numbers of species declining with

increasing altitude and latitude.

Two historical interpretations seem possible. First, the ancestral stock of the

Metriini and Ozaenini occupied the whole of Pangaea before its breakup, toward

the end of the Palaeozoic. With breakup, the northern vicar evolved into the

Quae st. Ent .. 1990. 26(1)

Middle American Genera of the Tribe Ozaenini

103

Metriini, and the southern one produced the Ozaenini (Erwin, 1979b: 577).

Second, the ancestral stock of the Metriini + Ozaenini was Gondwanian, appearing

after breakup of Pangaea. An initial split produced a less progressive line, the

Metriini, which was replaced gradually in the tropics by the more progressive

sister group, the Ozaenini. Metriines were replaced eventually, throughout the

tropics, and survive today only in a small area of the north temperate zone, beyond

the range of their ozaenine sister group. Either of these interpretations recognizes

Metrius as a relict group.

A more specific hypothesis accounts for the occurrence of Metrius (and other

relict taxa) in coastal areas in western United States. According to some

geologists, various small Pacific terranes have drifted eastward, eventually

encountering and becoming part of the west coast of North America. Perhaps these

tectonic plates carried with them the remnants of old taxa that populated the coastal

areas (Downes and Kavanaugh, 1988: 8). We give little credence to this

possibility.

The Genera

Patterns. — Four of the Middle American genera ( Physea , Pachyteles , Ozaena

and Platycerozaena) occur in both the South and North American continents.

Entomoantyx is confined to the Middle American part of the North American

continent. In terms of northward limits, Physea , Pachyteles , and Ozaena reach

southwestern United States. The range of Entomoantyx extends north of the

Isthmus of Tehuantepec in northeastern Mexico, and Platycerozaena reaches only

Nicaragua, at the southern edge of Nuclear Middle America. Pachyteles is the only

ozaenine genus known from the West Indies, where it is represented in the Greater

Antilles by two species. In South America, Physea , Pachyteles, and Ozaena reach

Argentina, whereas Platycerozaena reaches only central Brazil.

In terms of diversity, all of the bi-continental genera are more speciose on the

South American than the North American continent, though for Physea and

Platycerozaena the difference is slight.

Historical interpretation. — Accepting a Gondwanian origin of the Ozaenini,

we believe that occurrence of the tribe in Middle America and southwestern United

States on the North American continent, must have been accomplished by

dispersal — in part over sea and in part over land, in Tertiary time. There could

have been an over land movement, if in fact there was an early Tertiary inter¬

continental connection, as has been postulated by some authors (for a discussion,

see Donnelly, 1988, and Savage, 1982).

Based on phylogenetic position, diversity pattern and extent of northward

penetration, we believe the genera arrived in the following sequence: first, the

ancestral stock of Entomoantyx , a genus which we take to be a relict; second,

Pachyteles and Physea ; third, Platycerozaena (few Middle American species,

limited penetration); and fourth, Ozaena (only one species, which occurs in northern

South America, as well).

Number of incursions postulated varies. For the bi-continental genera, we

indicate: one each, for Physea and Ozaena ; for Platycerozaena, two (one endemic

species in Middle America, and one shared between the northern and southern

continents); for Pachyteles, several incursions, with several each for the

nominotypical subgenus and for Goniotropis.

For time of arrival, we postulate Late Cretaceous to Early Tertiary for

Entomoantyx’, Middle Tertiary for Physea’, Pliocene and Pleistocene for

Platycerozaena ; and Pleistocene for Ozaena. Because Pachyteles is represented by

several of the same species on both continents, as well as by an array of endemic

Quaes^En^^990^6{\)

104

Ball and McCleve

species in Middle America, we postulate a range of times from Early Tertiary to

Pleistocene and possibly even Recent.

The Greater Antilles were invaded probably at two different times by

Pachyteles : once early, possibly in mid-Tertiary (this invasion represented by an

endemic flightless species, living in the Jamaican highlands, undescribed, and

possibly now extinct); and once recently (Pleistocene or Recent), and represented

by P. gyllenhali , on the island of Cuba. The Lesser Antilles were probably invaded

comparatively recently by one or two species of Pachyteles.

Mid-Tertiary arrival of the ancestral stock of the species of Pachyteles on

Jamaica could have been facilitated by a land bridge, now foundered, but

represented by the Nicaraguan Rise (Donnelly, 1988), or possibly by a mobile

Jamaica that was closer to the mainland than it is now (Rosen, 1985, and references

therein). We are inclined to doubt the latter possibility.

In conclusion, then, we postulate that the Middle-North American fauna has

developed partly by incursions of taxa from South America, and partly by in situ

differentiation of invading stocks. This is a common biogeographic pattern,

described by many biogeographers.

Ozaenine Species of Southwestern United States and Vicinity

The pattern. — This is described partly in terms of extent of range and

habitats. The latter are numbered below, according to the classification of Brown

and Lowe (1980) and Brown, et al. (1980). Of the ozaenine species that are in or

near southwestern United States, one {Physea hirta) reaches its northern limits in

southern Texas, on the Gulf Versant. The remaining six are in the west, entering

United States in Arizona, with the range of Pachyteles gyllenhali extended

northward to approximately 33°30'N.

The northwestern species comprise two groups in terms of extent of range: those

confined to the west ( Physea latipes, Pachyteles parca, P. kuntzeni , and P .

enischnus; Map 1), and those with ranges extended southward for various

distances (P. gyllenhali and Ozaena lemoulti). Physea hirta belongs to this latter

group, also.

In terms of habitat, the species confined to the west occupy dry habitats

primarily, including tropical Sinaloan thorn scrub (134.3), Sinaloan deciduous

forest (124.6), Madrean evergreen woodland (123.3), Sonoran desert scrub

(154.11), Chihuahuan desert scrub (153.2), and interior chaparral (132.2).

However, it is important to realize that desert areas are probably marginal for

ozaenines, because no extensive diversification of the group has taken place in

such areas, and no species are confined to desert habitats. The wide-ranging

western species occupy some of these habitats, and as well Riparian cottonwood-

willow series (224.53), and southward, evergreen tropical and cloud forest.

Another aspect of the pattern is distribution of putative sister species.

Physea hirta and P. latipes are probably sister species, and their range overlap is

only partial (Map 1). Pachyteles parca and P. kuntzeni seem to be sister species,

whose ranges are in contact, probably narrowly, in southern Arizona and in northern

Mexico. P . enischnus has a probable sister species (P. filiformis Chaudoir)

filiformis Chaudoir, Pachyteles.in eastern Mexico. The sister species of P. gyllenhali

as not been postulated, though we suspect it will be South American. That of

Ozaena lemoulti is probably the more southern Brazilian O. elevata Banninger, with

the area of disjunction being the Amazon Basin.

The fact that most of the northern ozaenine species have vicariant or

parapatric sister species suggests relatively recent differentiation of each of the

stocks to which these species belong. From the standpoint of interpretation, it is

Middle American Genera of the Tribe Ozaenini

105

Map 1. Geographical positions of known localities in Mexico and southwestern United States for

five species of Ozaenini.

particularly interesting that each of the three species endemic on the Pacific

Versant ( Physea latipes, Pachyteles parca, and P. enischnus ) have putative eastern

counterparts.

In summary, the species at the northern end of the range of the Ozaenini occupy

there principally dry forests and marginally, desert habitats. Most of the endemic

Middle American-U.S. taxa are confined to such habitats. The two taxa whose

ranges extend into South America plus Physea hirta occur in wet tropical forests, as

well. The endemic Middle American-U.S. taxa exhibit east-west disjunction or

near-disjunction, of closely related species.

Geographical history. — The general pattern described above is like the

distribution patterns of many other taxa that occur in the same area (see Liebherr,

1986: 161-172, for details of the Agonum extensicolle group, and associated

references to other authors and taxa. See also Ball and Nimmo, 1983, and Ball and

Maddison, 1987).

The explanation seems rather simple. Each ancestral stock of the Recent

endemic Middle American elements was divided and thus isolated, to the east and

west of the north-south trending Sierra Madre Occidental, as a result of drying of

the climate in the later part of the Tertiary Period. In isolation, the now vicariant

elements differentiated from one another, becoming specifically distinct. Also, the

Quaest. Ent., 1990, 26(1)

106

Ball and McCleve

western species probably became adapted to some extent to desert conditions, and

succeeded in establishing in habitats marginal to their ancestral dry forests.

In Holocene time, and possibly earlier during the wetter glacials, the erstwhile

vicariants P. parca and P. kuntzeni came into contact with one another, their

ranges overlapping. During the wetter glacials of the Pleistocene, probably, the

ranges of Pachyteles gyllenhali and Ozaena lemoulti expanded, (the former species

reaching Cuba in the Caribbean Ocean, and the Tres Marias Islands in the Pacific),

and these taxa reached the northern limits of the tribe Ozaenini in the New World,

and came to overlap the ranges of the more northern endemics. Possibly the range of

Physea hirta expanded similarly, with this predominantly eastern species

spreading through the Isthmus of Tehuantepec to the lowlands on the Pacific

Versant of Mexico.

More generally, we believe that these northern ozaenine stocks provide some

perception about how the dry adapted fauna of western Mexico evolved. Certain

species, ranging northward from the wet tropics, are able to enter drier habitats. In

the course of changing circumstances, the expanded ranges of these species become

disjunct. In the fullness of time, and under the influence of natural selection, the dry

forest stocks become adapted to such conditions to the extent that they either

become confined to them, or are able to enter still drier habitats. So far, ozaenines

have not penetrated the extreme habitats encountered in the Sonoran and

Chihuahuan deserts. Such penetration remains as a future evolutionary possibility.

Corollaries. — If this zoogeographic scenario is correct, the following should

be found to be true.

1 . Future work will not refute the hypothesized sister-group relationships.

2. Pachyteles parca , most collections of which have been made in Arizona, will

be found further south along the Pacific coast, in Sinaloan deciduous forest.

3. The range of Pachyteles kuntzeni is in fact limited in northwestern Mexico,

to the eastern slopes of the Sierra Madre Occidental, as is suggested by present

limited data. It will be found, farther south, in tropical forest, as well.

CONCLUDING STATEMENT

Nearly 30 years ago, when Ross H. Arnett, Jr., published his treatment of the

beetle genera of North America north of Mexico, knowledge of the included

Ozaenini was hardly more than what was known in the previous century, when most

of the species of the group resident in southwestern United States were described.

In this work, we add a few points about relationships and distribution of these

species, but much more must be done, both ecologically and systematically, to bring

this small but markedly divergent faunule (seven species, only, in three genera) to

that stage of understanding at which ecologists can make use of the species as

elements in ecosystems, et cetera.

At a more general level, we have added to understanding of the ozaenine

genera of Middle America by detailed comparisons of a number of systems of

structural features, and by entering into the structural characterizations the

important data that have been generated by others concerning the defensive

secretions of the pygidial glands. For the first time, we have used in detail the

features of Metrius in out-group comparison to polarize in an evolutionary context

the features of the ozaenine genera. Still to be done is to work out relationships of

these genera on a worldwide basis, and to relate the reconstructed phylogenetic

patterns to the movement of continents as described by plate tectonic theory.

Quaest. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

107

In our investigations of structural features, we have uncovered a wealth of

detail that could be analyzed profitably by functional morphologists: mouthparts

and ovipositors are two such systems.

We hope we have provided an adequate basis for the next stage of systematic

analysis of the New World Ozaenini, which must be treatment of the species,

particularly of Pachyteles ( sensu lato ). We hope that those who use this publication

find in the wealth of implied questions about ozaenines adequate recompense for

the lack of answers that we have been able to provide.

ACKNOWLEDGEMENTS

We thank the entomologists whose names appear in the list of codens and

addresses of collections from which we received the material on which this study is

based. The senior author adds a special note of appreciation to those who made

possible his visits to study their collections, and who extended warm hospitality

to him during those visits: M. E. Bacchus, P. M. Hammond, R. D. Pope, and N. E.

Stork (BMNH); D. H. Kavanaugh (CASC); S. R. Shaw and D. R. Maddison

(MCZC); and L. Sims and T. L. Erwin (USNM). Lee H. Herman (AMNH) made

special efforts to arrange for the loan of the type of Pachyteles beyeri Notman.

Most of the line drawings of the male genitalia were prepared by D. R.

Maddison, while he was a graduate student at the University of Alberta. D. Shpeley

and G. D. Bray brook worked together in preparing the material and photographs of

structural features, using the Department of Entomology's scanning electron

microscope. J. S. Scott prepared the photographs of habitus and the plates of

illustrations.

At our request, a draft of the manuscript was reviewed by Yves Bousquet

(Biosystematics Research Centre, Agriculture Canada, Ottawa, Ontario) and by

Donald R. Whitehead (Systematic Entomology Laboratory, United States

Department of Agriculture, Washington, D. C.). Although we were not able to take

advantage of all of their proposals for improvement, we adopted many of them and

made corrections, as required.

We hope that our colleagues and associates who contributed so generously to

this study find adequate recompense for their efforts in the resulting publication.

Financial support for field and museum work was received through grants to

the senior author: A- 1399, Natural Sciences and Engineering Research Council of

Canada; and GB-3312, National Science Foundation, U.S.A. Publication costs were

met with funds from NSERCC Grant A- 1399. We are grateful for this important

assistance.

REFERENCES CITED

Allen, R. T. and G. E. Ball, 1980. Synopsis of Mexican taxa of the Loxandrus

series. (Coleoptera: Carabidae: Pterostichini). Transactions of the American

Entomological Society, 105: 481-576.

Andrewes, H. E. 1929. Coleoptera Carabidae. Vol. 1.- Carabinae. In, The Fauna of

British India, including Ceylon and Burma (J. Stephenson, Editor). Taylor and

Francis, London, xviii + 431 pp., 5 plates.

Audouin, J. V. and A. Brulle. 1834. Histoire naturelle des Insectes, traitant de leur

organization et de leurs moeurs en general, et comprenant leur classification et

la description des especes. Volume 4, Coleopteres I. . D. Pillot, Paris. 479 pp.

Banninger, M. 1927. Die Ozaenini. Deutsche Entomologische Zeitschrift for 1927,

177-216.

Quae st. Ent., 1990, 26(1)

108

Ball and McCleve

Banninger, M. 1932. Uber Carabinae, Erganzungen und Berichtigungen. (Col.) 17

Beitrag. Deutsche Entomologische Zeitschrift, (1931): 177-212.

Banninger, M. 1949. Ueber Carabinae (Col.) Erganzungen und Berichtgungen III,

mit Bemerkungen zu R. Jeannel's neuer Einteilung der Carabiden. Mitteilungen

der Miincher Entomologischen Gesellschaft e. V., 35-39, 1945-1949, pp. 127-

157.

Banninger, M. 1956. Uber Carabinae, Erganzungen und Berichtigungen (IV).

Entomologischen Arbeiten aus dem Museum G. Frey, Tutzing, 7 (1): 398-411.

Ball, G. E. 1960. Carabidae, Fascicle 4, pp. 55-210. In , The beetles of the United

States (A manual for identification) (R. H. Arnett, Editor). The Catholic

University of America Press, Washington, D. C. XI + 1112 pp.

Ball, G. E. 1979. 2.21. Conspectus of carabid classification: history,

holomorphology and higher taxa, pp. 63-111. In, Carabid beetles, their

evolution, natural history and classification, etc . (T. L. Erwin, G. E. Ball, D. R.

Whitehead and A. L. Halpem, Editors). Dr. W. Junk bv Publishers. The Hague,

The Netherlands. X + 644 pp.

Ball, G. E. 1985. The contributions of Philip J. Darlington, Jr. to the study of

carabid beetles in the Americas, and a bibliography of his publications, pp.

17-30. In, Taxonomy, phylogeny and zoogeography of beetles and ants: a

volume dedicated to the memory of Philip Jackson Darlington, Jr. (1904-

1983). (G. E. Ball, Editor). Dr. W. Junk Publishers, Dordrecht, Boston,

Lancaster. XIV + 514 pp.

Ball, G. E. and D. R. Maddison. 1987. Classification and evolutionary aspects of

the species of the New World genus Amblygnathus Dejean, with description

of Platymetopsis new genus, and notes about selected species of

Selenophorus Dejean (Coleoptera: Carabidae: Harpalini). Transactions of the

American Entomological Society, 113: 189-307.

Ball, G. E. and A. P. Nimmo. 1983. Synopsis of the species of subgenus

Progaleritina Jeannel, including reconstructed phylogeny and geographical

history (Coleoptera: Carabidae: Galerita Fabricius). Transactions of the

American Entomological Society, 109: 295-356.

Basilewsky, P. 1962. Les Ozaeninae d'Afrique et de Madagascar. Revue de

Zoologie et Botanique Africaine, 66 (3-4): 291-314.

Bates, H. W. 1874. Notes on Cicindelidae and Carabidae and descriptions of new

species (No. 17). Entomologists Monthly Magazine, 11: 22-28.

Bates, H. W. 1881. Insecta, Coleoptera, Carabidae, Volume 1, part 1, pp. 1-40,

plates i-ii. In, Biologia Centrali-Americana, 1879-1911, Coleoptera, 7 volumes

in 17 parts (F. D. Godman and O. Salvin, Editors). London.

Bell, R. T. 1967. Coxal cavities and the classification of the Adephaga

(Coleoptera). Annals of the Entomological Society of America, 60 (1): 101 -

107.

Bell, R. T. 1983. What is Trachypachusl (Coleoptera: Trachypachidae), pp. 590-

596. In, The Baron Maximilien de Chaudoir (1816-1881): a symposium to

honor the memory of a great coleopterist during the centennial of his death (D.

R. Whitehead, Editor). The Coleopterists Bulletin, 36 (3&4): 459-609.

Bils, W. 1976. Das Abdomenende weiblicher, terrestrich lebender Adephaga

(Coleoptera) und seine Bedeutung fur die Phylogenie. Zoomorphologie, 84

(2): 113-193.

Blackwelder, R. E. 1944. Checklist of the coleopterous insects of Mexico, the

West Indies and South America. Part 1. Bulletin of the United States National

Museum, No. 185, pp. 1-188.

Blanchard, E. See below, under Cuvier.

Quaest. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

109

Bousquet, Y. 1986. Description of first instar larva of Metrius contractus

Eschscholtz (Coleoptera: Carabidae) with remarks about phylogenetic

relationships and ranking of the genus Metrius Eschscholtz. The Canadian

Entomologist, 118: 373-388.

Brown, D. E. and C. A. Lowe. 1980. [Map of} Biotic communities of the

southwest. General Technical Report RM 78. Rocky Mountain Forest and

Range Experiment Station Forest Service, U.S.D.A. U. S. Government Printing

Office, 1980. 679.042.

Brown, D. E., C. A. Lowe, and C. P. Pase. 1980. A digitized systematic

classification for ecosystems with an illustrated summary of the natural

vegetation of North America. U.S.D.A. Forest Service General Technical

Report RM-73, 93 pp. Rocky Mountain Forest and Range Experiment

Station, Fort Collins, Colorado.

Brulle, A. 1834. See Audouin and Brulle, above.

Castelnau, F. L. Laporte, de. 1834. Etudes entomologiques ou description des

insectes nouveaux; et observations sur leur synonymie. Paris, viii + pp. 1 —

94.

Castelnau, F. L. Laporte, de. 1835. Etudes entomologiques ou description d'

insectes nouveaux; et observations sur leur synonymie, part 1. Paris, pp. 95-

195. 4 colored plates.

Chaudoir, M. de. 1848. Memoire sur la famille des carabiques. Bulletin de la

Societe Imperiale des Naturalistes de Moscou, 21 (1): 3-134.

Chaudoir, M. de. 1852. Memoire sur la famille des Carabiques, Part 3. Bulletin de la

Societe Imperiale des Naturalistes de Moscou, 25 (1): 3-104.

Chaudoir, M. de. 1854. Memoire sur la famille des carabiques. Bulletin de la

Societe Imperiale des Naturalistes de Moscou, 27: 112-144, 279-352.

Chaudoir, M. de. 1868. Revision du groupe des ozenides. Annales de la Societe

Entomologique Belgique, 11: 43-74.

Chenu, J. C. 1851. Encyclopedic d’histoire naturelle ou Traite complet de cette

Science d'apres les travaux des naturalistes les plus eminents de tous les pay et

de toutes les epoques. Coleopteres Cicindeletes, carabiques, etc. Volume 1.

Paris. 312 pp., 12 + 28 plates.

Chevrolat, L. A. A. See below, under d'Orbigny.

Crowson, R. A. 1955. The natural classification of the families of Coleoptera.

Nathaniel Lloyd and Co., Ltd. 187 pp., 213 figs.

Crowson, R. A. 1981. The biology of the Coleoptera. Academic Press, London and

New York, xi + 802 pp.

Csiki, E. 1927. Coleopterorum Catalogus, pars. 92, Carabidae: Carabinae II, pp.

315-622 (Vol. 1).(S. Schenkling, Editor). W. Junk, Berlin W 15.

[Cuvier, G.]. 1842. Le regne animal distribue d apres son organisation, pour servir

de base a l'histoire naturelle des animaux, pp. 121-128. Volume 1, Disciples'

Edition, 1838-1845. Paris. 557 pp.

Darlington, P. J., Jr. 1950. Paussid beetles. Transactions of the American

Entomological Society, 76: 47-142.

Dejean, P. F. M. A. 1825. Species general des Coleopteres de la collection de M. le

Comte Dejean. Volume 1. Mequignon-Marvis, Paris, xxx + 463 pp.

Dejean, P. F. M. A. 1831. Species general des Coleopteres de la collection de M. le

Comte Dejean. Mequignon-Marvis, Paris, viii + 883 pp.

Dejean, P. F. M. A. 1834. Catalogue des Coleopteres de la collection de M. le

Comte Dejean, Deuxieme edition, livr. 3, pp. 177-256. Paris.

Dejean, P. F. M. A. 1837. Ibid., Troisieme edition, revue, corigee et augmented, livr.

1-4, pp. 1-468. Paris.

Quae st. Ent., 1990, 26(1)

110

Ball and McCleve

Dejean, P. F. M. A. and J. B. A. Boisduval. 1829. Iconographie et histoire naturelle

des coleopteres d'Europe. Tome 1. Mequignon-Marvis, Paris. XIV + 400 pp,

60 plates.

Deuve, T. 1988. Etude phylogenetique des Coleopteres Adephaga: redefinition de

la famille Harpalidae, sensu novo, et position systematique des

Pseudomorphinae et Brachinidae. Bulletin de la Societe Entomologique de

France, 92 (5-6): 161-182.

Donnelly, T. W. 1988. 2. Geologic constraints on Caribbean biogeography, pp. 15-

37. In, Zoogeography of Caribbean insects (J. K. Liebherr, Editor). Comstock

Publishing Associates, Cornell University Press, Ithaca and London, xi + 285

pp.

Downes, J. A. and D. H. Kavanaugh. 1988. Origins of the North American insect

fauna: introduction and commentary, pp. 1-12. In, Origins of the North

American insect fauna (J. A. Downes and D. H. Kavanaugh, Editors).

Entomological Society of Canada, Memoirs, No. 144: 168 pp.

Duponchel, E. See d'Orbigny, C. D., 1845, below.

Dupuis, P. 1911. Coleoptera Adephaga Carabidae Subfam. Metriinae and

Mystropominae, asc. 116, 3 pp., 2 plates. In, Genera Insectorum (P. Wytsman,

Editor). Verteneuil et Desmet, Bruxelles.

Eisner, T., T. H. Jones, D. J. Aneshansley, W. R. Tschinkel, R. E. Silberglied, and J.

Meinwald. 1977. Chemistry of defensive secretions of bombardier beetles

(Brachinini, Metriini, Ozaenini, Paussini). Journal of Insect Physiology, 23:

1383-1386.

Eisner, T. and D. J. Aneshansley. 1981. Spray aiming in bombardier beetles: jet

deflection by the Coanda effect. Science, 215: 83-85.

Emden, F. I. van. 1936. Eine interessante zwischen Carabidae und Paussidae

vermittelnde Kaferlarve. Arbeiten uber physiologische und angewandte

Entomologie aus Berlin-Dahlem, 3: 250-256.

Emden, F. I. van. 1942. A key to the genera of larval Carabidae. Transactions of the

Royal Entomological Society of London, 92: 1-99, 100 figs.

Erwin, T. L. 1979a. The American connection, past and present, as a model

blending dispersal and vicariance in the study of biogeography ,pp. 355-367.

In, Carabid beetles: their evolution, natural history and classification (T. L.

Erwin, G. E. Ball, D. R. Whitehead and A. L. Halpem, Editors). W. Junk, b.v.,

Publishers, The Hague. X + 644 pp.

Erwin, T. L. 1979b. Thoughts on the evolutionary history of ground beetles:

hypotheses generated from comparative faunal analyses of lowland forest sites

in temperate and tropical regions (Coleoptera: Carabidae), pp. 539-592. Ibid.

Erwin, T. L. 1985. The taxon pulse: a general pattern of lineage radiation and

extinction among carabid beetles, pp. 437-493. In, Taxonomy, phylogeny and

zoogeography of beetles and ants. A volume dedicated to the memory of

Philip Jackson Darlington, Jr. (1904-1983). (G. E. Ball, Editor). Dr. W. Junk

Publishers, Dordrecht, Boston, Lancaster. XIII + 514 pp.

Erwin, T. L. and A. L. Halpem. 1978. Max Banninger: his collection and

publications (Coleoptera: Carabidae). The Coleopterists Bulletin, 32: 357-

366.

Erwin, T. L. and L. L. Sims. 1984. Carabid beetles of the West Indies (Insecta:

Coleoptera): a synopsis of the genera and checklist of tribes of Caraboidea,

and of the West Indian species. Quaestiones Entomologicae, 20 (4): 350-466.

Erwin, T. L., D. R. Whitehead, and G. E. Ball. 1977. Family 4. Carabidae. The

ground beetles, pp. 4.1-4.68. In, Checklist of the beetles of Canada, United

States, Mexico, Central America, and the West Indies (R. E. Blackwelder and

Quaest. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

111

R. H. Arnett, Jr., Editors). World Digest Publications, Oxycopis Pond,

Wallace Road, Kinderhook, New York, 12106.

Evans, M. E. G. 1986. Carabid locomotor habits and adaptations, pp. 59-77. In,

Carabid beetles, their adaptations and dynamics. XVIIth International

Congress of Entomology, Hamburg, 1984 (P. J. den Boer, M. L. Luff, D.

Mossakowski and F. Weber, Editors). Gustav Fischer, Stuttgart- New York.

Contents + 351 pp.

Evans, M. E. G. and T. G. Forsythe. 1985. Feeding mechanisms and their variation

in form, of some adult ground beetles (Coleoptera: Caraboidea). Journal of

Zoology, London (A), 206: 113-143.

Forbes, W. T. M. 1926. The wing folding patterns of the Coleoptera. Journal of the

New York Entomological Society, 34: 42-139.

Forsyth, D. J. 1972. The structure of the pygidial glands of the Carabidae.

Proceedings of the Zoological Society of London, 32 (4): 253-309.

Forsythe, T. G. 1982. Feeding mechanisms of certain ground beetles (Coleoptera:

Carabidae). The Coleopterists Bulletin, 36 (1): 26-73.

Forsythe, T. G. 1987. Ground beetles: form and function. Antenna, 11 (2): 57-61.

Goulet, H. 1979. 2.25. Contributions of characters of larvae to systematics of

Carabidae, pp. 205-208. In , Carabid beetles: their evolution, natural history

and classification (T. L. Erwin, et al.. Editors). Dr. W. Junk bv Publishers. The

Hague, The Netherlands. X + 644 pp.

Gray, G. R. 1832. Notices of new genera and species. In, The Animal Kingdom

arranged in conformity with its organization by the Baron Cuvier, vol. 14

(Insecta, vol. 1), (Griffith and Pidgeon, Editors). London. 570 pp.

Hlavac, T. . 1971. Differentiation of the carabid antenna cleaner. Psyche, 78: 51-66.

Hope, F. W. 1838. The Coleopterists Manual, Part the Second, containing the

predaceous land and water beetles of Linnaeus and Fabricius. H. G. Bohn,

London, xvi + 168 pp., 3 colored plates.

Horn, G. H. 1868. New species of Coleoptera from the Pacific District of the

United States. Transactions of the American Entomological Society, 2: 129-

178.

Horn, G. H. 1881. On the genera of Carabidae with special reference to the fauna of

Boreal America. Transactions of the American Entomological Society, 9: 91-

196, plates 1-10.

Horn, G. H. 1894. The Coleoptera of Baja California. Proceedings of the California

Academy of Sciences, (2), 4: 302-449.

Iakobson, G. G. 1906. Zhuki Rossii i Zapadnoi rukovdstvo k opredieleniiv

Zhukov. Lief. 4, pp. 241-320. A. F. Devrient, St. Petersburg.

Jeannel, R. 1941. Coleopteres carabiques, premiere partie. Faune de France, 39: 1-

571, 213 Figs. Paris.

Jeannel, R. 1946. Coleopteres carabiques de la region malgache (premiere partie).

Faune de l'Empire francaise, 6: 1-372.

Johnson, J. W. 1978. A preliminary list of South Texas Carabidae. Cordulia, 4 (2):

66-69.

Klug, J. C. F. 1834. Jahrbiicher der Insektenkunde, mit besonderer Rucksicht auf die

Sammlung im Konigl. Museum in Berlin herausgegeben. Volume 1. Enslin,

Berlin. 96 pp., 2 plates.

Kolbe, H. 1927. Die Einstammigkeit der Paussiden und die primitiven Gattungen

dieser myrmekophilen Coleopteren familie. Zoologische Anzeiger, 72: 205-

218.

Kryzhanovskij, O. L. 1976. Revised classification of the family Carabidae.

Entomological Reviews 1: 80-91 (translated from Russian).

Quaest. Ent., 1990, 26(1)

112

Ball and McCleve

Lacordaire, T. 1854. Histoire naturelle des insectes, genera des coleopteres ou

expose methodique et critique de tous les genres proposees jusqu'ici dans cet

ordre d'insectes. Volume 1. Cicindeletes-Palpicornes. Libraire

encyclopedique de Roret, Paris, x + 486 pp.

Latreille, P. A. 1817 [Insectes]. In, Le regne animal (G. C. L. D. Cuvier), Vol. 3.

Paris. 653 pp.

Latreille, P. A. 1829. Les crustaces, les arachnides et les insectes, distribues en

families naturelles, ouvrage formant les tomes 4 et 5 de celui de M. le Baron

Cuvier sur le regne animal (deuxieme edition). Vol. 1. Paris. 584 pp.

Latreille, P. A. and P. F. M. A. Dejean. 1822. Histoire naturelle et iconographie des

insectes coleopteres d'Europe. Livr. 1. Crevot, Paris. 89 pp., plates 1-5.

LeConte, J. L. 1853. Notes on the classification of the Carabidae of the United

States. Transactions of the American Philosophical Society (2), 10: 363-403.

LeConte, J. L. 1861. Classification of the Coleoptera of North America. Prepared

for the Smithsonian Institution. Smithsonian Miscellaneous Collections, III (No.

136): 1-286. Part 1.

LeConte, J. L. 1884. Short studies of North American Coleoptera (No. 2).

Transactions of the American Entomological Society, 12: 1-32.

LeConte, J. L. and G. H. Horn. 1883. Classification of the Coleoptera of North

America. Prepared for the Smithsonian Institution. Smithsonian Miscellaneous

Collections, 26, part 4 (No. 507), pp. i-xxxvii, 1-567.

Leng, C. W. 1920. Catalogue of the Coleoptera of America north of Mexico.

Sherman, Mt. Vernon, New York. X + 470 pp.

Liebherr, J. K. 1986. Cladistic analysis of North American Platynini and revision

of the Agonum extensicolle group (Coleoptera: Carabidae). University of

California Publications in Entomology, 106: x + 198 pp.

Lindroth, C. H. 1969. The ground-beetles (Carabidae excl. Cicindelinae) of Canada

and Alaska, Part 1, pp. I-XLVIII. Opuscula Entomologica, Supplementum 35.

Luna de Carvalho, E. 1959. Notas sobre Paussideos (Col. Carab. Isochaeta).

Publicacoes culturais da Companhia de Diamantes de Angola, 48: 47-90.

Moore, B. P. 1979. 2.24. Chemical defense and its bearing on phylogeny, pp. 193-

203. In, Carabid beetles: their evolution, natural history and classification. ..(T.

L. Erwin, G. E. Ball, D. R. Whitehead, and A. L. Halpern, Editors). W. Junk,

b.v., Publishers. The Hague. X + 635 pp.

Moore, B. P., T. A. Weir, and J. E. Pyke. 1987. Rhysodidae and Carabidae, pp.

20-320. In, Volume 4, Coleoptera: Archostemata, Myxophaga and Adephaga,

viii + 444 pp. In, Zoological catalogue of Australia (D. W. Walton, Executive

Editor). Australian Government Publishing Service, Canberra.

Nagel, B. P. 1979a. The classification of Carabidae, pp. 7-14. In, On the evolution

of behaviour in carabid beetles. (P. J. den Boer, H.-U. Thiele, and . Weber,

Editors). Miscellaneous Papers 18 (1979). Agricultural University,

Wageningen, The Netherlands. A. Veenman and Zonen B.V.- Wageningen, The

Netherlands. Contents + 222 pp.

Nagel, B. P. 1979b. Aspects of the evolution of myrmecophilous adaptations in

Paussinae (Coleoptera, Carabidae), pp. 15-34. Ibid.

Notman, H. 1919. Records and new species of Carabidae. Journal of the New York

Entomological Society, 27 (2-3): 225-237.

Ogueta, E. 1965a. Descripciones, redescripciones y comentarios sobre Ozaenini I.

(Coleop. Carabidae). Revista de la Sociedad Entomologica Argentina, 26:

111-118.

Ogueta, E. 1965b. Descripciones, redescripciones y comentarios sobre Ozaenini II.

El genero Ozaena Olivier (Coleoptera: Carabidae). Physis, 25 (68): 73-88.

Quaest. Ent., 1990, 26(1)

Middle American Genera of the Tribe Ozaenini

1 13

Ogueta, E. 1965c. Descripciones, redescripciones y comentarios sobre Ozaenini III.

El genero Platycerozaena Banninger (Coleoptera: Carabidae). Physis, 25 (70):

361-368.

Olivier, A. G. 1812. Encyclopedic methodique. Histoire naturelle. Insectes, vol. 8

(pt. 2), pp. 361-722.

d'Orbigny, C. D. 1845. Dictionnaire universel d'histoire naturelle. Volume 6. MM

Renard, Martinet et cie. Paris. 792 pp.

d'Orbigny, C. D. 1846. Ibid. Volume 7. Ibid. 808 pp.

d'Orbigny, C. D. 1847. Ibid. Volume 9. Ibid. 776 pp.

d'Orbigny, C. D. 1848. Ibid. Volume 12. Ibid. 816 pp.

Perty, J. A. M. 1830. Insecta Brasiliensia. In , Delectus animalium articulatorum, quae

in itinere per Brasiliam annis MDCCCXVII-MDCCCXX jussu et auspiciis

Maximiliani Josephi I. Bavariae regis augistissimi peracto...(fasc. 1) pp. 1-

60, illustrated.

Regenfuss, H. 1975. Die Antennen-Putzeinrichtung der Adephaga (Coleoptera),

parallele evolutive Verwollkammung einer komplexer Struktur. Zeitschrift fur

zoologische Systematik und Evolutions-forschung, 13: 278-299.

Reichardt, H. 1977. A synopsis of the genera of Neotropical Carabidae.

Quaestiones Entomologicae, 13 (4): 347-485.

Roach, B., K. R. Dodge, D. J. Aneshansley, D. Wiemer, J. Meinwald and T. Eisner.

1979. Chemistry of defensive secretions of ozaenine and paussine

bombardier beetles (Coleoptera: Carabidae). The Coleopterists Bulletin, 33

(1): 17-19.

Rosen, D. E. 1985. Geological hierarchies and biogeographic congruence in the

Caribbean. Annals of the Missouri Botanical Garden, 72: 636-659.

Savage, J. M. 1982. The enigma of the Central American herpetofauna: dispersal or

vicariance? Ibid, 69: 464-547.

Schaum, H. R. 1856-1860. Erste Abtheilung Coleoptera Erster Band, 791 pp. In,

Naturgeschichte der Insecten Deustchlands (W. F. Erichson). Nicolaische

Verlagsbuchandlung (G. Pathey), Berlin.

Schaum, H. R. 1864. Beitrage zur Kenntniss der Carabicinen. Berliner

Entomologische Zeitschrift, 8: 114-126.

Shpeley, D. and G. E. Ball. 1978. Anisocnemus, a Neotropical genus: classification

and geographical distribution (Coleoptera: Carabidae: Harpalini). The

Coleopterists Bulletin, 32 (2): 77-92.

Sloane, T. G. 1920. The Carabidae of Tasmania. Proceedings of the Linnaean

Society of New South Wales, 45: 1 13-178.

Sloane, T. G. 1923. The classification of the family Carabidae. Transactions of the

Royal Society of London, 1923, pp. 234-250.

Smith, M. R. 1951. Family Formicidae, pp. 778-875. In, Hymenoptera of America

north of Mexico (C. F. W. Musebeck, K. V. Krombein and H. K. Townes,

Editors). United States Department of Agriculture, Monograph No. 2.

Washington, D. C. 1420 pp.

Solier, A. J. J. 1836. Memoire sur quatre genres de la famille des carnassier-

terrestres. Annales de la Societe Entomologique de France, 5: 589-602.

Solier, A. J. J. 1849. Orden III. Coleopteros. In, Gay, Historia fisica y politica de

Chile. Vol. 4, pp. 105-380, 414-511.

Stork, N. E. 1980. A scanning electron microscope study of tarsal adhesive setae in

the Coleoptera. Zoological Journal of the Linnean Society, 68: 173-306.

Stork, N. E. 1985. Dhanya, a South-east Asian genus of ozaenine ground beetles.

Journal of Natural History, 19: 1113-1138.

Quaest. Ent., 1990, 26(1)

1 14

Ball and McCleve

Thompson, R. G. 1979. Larvae of North American Carabidae with a key to the

tribes, pp. 209-291. In , Carabid beetles: their evolution, natural history and

classification. ..(T. L. Erwin, G. E. Ball, D. R. Whitehead, and A. L. Halpem,

Editors). W. Junk b.v., Publishers, The Hague. X + 635 pp.

Thomson, J. 1856. Description de quelques Coleopteres nouveaux ou peu connus.

Annales de la Societe entomologique de France (3) 4: 317-335, 2 colored

plates.

Ward, R. D. 1979. Metathoracic wing structures as phylogenetic indicators in the

Adephaga, pp. 181-191. In , Carabid beetles, etc. For complete citation, see

Thompson, R. B., above.

Winkler, A. Catalogus Coleopterorum regionis palaearcticae, Pars 1, pp. 1-112.

Albert Winkler, Wien. 1924-1932. viii + 1698 pp.

Zacharuck, R. Y. 1985. Antennae and sensilla,pp. 1-69. In, Comprehensive insect

physiology, biochemistry and pharmacology. Volume 6. (G. A. Kerkut and L. I.

Gilbert, Editors). Pergamon Press, Oxford, xvi + 710 pp.

Quae st. Ent., 1990, 26(1)

_

Middle American Genera of the Tribe Ozaenini

115

INDEX TO NAMES OF TAXA

(Junior synonyms in italics)

Family Group Taxa

Bipartis 37

Brachinides 37

Brachinitae 39

Carabidae 38, 40, 80

Carabina 38

Carabini 37

Cicindisini 38, 40

Elaphrini 37-38

Enceladini 38

Eustrini 40

Harpalinae 38

Hiletini 38

Loricerini 37-38

Metriinae 38

Metriini 35, 38-41, 80, 103

Metriitae 39

Migadopini 38

Mystropominae 37

Mystropomini 37-38

Mystropomitae 37

Nototylini 38, 40

Omophronini 37-38

Ozaenidae 37, 40

Ozaenina 35, 37, 40

Ozaeninae 37, 39

Ozaenini 38-41, 48, 76-77, 80-81,

103

Ozaenitae 40

Ozenides 37

Pachytelini 40

Panagaeini 38

Paussidae 41

Paussina 40

Paussinae 37, 39

Paussini 38-41

Paussitae 37, 39

Promecognathini 37-38

Protopaussini 38-41

Scaritides 37

Scaritini 38

Septistemia 39

Siagonini 38

Genera and Subgenera

Afrozaena Jeannel 40, 89

Anentmetus Andrewes 84

Atta Fabricius 80, 86-90

Brachinus Weber 40

Catapiesis Brulle 37

Dhanya Andrewes 40, 53, 69-70,

75

Entomoantyx, new genus 41, 45-46,

48, 52-53, 58, 69-71, 75-77,

80, 82, 84, 103

Eustra Schmidt-Goebel 40

Goniotropis Gray 52-53, 70-71,

75-76, 88, 90, 92, 94, 103

Ictinus Castelnau-Laporte 98

Itamus Schmidt-Goebel 40

Melisodera Westwood 37

Metrius Eschscholtz 35, 40-41,

45-46, 48, 52-3, 58, 69-71,

74-77, 80-81, 83, 103, 106

Morions Latreille 37

Mystropomus Chaudoir 41, 53, 84

Nomius Castelnau 37

Ozaena Bates 101

Ozaena Banninger 101

Ozaena Olivier 35, 37, 41, 46, 48,

52-53, 58, 69-71, 75-77, 80,

98-99, 101-103

Ozena Chenu 98

Paussus Linnaeus 40

Pachy teles Perty 38, 41, 45-46,

52-53, 58, 69, 74 76, 80-82,

88-89, 92, 94, 97, 103

Pachyteles (sensu lato) 80, 93-94

Pachyteles (sensu stricto) 52-53,

58, 70-71, 75, 77, 90, 94, 97

Physea Brulle 41, 45^16, 48, 52-

53, 58, 69-71, 74-78, 80-81,

84, 86, 103

Physeomorpha Ogueta 41, 77, 80,

86

Platycerozaena Banninger 41, 45-

46, 48, 52-53, 58, 69-71, 74-

78, 80, 99, 101-103

Pseudozaena (sensu stricto). 40

Pseudozaena Castelnau 40

Scythropasus Chaudoir 88, 90

Sphaerostylus Chaudoir 40

Trachelizus Brulle 84

Trachelyzus Chenu 84

Trachypachus Motschulsky 48

Tropopsis Sober 82, 88

Quaest. Ent., 1990, 26(1)

116

Ball and McCleve

Species and Subspecies

beyeri Banninger, Goniotropis 9 1

beyeri Notman, Pachyteles 91

braziliensis Gray, Goniotropis 90

brevicornis (Bates), Platycerozaena

34

contractus Eschscholtz, Metrius 33,

42-44, 46, 49, 54-55

cyanipennis (Chaudoir), Entomoan-

tyx 42-44, 46, 49, 54-56, 61,

64, 66, 68, 82, 84

cyanipennis (Chaudoir), Tropopsis

84

cyanipennis Chaudoir, Ozaena 84

cyanipennis Chaudoir, Pachyteles

84

cyanoptera (Thompson), Pachyteles

84

cyanoptera Thompson, Ozaena 84

dentipes Olivier, Ozaena 38

elevata Ball, Ozaena 32, 100

elevata Banninger, Ozaena 104

elongata Chaudoir, Scythropasus 93

elongatus (Chaudoir), Pachyteles

63-64, 90, 93

elongatus Chaudoir) Scythropasus

93

enischnus, new species, Pachyteles

34, 62, 68, 96, 104

filiformis Chaudoir, Pachyteles 97

gyllenhali (Dejean), Pachyteles 34,

62, 66, 75, 94, 104

gyllenhali Dejean, Ozaena 94

halffteri Ogueta, Ozaena 100

hirta LeConte, Physea 42-44, 46,

49-50, 54-55, 61, 64, 66, 86-

87, 104

kuntzeni (Banninger), Pachyteles 33,

42^14, 46, 50, 55-56, 62, 64, 80,

91-93, 100, 104, 106

latipes Schaum, Physea 68, 86-87,

104

lemoulti Banninger, Ozaena 34, 42-

44, 46, 50, 54-56, 63, 66, 75,

99-100, 104

magna (Bates), Platycerozaena 66

marginicollis Solier, Tropopsis 82

marginicollis, (Solier), Pachyteles

66

mexicana Chaudoir, Ozaena 97

mexicanus (Chaudoir), Pachyteles

62, 66, 97

panamensis (Bates), Platycerozaena

42-44, 46, 50, 54-56, 63, 66

parca LeConte, Pachyteles 33, 54-

56, 62, 64, 68, 90, 93, 104, 106

rufus Brulle, Trachelizus 85

setosa Chaudoir, Physea 87

striola Perty, Pachyteles 42-44, 46,

49, 54-56

tenebrioides Castelnau-Laporte,

Ictinus 98

testaceus Horn, Pachyteles 94

testudinea Klug, Physea 85

verticalis (Chaudoir), Pachyteles 94

verticalis Chaudoir, Ozaena 94

Quaest. Ent., 1990, 26(1)

BOOK REVIEW

MANUAL OF NEARCTIC DIPTERA. Volume 3. J. F. McAlpine and D.M. Wood

(editors). 1989. Minister of Supply and Services Canada, vi + pp. 1333-

1581. $75.95 (Canada), US $91.15 (outside Canada).

This third (and last) volume of the Manual of Nearctic Diptera contains

chapters 114-116 on the phylogeny and classification of subgroups of the Diptera,

together with two pages of corrections to volumes 1 and 2 and a very long

composite index of the taxonomic names of Diptera and morphological terms used

in all three volumes. This third volume is mainly illustrated by phylogeny diagrams

(the reader being referred to the previous volumes for morphological illustrations),

but a few important new illustrations of larval head structure illustrate chapter 1 14.

Until publication of this work the most comprehensive modem reviews of the

phylogeny of the Diptera were contained in various works of B. B. Rohdendorf

and W. Hennig, both of whom died in the seventies. The work of both authors has

certain defects. Rohdendorf s system contained much that was arbitrary and not in

accordance with cladistic methodology (which he rejected). Hennig pioneered

cladistic methods, but in his later years wrote too quickly without sufficient

morphological studies; as a result his later works contain much that is superficial or

erroneous. The authors of Manual 3 take Hennig's work as the main starting point

of their studies, but find much that is in need of revision. I agree that extensive

revision of Hennig's system is needed, and welcome the publication of this work

which will hopefully stimulate renewed interest in this field of fundamental

importance to the whole of dipterology.

Of course this is not to say that I agree with everything proposed, and the

authors themselves recognize that there are many areas of the system where only

tentative proposals can be offered. In this review I will comment on the validity of

new proposals to the extent that this is possible on the basis of existing literature

and my previous studies in this field. But there are too many new observations for

me to try and check them against insects in the context of writing a review.

Evaluation of some new proposals will require a period of years, as the literature

develops.

Before discussing each chapter, there are two points of a formal nomenclatural

nature and one of a methodological nature which need comment.

Some of the superfamily names used in this work, especially within the

Schizophora, are contrary to longstanding usage, e.g. Sciaroidea (for

Mycetophiloidea), Ephydroidea (for Drosophiloidea), Carnoidea (for

Chloropoidea), Nerioidea (for Micropezoidea) and Oestroidea (for Tachinoidea).

It is explained that this is an application of Article 36(a) of the 1985 International

Code of Zoological Nomenclature, in which it is stated that "a name established for

a taxon at any rank in the family group is deemed to be simultaneously established

with the same author and date ... at other ranks in the family group". Thus, the

oldest family-group name is used, even if this was previously used only as a family,

not as a superfamily, name. It remains to be seen whether other dipterists will

accept these changes, or whether reference will be made to the Commission to

conserve long-accepted superfamily names. The question of priority of

superfamily names was given scant consideration in previous literature, since the

dates of first proposal were not known for many of the older family-group names.

This situation has changed as a result of bibliographic studies by C. W. Sabrosky,

who gave advice to the authors on this matter. If we are to strictly apply the

118

Book Review

priority principle to superfamily names in the future, it is essential that Sabrosky's

work be published.

A second formal question concerns the formation of names for higher taxa

(above the family group). Such names are not regulated by the Code. The authors

of the Manual have made wide use of the suffix -morpha attached to the root of

generic names (following the precedent of Rohdendorf), even in cases where other

names are well established and have priority (e.g. Muscomorpha instead of

Cyclorrhapha). In my opinion this extension of the principle upon which family-

group names are formed to higher levels is misguided, and will cause instability

because changes in suffices have to be made whenever new research causes the

relative ranking of groups to be revised. Therefore I intend to continue to use

names formed upon other principles when these have priority, and to apply names

formed by adding suffices to generic roots in their original senses irrespective of

ranking changes. In this connection it should be noted that names formed with the

suffix -formia have priority over many of those formed with the suffix -morpha.

I have found one difficulty regarding the phylogeny diagrams in this volume.

All diagrams are of the type in which characters assumed to be autapomorphies are

ascribed to each branch of a dendrogram. This type of illustration is of course

widely used and valid. However, there is a difficulty when many of the characters

used are subject to homoplasy and the interpretation of their distribution is

debatable. In such cases it is necessary to know the complete distribution of the

characters in order to be able to judge the validity of proposed groupings. This is

especially a problem in the treatment of "Acalyptratae", in which many groupings

are based solely on characters known to be subject to homoplasy. I recommend that

in future treatments diagrams with bars across showing the total distribution of

characters should also be included, so that readers can retrieve this information

directly from the illustration without the need to search the text and other literature

for information on the wider distribution of characters.

Chapter 114. Phylogeny and Classification of the Nematocera (by

D. M. Wood and A. Borkent)

This chapter includes discussion of the origin of the Diptera, as well as the

relationships between the groups traditionally included in the "Nematocera"

(probably a paraphyletic grouping). A system of seven infraorders is proposed

(Tipulomorpha, Blephariceromorpha, Axymyiomorpha, Bibionomorpha,

Psychodomorpha, Ptychopteromorpha and Culicomorpha). The major innovation

of this system in comparison with Hennig's (1973) treatment is the new concept of

Psychodomorpha, containing the Trichoceridae (removed from Tipulomorpha) and

four families removed from Bibionomorpha, the Perissommatidae, Anisopodidae,

Scatopsidae and Synneuridae (the last should be called Canthyloscelididae on

grounds of priority). Groups included in Psychodomorpha by Hennig (1973) but

removed by Wood & Borkent are the Blephariceridae, Deuterophlebiidae and

Nymphomyiidae (grouped as infraorder Blephariceromorpha) and the

Ptychopteridae and Tanyderidae (grouped as infraorder Ptychopteromorpha). The

enigmatic family Axymyiidae (formerly in Bibionomorpha) is also segregated as

the new infraorder Axymyiomorpha.

In the discussion of the origin of the Diptera, Wood and Borkent advance the

hypothesis that the Nannochoristidae are the sister-group of the Diptera +

Siphonaptera despite certain contrary evidence. I do not find the evidence they

offer for regarding the Siphonaptera as the sister-group of the Diptera (larval

thoracic legs absent, pupal mandibles immovable) convincing, as these characters

are subject to homoplasy. There is a series of characters which suggests that the

Quaest. Ent., 1990, 26(1)

-

Book Review

119

Siphonaptera are more closely related to the Mecoptera, especially the structure of

the spermatozoa (see Christensen 1975, 1981). The question of the relationships

of the Nannochoristidae is addressed in greater detail in a new work by Willmann

(1989), which also redescribes and reinterprets Mesozoic fossils relevant to the

origin of the Diptera. Willmann's outstanding work will obviously provide the

main starting point for future investigations of the origin and relationships of the

Diptera as well as Mecoptera. Willmann treats the Nannochoristidae as the sister-

group of all other Mecoptera in the recent fauna, while leaving open the question of

the position of the Siphonaptera within the Antliophora (= Diptera + Mecoptera +

Siphonaptera).

Although Wood & Borkent's opinion that the Nannochoristidae should be

removed from the Mecoptera because they are more closely related to the

Siphonaptera + Diptera seems unlikely, it should be noted that use of the

Nannochoristidae as an outgroup for assessing character polarity in the Diptera is

not in contention. The Nannochoristidae are recognized by Willmann and other

mecopterists as the relatively plesiomorphous subgroup of the Mecoptera in most

respects. They remain an important basis of outgroup comparison in studies of

relationships within the Diptera and Siphonaptera, irrespective of what view is

taken of their position within the Antliophora.

The changes in the content of the infraorders of "Nematocera" proposed by

Wood & Borkent in my opinion represent a considerable advance over the systems

proposed by Hennig and Rohdendorf. Their work should provide one of the main

starting points for further investigations in this field. My main criticism is that the

changes do not go far enough.

The new concept of Psychodomorpha is justified by Wood & Borkent on the

basis of a complex of characters (nos. 38-42) of the larval head (labrum with

posteriorly pointed hairs, "premandibles" dentate or pectinate, torma articulating

with dorsal labral sclerite, mandible moving in nearly vertical plane and striking

hypostoma, mandible chela-shaped). I support the view that this character complex

is apomorphous and characterizes the groundplan of a major subgroup of the

Diptera. However, I think that the group to whose groundplan these characters

belong may be more extensive than Wood & Borkent's Psychodomorpha. There

are grounds for suggesting that the larval head structures of the groups called

Tipulomorpha, Ptychopteromorpha and Culicomorpha by Wood & Borkent

represent further modifications of the same groundplan condition.

Wood & Borkent show the Tipulomorpha (Tipulidae s.l.) as the sister-group

of all other "Nematocera" on their phylogeny diagram, the grouping of all other

Nematocera being supported by the apomorphous state of character 1 (prostheca

arising directly from median surface of larval mandible). The opposing state

(mandible with prostheca on articulated lobe) is assumed to belong to the

groundplan of the Tipulomorpha (and of the Diptera as a whole) on the basis of

outgroup comparison with Nannochoristidae. I am sceptical whether the presence

of an articulated prosthecal lobe in Tipula is a genuinely plesiomorphous character.

The larval head capsule of Tipula is of a highly modified type with the posterior

margins of the capsule strongly indented. The interpretation that an articulated

prosthecal lobe belongs to the groundplan of the Tipulomorpha will only be

convincing, if it is also shown to be present in other groups of Tipulomorpha with

less modified head capsule. I favour a quite different interpretation of the

relationships of the Tipulidae s.l.

It appears to me that the apparent synapomorphies between Tipulidae (s.l .)

and Trichoceridae (especially reduction of the male cerci, development of

gonopods from posterolateral zones of proliferation, female cerci with single

Quae st. Ent., 1990, 26(1)

120

Book Review

article, only 3 branches of radial sector reaching wing margin, forwards

displacement of distal section of m\+2) cannot be dismissed as due to homoplasy.

The view that the Trichoceridae and Tipulidae s.l . are sister-groups is reconcilable

with Wood & Borkent's justified emphasis on the synapomorphies between the

larval head structure of Trichoceridae and that of other Psychodomorpha, if we

assume that the considerable diversity of head structure shown by the larvae of

Tipulidae s.l . represents a transformation series from a groundplan structure similar

to that of Trichocera. On this interpretation the horizontal plane of movement of the

mandibles of some Tipulidae s.l. is assumed to be secondary. Final resolution of

how the polarity of characters of the larval head structure in Tipulidae s.l. should

be interpreted must obviously await more detailed comparative morphological

studies than we presently have available. Meanwhile, readers should note that

Wood & Borkent's interpretation of the position of the Tipulidae s.l . as the sister-

group of all other "Nematocera" is poorly supported, and other interpretations are

possible. I would place the Tipulidae s.l. within their Psychodomorpha as the

sister-group of the Trichoceridae.

The grouping of Ptychopteridae and Tanyderidae (as Ptychopteromorpha)

following Hennig (1973) is supported only by character 52 (male tarsal claws

folding against basal swelling on tarsomere 4), a character found in the Tanyderidae

and Ptychoptera (but not in other Ptychopteridae). Whether this feature belongs

to the groundplan of the Ptychopteridae is doubtful, since it has not been found in

any of the Mesozoic Ptychopteroidea (information from N. S. Kalugina). In my

opinion the relationships of the Ptychopteroidea (Ptychopteridae and related

fossil groups) and Tanyderidae should be considered separately. Both groups are

archaic, represented in the earliest Mesozoic fossil assemblages. In their

discussion Wood & Borkent present new evidence that the Ptychopteridae alone

may be the sister-group of the Culicomorpha based on the structure of the larval

labrum and mandibles. This seems to me more convincing evidence of the

relationships of the Ptychopteridae than the dubious tarsal character. At the same

time we must not lose sight of the fact that the larval head structures of the

Tanyderidae, Ptychopteridae and Culicomorpha may be derived from the same

groundplan structure as that postulated for the groundplan of the Psychodomorpha.

Wood & Borkent do in fact unite these groups at a higher level on their phylogeny

diagram, but do not name the more inclusive group.

The inclusion of the Scatopsoidea (Scatopsidae + Canthyloscelididae) in the

Psychodomorpha on the basis of larval head structure seems to me fully justified.

Wood & Borkent place this group as the sister-group of the Anisopodidae.

However, I am aware of one unique character which suggests that they may be the

sister-group of the Psychodidae. This is that the 8th pair of abdominal (the larval

hind) spiracles lacking in most adult Diptera persist in the adult male but are

displaced to a dorsal position within the 9th tergite (epandrium). Further

investigation of the position of the Scatopsoidea within the Psychodomorpha is

needed. The transference of the Perissommatidae to the Psychodomorpha based on

new observations of the larval head structure also appears fully justified.

Krivosheina (1988) has also recently examined the larvae of Perissommatidae, and

concludes that they "have characters relating them variously with the Scatopsidae,

Trichoceridae, and to a considerable degree with the Psychodidae" ( i.e . with

families included by Wood & Borkent in the Psychodomorpha). It seems that there

was no contact between Krivosheina and Wood & Borkent, but both reached

similar conclusions independently.

Wood & Borkent expand the concept of Blephariceromorpha (Blephariceridae

+ Deuterophlebiidae) to include also the Nymphomyiidae. This is controversial.

Quaest. Ent., 1990, 26(1)

Book Review

121

and rests on interpreting the abdominal prolegs of larval Deuterophlebiidae and

Nymphomyiidae as of common origin and belonging to the groundplan of the

Blephariceromorpha. Unfortunately, the development of prolegs is subject to much

homoplasy in Diptera, so I do not have confidence in this character in isolation. In

Rohdendorf's (1964) system the Nymphomyiidae (as Archidiptera) were

considered the sister-group of all other recent Diptera, a view which is best

justified by the primitive structure of the adult nervous system (retaining 8

separate ganglia, as in larvae). Wood & Borkent argue that this character may be

neotenous, a possibility which certainly cannot be excluded. The position of the

Nymphomyiidae remains in doubt. The two interpretations currently held are both

essentially based on the distribution of single characters. Further morphological

studies are needed, so that additional evidence can be brought to bear on the

problem.

The treatment of the Bibionomorpha is a weak part of Wood & Borkent's

work, and no constitutive (autapomorphous) characters of this group are

suggested. After removal of the heterogenous elements included by previous

authors, the Bibionomorpha in Wood & Borkent's sense consists of two certainly

monophyletic groups: the Pachyneuroidea + Bibionoidea (which I would

amalgamate) and the Sciaroidea. These groups have been closely associated in all

recent systems and may well be monophyletic, but a critical assessment is still

impeded by lack of sufficiently detailed studies of primitive Sciaroidea

(especially Ditomyiidae). The monophyly of the Pachyneuroidea + Bibionoidea is

demonstrated by the synapomorphous structure of the larval labium and

hypopharynx, which Wood & Borkent do not discuss.

The recognition of Axymyiidae as a group of high rank (Axymyiomorpha) is

probably the best treatment on present information, since there are no convincing

grounds for including the group within any other infraorder. Krivosheina (1989),

who has made special studies of this group, has also accepted the concept of

Axymyiomorpha.

The strongest part of Wood & Borkent's work is no doubt the treatment of

the Culicomorpha, a group on which both authors have worked for many years. I

agree with them that the content of this group is no longer contentious. Their

discussion is authoritative and will provide a sound basis for future studies.

Chapter 115. Phylogeny and classification of the "Orthorrha-

phous" Brachycera (By N.E. Woodley)

This chapter treats all Brachycera except the Cyclorrhapha (= Muscoidea in

the sense of this chapter, Muscomorpha in the sense of chapter 116). Four

infraorders are recognized, the Xylophagomorpha, Stratiomyomorpha,

Tabanomorpha and Muscomorpha (in a new wide sense, different from the usage in

chapter 116). The different usages of the same names in the two chapters is

confusing, and well illustrates how the use of suffices to denote relative ranking

causes instability.

Woodley's discussion is generally sound, and I do not find much to disagree

with. But I think that his Xylophagomorpha and Tabanomorpha can be combined at

infraordinal level, thus reducing the number of infraorders to three. I base this

suggestion on the structure of the male genitalia in the Coenomyiinae

(Xylophagidae), which agrees substantially with that of Rhagionidae

(Tabanomorpha). I infer from the work of Nagatomi (1984) that there is a major

subgroup of the Brachycera corresponding to the Xylophagomorpha +

Tabanomorpha sensu Woodley characterized by fusion of the ejaculatory apodeme

Quae st. Ent., 1990, 26(1)

122

Book Review

with the base of the aedeagus and sheathing of the aedeagus and gonites ("tines")

by a dorsal and pair of lateral/ventral processes.

The Stratiomyomorpha should probably also include the Pantophthalmidae,

listed by Woodley as incertae sedis. The distal parts of the powerful mouthhooks

of pantophthalmid larvae bear palpi, indicating that they are of maxillary origin as

in the Stratiomyidae and Xylomyidae. Also the structure of the male genitalia

described by Nagatomi (1984) is incompatible with inclusion of this family in the

Xylophagomorpha + Tabanomorpha.

Regarding the genera Exeretoneura and Heterostomus, also listed as incertae

sedis by Woodley, Nagatomi's descriptions of the male genitalia suggest that both

belong somewhere in the Xylophagomorpha + Tabanomorpha. Woodley's

placement in the vicinity of Xylophagidae seems appropriate pending further

studies.

More enigmatic is the position of the Vermileonidae, listed by Woodley as a

family incertae sedis within the Tabanomorpha. I agree with Woodley that this

family cannot belong in the Asiloidea, where it was placed by Teskey in Volume 1

of this Manual. If Kovalev (unpublished MSS) is correct in referring the lower

Jurassic Protobrachyceron (the earliest described fossil brachyceron) to this

family, then it is possible that it merits higher rank in the system. Woodley's

proposal to place the Vermileonidae provisionally in the Tabanomorpha seems

reasonable pending further studies.

The infraorder Muscomorpha is proposed by Woodley in a new sense,

inclusive of the Nemestrinoidea, Asiloidea, Empidoidea and Muscoidea. The

concept seems to me well justified, but not the nomenclature. The numerous

different senses in which the name Muscomorpha has been used cause confusion. A

new name would have been preferable. The superfamilies Asiloidea, Empidoidea

and Muscoidea are ranked more highly by most other authors (including McAlpine

in chapter 1 16), who hence use other suffices if they believe in forming names of

higher taxa in this way. I recommend forgetting about rank and suffices and using

the earliest appropriate names (Pleroneura, Orthogenya and Cyclorrhapha). These

names can remain applied to the same groups, irrespective of different relative

ranking by different authors.

Woodley (correctly in my opinion) restricts the concept of Nemestrinoidea to

the Nemestrinidae + Acroceridae, referring the Bombyliidae to the Pleroneura

("Asiloidea"). Some previous authors, including Hennig (1973), placed the latter

family in the Nemestrinoidea on account of the hypermetamorphic larval

development, but subsequent morphological studies leave no doubt that it belongs

to the Pleroneura. Woodley considers that all Muscomorpha except

Nemestrinoidea form a monophyletic group characterized by a setiform empodium.

I agree with this view, and suggest that we follow the precedent of Lameere (1906)

in applying the name Heterodactyla to this group.

In his phylogeny diagram Woodley shows a trichotomous subdivision of the

Heterodactyla into Asiloidea (Pleroneura), Empidoidea (Orthogenya) and

Muscoidea (Cyclorrhapha). This fence-sitting on the issue of the validity of the

concept Eremoneura (Orthogenya + Cyclorrhapha) presumably reflects the fact that

he has not worked personally on these groups and does not wish to involve himself

in controversy. At the end of the chapter he quotes verbatim the characterization of

the Eremoneura which I gave in 1984, with the disclaimer that "I cannot personally

evaluate these characters and their distributions within the Brachycera". I suggest

that these characters and their distributions have already been evaluated, and that

the grounds for grouping the Orthogenya with the Cyclorrhapha are in fact

overwhelming. This grouping is indicated by numerous autapomorphies involving

Quae st. Ent., 1990, 26(1)

Book Review

123

the structure of the male and female terminalia, the wing venation and the

chaetotaxy, probably also by the larval head structure (but interpretation of the

character sequence in larvae remains problematical due to inadequate information on

the larvae of Orthogenya). Woodley offers two characters in support of the

traditional grouping of the Orthogenya with the Pleroneura; presence of three

antennal flagellomeres, and presence of acanthophorites (spinous halves of 10th

tergite) in the female. In my opinion neither of these characters provides reliable

evidence for such a relationship. Presence of three (as in most Orthogenya) or four

(as in most Cyclorrhapha) flagellomeres does not indicate that the aristate antennae

in these groups originated independently. Several cases are now known in which

the change from a 3- to 4-articled flagellum or vice versa must have occurred.

Within the Orthogenya, 4-articled flagella are certainly known in Dryodromia and

Meghyperus ; within the Cyclorrhapha 3-articled flagella are known in Opetia (the

probable sister-group of all other Cyclorrhapha) and in one subgroup of Diopsidae.

Thus there is no fundamental difference between the aristate antennae of Orthogenya

and Cyclorrhapha. Development of spines on the female 10th tergite is also a

character prone to homoplasy. For instance, my studies indicate that such spines

have evolved several times within the family Anthomyiidae alone. So even if more

complete information causes us to revise the prevailing opinion that

acanthophorites do not belong to the groundplan of the Orthogenya, I fail to see

how this could cast doubt on the validity of the concept of Eremoneura. In this

connection I draw attention to the recent thesis by Wiegmann (1989), who sees the

problem not as determining whether the Eremoneura are monophyletic but whether

the Orthogenya are monophyletic or paraphyletic with respect to the Cyclorrhapha.

Chapter 116. Phylogeny and Classification of the Muscomorpha

(by J. F. McAlpine)

The Muscomorpha in the sense of this chapter is the group normally called

Cyclorrhapha, here ranked as an infraorder (both naming and ranking being

inconsistent with the previous chapter). In order to avoid confusion I shall refer to

this group as the Cyclorrhapha. This chapter is the longest of the three, and sets out

J. F. McAlpine's views on the origins of the Cyclorrhapha and the relationships

between included families in far more detail than previously available. I welcome

its publication as a major contribution to this field, although I do not accept the

author's views regarding the origin of the Cyclorrhapha and homologization of the

male genitalia and proctiger. McAlpine's system of superfamilies and families

contains a variety of differences from previous treatments by me (Griffiths [1972])

and by Hennig (1973, with modifications in subsequent papers). Some of the

changes are clearly justified, a few seem to me retrograde.

I included extensive comments on McAlpine's interpretation of the homologies

of the male genitalia and proctiger in my review of Volume 1 of this Manual

(Griffiths 1981). His views seem little changed, so the criticisms remain. But it

does not seem necessary to occupy space in this review by repeating them.

Readers who want a summary of my views may refer to that review and also to my

characterization of the Eremoneura (Griffiths [1984]) quoted at the end of Chapter

115. Of course the description of some of the characters used by McAlpine to

characterize the groundplan of the Cyclorrhapha is affected by how homologies are

interpreted, and I would describe some of the genitalia characters quite differently.

Despite disagreements over the interpretation of certain characters, I recognize

McAlpine's long and detailed review of the groundplan characters of the

Cyclorrhapha as an important contribution and certainly justifying his conclusion

Quae st. Ent., 1990, 26(1)

124

Book Review

that the monophyly of this group is "one of the best substantiated and most

universally accepted assumptions in the phylogeny of the Diptera".

McAlpine goes on to present an interesting argument that the Cyclorrhapha are

more closely related to Stratiomyoidea (= Stratiomyomorpha in the sense of

chapter 115) than to the Orthogenya or Pleroneura. In my opinion this cannot be

correct in view of the overwhelming evidence for the monophyly of the Eremoneura

(Orthogenya + Cyclorrhapha). I offer the following comments on the list of 12

suggested synapomorphies: The first five characters involve alleged

synapomorphies in the larval head structure and feeding mechanism. However,

Schremmer (1951) established that the mouthparts of larval Stratiomyoidea are of a

fundamentally different type from those of all other Brachycera (including

Cyclorrhapha) with the palpus-bearing part of the maxilla involved in the formation

of the mouthhooks. Characters 8 and 9 can be dismissed also. The presence of a

cone-shaped condyle inserted into the base of the first flagellomere is not a

groundplan character of the Cyclorrhapha. Nor is the first flagellomere of

Cyclorrhapha of composite origin as in some Stratiomyoidea. Characters 7, 10 and

1 1 are inconclusive, since not confined to the Stratiomyoidea and Cyclorrhapha.

There remain only two characters (6 and 12): formation of a puparium, and male

with ejaculatory apodeme free from aedeagus and body wall. The first feature is

found only in Stratiomyidae among the Stratiomyoidea, where it is normally

assumed to have evolved independently of the Cyclorrhapha. The ejaculatory

apodeme character requires clarification. The ejaculatory apodeme is primitively a

separate sclerite in Diptera, but connected by muscles to some part of the outer

body wall around the base of the aedeagus. The apomorphous modification in

Cyclorrhapha is that the muscles on this apodeme connect only to the walls of the

ejaculatory bulb, which has allowed the apodeme to become withdrawn from the

base of the aedeagus. A similar modification is reported to have occurred in

Stratiomyidae, but no detailed morphological description is available. Nor is it

known whether a free ejaculatory apodeme belongs to the groundplan of the

Stratiomyoidea. Even if the condition in Stratiomyidae proves to be the same as

in Cyclorrhapha, I think that homoplastic modification will have to be assumed in

view of the extensive evidence for the monophyly of the Eremoneura. A free

ejaculatory apodeme also has evolved independently in some Scatopsidae.

McAlpine accepts the traditional division between Aschiza and Schizophora

as the primary subdivision of the Cyclorrhapha. The recent suggestion that Opetia

(Opetiidae) may be the sister-group of all other Cyclorrhapha unfortunately is not

discussed. Although the larvae of Opetia are unknown, Wiegmann (1989) points

out that the lack of pupal muscle plaques on the adult abdomen indicates that the

pupa is contained within a puparium. Thus there seems no possibility that Opetia

is misplaced in the Cyclorrhapha. Either it is the sister-group of all other

Cyclorrhapha, or its lack of hypopygial rotation is secondary. Presumably

McAlpine holds the latter opinion, since he lists Opetiidae as a synonym of

Platypezidae.

If we accept that Opetia is probably the sister-group of all other

Cyclorrhapha, the question arises whether the Aschiza exclusive of Opetia form a

monophyletic group. The evidence in favour of this interpretation according to

McAlpine's extensive tabulation of the "character states in ground plans of Aschiza

and Schizophora" is fusion of the larval hypopharyngeal and tentopharyngeal

sclerites and enlargement of the pupal respiratory horns. On the other hand it may

be pointed out that the immature stages of some families of Aschiza (especially

Platypezidae) are poorly studied, so the existence of these synapomorphies needs

confirmation. An equally plausible hypothesis is that the Platypezidae are the

Quae st. Ent., 1990, 26(1)

Book Review

125

sister-group of all other Cyclorrhapha (exclusive of Opetia ), since only in

platypezids is hypopygial rotation partly reversible. Pending further studies I

think we should keep an open mind.

It is interesting that McAlpine regards the Lonchopteridae as the sister-group

of the group usually called Hypocera or Phoridea (Ironomyiidae +

Sciadoceridae/Phoridae). Formerly (Griffiths 1972) I followed the opinion that the

Lonchopteridae (= Acroptera, Anatriata) are the sister-group of all other

Cyclorrhapha (Atriata), but now regard this as improbable. The sole apomorphous

character upon which McAlpine bases his grouping of the Lonchopteridae with the

Phoridea is the dichoptic condition in males. However, the apparently

synapomorphous structure of the male postabdomen (loss of 7th tergite, 7th

stemite and inverted 8th tergite) also supports this grouping. These sclerites are

present in the groundplan of all other subgroups of Cyclorrhapha except Opetia .

The position of the cleavage lines on the puparium may also represent a

synapomorphy of the Lonchopteridae and Phoridea. McAlpine interprets the

pattern in Lonchopteridae as closest to the groundplan of the Cyclorrhapha, but it

is more parsimonious in terms of his phylogeny diagram to interpret is as

apomorphous. In all other Cyclorrhapha the operculum which is broken off when the

adult emerges includes the dorsal half of the thoracic segments of the puparium.

McAlpine places the Platypezidae as the sister-group of the Lonchopteridae +

Phoridea (forming the superfamily Platypezoidea), but I regard this with

scepticism. He justifies this concept of Platypezoidea mainly on the basis of

chaetotaxy. However, the chaetotactic characters may not be apomorphous, since

the reduced chaetotaxy of the Syrphoidea is surely secondary. Some of the setae in

question (such as ocellar setae) belong to the groundplan of the Cyclorrhapha, if not

of the Eremoneura as a whole. The relationships between the Platypezidae and

other Cyclorrhapha remain controversial and in need of further study.

The treatment of the families of Schizophora is long and detailed. McAlpine

follows the traditional subdivision of this group into the Acalyptratae and

Calyptratae, rejecting my criticism of the former as a residual paraphyletic group

(Griffiths 1972). I have checked the apomorphous character states listed in the

extensive table of "character states in ground plans of Acalyptratae and

Calyptratae" and do not find the suggested apomorphies of the former convincing.

There is no groundplan difference in the development of the pupal respiratory horn,

which pierces the wall of the puparium in Heleomyzidae as well as in many

Calyptratae. Since there are several groups with holoptic male eyes also in the

Acalyptratae, it cannot be assumed that dichopticism belongs to the groundplan of

this group; and in any case the change from a holoptic to dichoptic condition is

very prone to homoplasy. Presutural dorsocentral and postsutural acrostichal setae

are both present in some Acalyptratae (e.g. Agromyzidae), so I am sceptical

whether the absence of these setae can be ascribed to the groundplan. Some other

chaetotactic characters suggested (lower surface of scutellum bare, latepimeron

bare, meron bare, laterotergite bare) seem to me trivial, and I cannot accept them as

significant without more information on their distribution. The relative size of the

lower calypter is prone to variation both in Acalyptratae and Calyptratae, and I

know of no basis in terms of outgroup comparison for inferring that possession of

a relatively large lower calypter is the groundplan condition of the Schizophora.

Only the two characters of the female reproductive system listed at the end of

McAlpine's table represent major structural differences. But I doubt that their

distribution validates the concept of Acalyptratae. Possession of a common duct

by 2 of the 3 spermathecae is in my opinion an autapomorphy of the Cyclorrhapha

exclusive of Opetia , not of the Acalyptratae alone, and certainly belongs to the

Quaest. Ent., 1990, 26(1)

126

Book Review

groundplan of the Calyptratae also (possession of 3 spermathecae with 3 separate

ducts in part of the Oestroidea being secondary). Information on the arrangement of

spermathecae in the families of "Aschiza" is meagre, but at least in some Phoridae

there are 3 spermathecae of which 2 share a common duct (information from

B. V. Brown). The question of the ventral receptacle requires further study. Such a

structure is certainly widespread in the "Acalyptratae", but the information on non-

sclerotized parts of the female reproductive system in other Cyclorrhapha seems to

me too meagre for us to determine whether or not homologous structures occur.

The basic subdivision of the Schizophora remains controversial. In my 1972

book I subdivided this group into 5 superfamilies (Lonchaeoidea, Lauxanioidea,

Drosophiloidea, Nothyboidea and Muscoidea) mainly on the basis of the structure

of the male postabdomen. Subsequently the description of Morgea (McAlpine

1981) has satisfied me that the Lonchaeoidea belong within what I called the

"Tephritidae family-group" (Tephritoidea in McAlpine's sense) within the

Muscoidea in my sense. McAlpine's remarks on the Lauxanioidea and the families I

included in the Nothyboidea indicate that he believes that the male postabdominal

structure of these groups too is derived from the muscoid type. I have not seen

some of the insects upon which his remarks are based, so refrain from agreeing or

disagreeing with him at this time. However, if it is confirmed that what I called the

muscoid type of postabdominal structure (with asymmetrically reduced 7th tergite)

is also basic to the Lauxanioidea and Nothyboidea, this would leave the

Ephydroidea (Drosophiloidea) as the sister-group of all other Schizophora. For

further discussion of the fundamental differences between the male postabdominal

structure and development in the Ephydroidea and Muscoidea (in my sense) see

pages 81-83 of that book (Griffiths 1972).

I have the following comments on the treatment of particular superfamilies of

"Acalyptratae". They have to be brief in order to contain the length of this review.

The treatment of the first superfamily, Nerioidea (= Micropezoidea) has my

support. This grouping has been accepted by all recent authors.

The "Diopsoidea" appear to me to be an assemblage of heterogenous long¬

bodied forms. For the Diopsidae and its close relatives Syringogastridae and

Centrioncidae, the new work of Feijen (1989) gives a more detailed and up-to-date

treatment. Feijen treats these families as monophyletic (grouped as the prefamily

Diopsioinea). The Megamerinidae are a possible sister-group of the Diopsioinea,

but the family is too poorly studied for a firm opinion to be given. Whether the

further relationships of the Megamerinidae and Diopsioinea are with the

Nothybidae, as McAlpine suggests, or with the Sciomyzoidea (as suggested in my

1972 book) should be addressed in future studies. Two other families included by

McAlpine in the "Diopsoidea", the Tanypezidae (including

Strongylophthalmyiidae) and Psilidae, have elongate ovipositors with partially

fused cerci and may belong to or be closely related to the Tephritoidea (see below).

McAlpine recognizes the Conopoidea (Conopidae s.l .) as closely related to

the Tephritoidea, in agreement with my opinion (Griffiths 1972).

The concept of Tephritoidea proposed by McAlpine is equivalent to the

Tephritidae family-group of my 1972 book with the addition of the Lonchaeidae.

This inclusion is justified, but I think that the Cryptochetidae should also be

included. McAlpine places the Cryptochetidae as the sister-group of the

Chloropidae, i.e. within the group Milichiidae + Chloropidae characterized by

reduced spermathecae with long fine ducts. This is wrong, because the female

reproductive system in Cryptochetidae is not of that type. On present information

I continue to regard the Cryptochetidae as probably derived from a lonchaeid-like

Quaest. Ent., 1990, 26(1)

,

Book Review

127

ancestor and would include them in Tephritoidea. But I agree with the exclusion of

Librella from the Cryptochetidae for the reasons given by McAlpine.

In connection with the Cryptochetidae, I was astonished to read on page 1406

that I mistook the female for the male terminalia of Cryptochetum nipponense

Tokunaga in my 1972 book. Evidently this is a reference to the unfounded claim

by D. K. McAlpine (1976) that both I and Okada (1956) had mistaken the sexes of

this species. For the record, the mistake was D. K. McAlpine's, as anyone can

confirm by reviewing other literature on this group. For instance, Pfennig's (1937)

figures of the male terminalia of C. buccatum Hendel are quite similar to my figures

of C. nipponense Tokunaga.

While McAlpine's concept of Tephritoidea is clearly an advance, I suggest

that there may be additional families which belong in this group: for instance, the

Camidae. The structure of Neomeoneurites Hennig (1972) casts doubt on the

traditional grouping of this family with the Milichiidae + Chloropidae. The

structure of both the male and female terminalia in Neomeoneurites (long coiled

aedeagus, extremely long ovipositor bearing fused cerci) indicate that the Camidae

belong to the Tephritoidea. Other groups which should be reviewed for possible

inclusion in the Tephritoidea on the basis of the structure of the male and female

terminalia are the Tanypezidae (including Strongylophthalmyiidae) and the

Psilidae.

McAlpine accepts the concept of Lauxanioidea introduced by Hennig (1958)

and followed by me (Griffiths 1972). His interpretation differs from mine at that

time in that he regards the male postabdominal structure of this group as derived

from what I called the muscoid type. This allows him to consider Cremifania (a

genus of typical muscoid structure which I removed to the Sciomyzoinea as family

Cremifaniidae) as a primitive chamaemyiid. He also claims to have evidence that

the postabdominal structure of Lauxaniidae is similarly derived.

As far as Cremifania is concerned, its position remains disputed.

Tanasiychuk (1986) did not accept it as a chamaemyiid. The undoubted

chamaemyiids with somewhat asymmetrical postabdomen ( Acrometopia and

Parochthiphila ) do not show the typical muscoid pattern of sclerites (asymmetrical

reduction of 7th tergite not demonstrable). I am not aware of any lauxaniid with an

asymmetrical postabdominal structure. If such exist, it would be helpful if the

species were stated so that McAlpine's arguments can be checked.

McAlpine's concept of Sciomyzoidea is close to that in other recent

treatments by Hennig and me. All recent authors include here the families included

by McAlpine. There is disagreement only regarding whether additional families,

such as Cremifaniidae and Megamerinidae, should also be included. As noted

above, if the Megamerinidae belong to the Sciomyzoidea, probably the

Diopsioinea should also be included here as a subordinate group. Since the

Megamerinidae are poorly studied, I note this possibility as one to be considered

in future studies, not as a firm opinion.

The concept of Opomyzoidea is new and unsatisfactory. That should not be

taken as a criticism, since the relationships of most of the included families have

been poorly studied and any superfamily arrangement at this time must be arbitrary

and tentative. The apomorphous characters given for the groundplan of the

Opomyzoidea all have wider distributions and may be subject to homoplasy.

Regarding the subgroups (suprafamilies sensu McAlpine) of Opomyzoidea, it

appears to me that the Opomyzoinea and Asteioinea are defensible groupings

which will serve as a good basis for further studies. But his Clusioinea and

Agromyzoinea seem to me most improbable groupings. The structure of the male

genitalia suggests that the Acartophthalmidae and Odiniidae belong in the vicinity

Quaest. Ent., 1990, 26(1)

128

Book Review

of the Tethinidae and other families included by McAlpine in the "Camoidea", as

they were treated in my 1972 book. I find McAlpine's argument that the characters

of the fossil Acartophthalmites demonstrate a relationship between

Acartophthalmidae and Clusiidae unconvincing, since the only synapomorphies

suggested between these families are in characters subject to homoplasy.

In connection with the comments on Agromyzidae, it should be noted that the

old report of sclerotized strips on the left side of the male abdomen was checked

many years ago and found to be erroneous. I am not aware of any members of this

family with any asymmetry in the structure of the male postabdomen. If McAlpine

has observed anything different, he should state what species he examined so that

specialists can check the observation.

McAlpine's concept of "Carnoidea" mainly includes families treated in my

1972 book as belonging to the Tephritoinea but not to the Tephritidae family-

group, that is families in which the aedeagus is long and flexible (or assumed to be

derived from such a type) and in which a retractile ovipositor is developed but not

showing the full suite of apomorphies (such as fused cerci) shown by females of the

Tephritidae family-group (Tephritoidea in McAlpine's sense). This concept of

relationship remains valid, but there remain problems regarding which families

belong in the Tephritoinea sensu lato. McAlpine's proposal to recognize a

separate superfamily for the Tephritoidea exclusive of the Tephritidae family-

group seems reasonable pending further investigations. But the superfaimily cannot

be called Carnoidea, because (as noted above) the Carnidae belong to the

Tephritidae family-group (Tephritoidea sensu McAlpine). I think the name

Chloropoidea is the appropriate one.

There are some other misplacements in McAlpine's treatment of the

"Carnoidea". The Cryptochetidae do not belong in the subgroup Milichiidae +

Chloropidae, as discussed above. More probably they are close to the

Lonchaeidae (Tephritoidea). The Risidae belong to the Ephydroidea (see

Chandler 1987), and in my opinion represent an aberrant subgroup of the

Ephydridae not a separate family. On the other hand, two families placed in

Opomyzoidea by McAlpine, Acartophthalmidae and Odiniidae, should be included

here according to the structure of their male and female terminalia.

The relationships of the two families separated by McAlpine at the base of the

"Camoidea", Australimyzidae and Braulidae, are unclarified. The case for including

them here (or in any other superfamily of "Acalyptratae") is quite weak, since based

only on characters subject to homoplasy. The Australimyzidae show some highly

plesiomorphous features, and may represent a group of higher rank (as treated in my

1972 book).

The superfamily Sphaeroceroidea is proposed by McAlpine for part of the

Anthomyzoinea in the sense of my 1972 book. The concept is reasonable pending

further investigations. While most of the Heleomyzidae of the Northern

Hemisphere probably represent a monophyletic group, the same cannot be said for

the Southern Hemisphere forms. D. K. McAlpine (1985), the lone current worker on

these southern groups, was unable to justify his suprageneric concepts in terms of

cladistic analysis, so we may well be dealing with an assemblage of diverse

origins. It is obvious that progress in clarifying the limits of and relationships

within the Sphaeroceroidea will be slow, so long as additional dipterists do not

take up the study of the "Heleomyzidae" of the Southern Hemisphere.

McAlpine's treatment of the Ephydroidea (= Drosophiloidea) generally has my

support, except that he tries to reverse the separation of the Campichoetidae from

Diastata proposed in my 1972 book. This seems to me retrograde. That these

groups are not monophyletic is confirmed in the important paper by Chandler

Quaes t. Ent., 1990, 26(1)

Book Review

129

(1987), which contains a review of the relationships between the families of

Ephydroidea (in which Risidae must also be included, as noted above).

Presumably McAlpine's manuscript was finalized before Chandler's work was

received, since he does not mention it. Future studies should take account of

Chandler's, as well as McAlpine's, treatment.

The treatment of the Calyptratae contains much less that is controversial than

the treatment of the "Acalyptratae". Numerous autapomorphies justify the concept

of Calyptratae, as shown in McAlpine's table of "character states in groundplans of

Acalyptratae and Calyptratae". Three subgroups are recognized (ranked as

superfamilies), the Hippoboscoidea, Muscoidea and Oestroidea. The first and last

are groups recognized (under a diversity of names) in all recent treatments. But

whether the Muscoidea (sensu McAlpine) is a monophyletic group requires further

study. The three characters of this group shown on his phylogeny diagram in my

opinion all belong to the groundplan of more inclusive groups.

CONCLUSION

The publication of this volume represents an important advance in our

understanding of the phylogeny of the Diptera. In writing this critical review I have

tried to distinguish what seems to me well established from what is controversial

or in certain cases demonstrably erroneous. I hope my remarks will assist future

workers in this most interesting field. J. Frank McAlpine is to be congratulated for

his persistence and hard work over many years in bringing the Manual project to

completion. This phylogeny volume will prove seminal, but should not be regarded

as the source of all truth on this subject. We are still at a stage where considerably

divergent opinions can reasonably be held concerning many areas of the system,

and many changes may be expected as a result of future research.

REFERENCES CITED

Chandler, P. J. 1987. The families Diastatidae and Campichoetidae (Diptera,

Drosophiloidea) with a revision of Palaearctic and Nepalese species of

Diastata Meigen. Entomologica Scandinavica 18: 1-50.

Feijen, H. R. 1989. Diopsidae. Flies of the Nearctic Region 9 (12). 122 pp.

Griffiths, G. C. D. 1972. The phylogenetic classification of Diptera Cyclorrhapha,

with special reference to the structure of the male postabdomen. Series

Entomologica no. 8. 340 pp. Dr. W. Junk N. V., The Hague.

Griffiths, G. C. D. 1981. Review of: McAlpine, J. F. et al. (eds.). Manual of

Nearctic Diptera. Volume 1. Entomological Society of Canada Bulletin 13:

49-55.

Griffiths, G. C. D. 1984. Note on characterization of Eremoneura, Orthogenya and

Cyclorrhapha. 1 p. Distributed at XVII International Congress of Entomology,

Hamburg.

Hennig, W. 1937. Milichiidae et Carnidae. Die Fliegen der Palaarktischen Region

5 (52). 39 pp.

Hennig, W. 1958. Die Familien der Diptera Schizophora und ihre phylogenetischen

Verwandtschaftsbeziehungen. Beitrage zur Entomologie 8: 505-688.

Hennig, W. 1972. Beitrage zur Kenntnis der rezenten und fossilen Carnidae, mit

besonderer Beriicksichtigung einer neuen Gattung aus Chile (Diptera:

Cyclorrhapha). Stuttgarter Beitrage zur Naturkunde, no. 240. 20 pp.

Hennig, W. 1973. Ordnung Diptera (Zweifliigler). Handbuch der Zoologie 4(2):

2/31 (Lieferung 20). 337 pp.

Quae st. Ent., 1990, 26(1)

130

Book Review

Kristensen, N. P. 1975. The phylogeny of hexapod "orders". A critical review of

recent accounts. Zeitschrift fur Zoologische Systematik und

Evolutionsforschung 13: 1-44.

Kristensen, N. P. 1981. Phylogeny of insect orders. Annual Review of

Entomology 26: 135-157.

Krivosheina, N. P. 1988. Approaches to solutions of questions of classification

of the Diptera. Entomologicheskoye Obozreniye 67: 378-390 (in Russian).

Also in English translation in Entomological Review 68 (1989): 111-124.

Krivosheina, N. P. 1989. Ontogeny and evolutionary ecology of the Diptera. Itogi

Nauki i Techniki, Seriya Entomologiya, no. 9. 164 pp. (in Russian).

Lameere, A. 1906. Notes pour la classification des Dipteres. Memoires de la

Societe Entomologique de Belgique 12: 105-140.

McAlpine, D. K. 1976. A new genus of flies possibly referable to Cryptochetidae

(Diptera, Schizophora). Australian Entomological Magazine 3: 45-56.

McAlpine, D. K. 1985. The Australian genera of Heleomyzidae (Diptera:

Schizophora) and a reclassification of the family into tribes. Records of the

Australian Museum 36: 203-251.

McAlpine, J. F. 1981. Morgea freidbergi new species, a living sister-species of

the fossil species M. mcalpinei, and a key to world genera of Pallopteridae

(Diptera). Canadian Entomologist 1 13: 81-91.

Nagatomi, A. 1984. Male genitalia of the lower Brachycera (Diptera). Beitrage

zur Entomologie 34: 99-157.

Okada, T. 1956. Systematic study of Drosophilidae and allied families of Japan.

183 pp. Gihodo, Tokyo.

Rohdendorf, B. B. 1964. The historical development of Diptera.

Trudy Paleontologicheskovo Instituta no. 100. 311pp. (in Russian). Also in

English translation (1974), xv + 360 pp. University of Alberta Press,

Edmonton.

Schremmer, F. 1951. Die Mundteile der Brachycerenlarven und der Kopfbau der

Larve von Stratiomys chamaeleon L. Oesterreichische Zoologische

Zeitschrift 3: 326-397.

Tanasiychuk, V. N. 1986. Silver flies (Chamaemyiidae). Fauna U.S.S.R., Diptera

14(7). 335 pp. (in Russian).

Wiegmann, B. M. 1989. A phylogenetic revision of the family Atelestidae

(Diptera: Empidoidea) and its implications for the origin of the

cyclorrhaphous Diptera. viii + 205 pp. University of Maryland, M.Sc.

thesis.

Willman, R. 1989. Evolution und Phylogenetisches System der Mecoptera

(Insecta: Holometabola). Abhandlungen der Senckenbergischen

Naturforschenden Gesellschaft, no. 544. 153 pp.

G. C. D. Griffiths

Department of Entomology

University of Alberta

Quaest. Ent., 1990, 26(1)

Book Review

131

BOOK REVIEW

TRAUTNER, J. and K. GEIGENMULLER. 1987. Carabid beetles, tiger beetles.

Illustrated key to the Cicindelidae and Carabidae of Europe / Sandlaufkafer,

Laufkafer. Illustrierter Schlussel zu den Cicindeliden und Carabiden Europas.

Verlag Joseph Margraf, Aichtel, FRG. 210 x 150 mm. Soft cover. 487 pp.

1200 text figures + 11 colour photographs. Price unknown.

Recently, through the generosity of a colleague in West Germany, I received a

collection of European carabids, as well as the above book to facilitate their

identification. Because of its wide geographical coverage and treatment of a large

and widespread group of beetles, this book would be quite valuable to many

systematists, ecologists and amateur beetle collectors.

The volume has several sections, beginning with an introduction, including a

very brief history of the classification of Carabidae, ecology, and natural history.

To increase its potential usage, the authors have divided the text into German and

English, on alternate pages or columns on single pages. Following this preliminary

material, the classification used in the book is given. The authors adopted the

system of Kryzhanovskiy except that his subfamilies are regarded as families ( i.e .

Omophronidae, Brachinidae), and his supertribes are elevated to subfamily rank.

No reasons are provided for these changes in rank. A key to families of European

terrestrial Adephaga is provided, followed by a key to subfamilies/tribes of

Carabidae ( sensu stricto ). The third key is the generic key, in which a total of 138

genera are actually keyed out, while 30 genera of blind, cave-inhabiting Bembidiini

and Trechinae are keyed into two groups and mentioned by name only. Thankfully

and also commendably, the page number of the treatment of each genus is given, thus

preventing more page flipping. What follows is the bulk of the book: 377 pages of

keys to species of the various genera. For most genera, each species is keyed out,

but for several of the larger genera (including Trechus Clairville, Duvalius

Delarouzee, Bembidion Latreille, Pterostichus Bonelli, Agonum Bonelli, Amara

Bonelli, Harpalus Latreille, and .i.Brachinus Weber;) reduced coverage is given.

For these and other genera (total of 27), completeness of treatment varies from only

references to more detailed taxonomic works, to only a key to subgenus, to keys to

species of a more restricted geographical area.

For each genus, the following data are provided: author and date of

publication of generic name, suprageneric classification, brief description

including mainly colour and size, brief habitat characterization, and pertinent

literature. The keys to species are for the most part clear, concise, and profusely

illustrated with line drawings. The authors use unambiguous characters and most

are effectively illustrated. Characters of the male genitalia are used only when

necessary. A habitus drawing accompanies the treatment of all but very few genera.

Distribution maps are provided for some genera but each species' couplet mentions

the species' geographical distribution.

The book treats species whose distributions are within Europe except:

Turkey, the western part of U.S.S.R., and the eastern parts of Bulgaria and

Romania. As mentioned above, some genera are treated on a smaller scale, and this

is mentioned with or without an accompanying map showing the smaller

geographical area covered in the key.

There are also some noticeable problems, most of which are minor. The

English translation of the original German is adequate, although in certain places it

Quaes t. Ent., 1990, 26(1)

132

Book Review

is difficult to understand. The page numbers for several genera in the index are not

actually the pages on which these taxa are treated in the text. Several of the

references to taxonomic papers cited in the text are not listed in the references. A

number of typographical errors are also evident, perhaps at least partly due to the

translation. Two of the most commonly cited authors in the text are C. Jeanne and

R. Jeannel, and in a few instances, these names are inadvertently switched. One

glaring and inexcusable error is the consistent placement of all authors' names in

parentheses. One would expect that the proper use of parentheses in dealing with

authors and dates of published taxonomic names would be common knowledge to

the authors of such a work. Hopefully this was just an oversight on their part.

This volume is intermediate between a field guide and taxonomic monograph.

Therefore, its readership will have a wide range of interests. The book's main

selling points are its good illustrations and easily followed keys. In fact, almost the

entire text is occupied by keys and figures, and the fact that both have been

prepared with care and accuracy ensures the value of this book. Not being an

expert on Carabidae, I was still able to determine easily the lot of beetles that I

received with the book. Although there are no actual descriptions with which one

could be certain of a determination, the couplets in the keys are such that

descriptions are unwarranted. Literature in which descriptions and/or more

complete taxonomic treatment may be found is always given, in any case.

In conclusion, I have found this book's many merits outweigh considerably the

few flaws. It is certainly not an easy task to prepare both brief and unambiguous

taxonomic keys, especially when dealing with such a large number of genera in

comparatively few pages. The authors have seemingly done a fine job of this, and

the resulting work is highly recommended for those with even a cursory interest in

this large and interesting group of beetles.

Darren A. Pollock

Department of Entomology,

University of Alberta

Quaest. Ent., 1990, 26(1)

BOOK NOTICES

FAUNA OF NEW ZEALAND. C. T. Duval, Series Editor. Science Information

Publishing Centre, DSIR, P.O. Box 9741, Wellington, New Zealand.

The year 1988 was the twenty fifth anniversary of the Systematics Group,

Department of Scientific and Industrial Research, of the Government of New

Zealand. It was marked, among other events, by the appearance of Numbers 13 and

14 of the Fauna of New Zealand. Number 15 was published, as well. In 1989,

Numbers 16-18 were published. These issues are of the same high quality and with

the same desirable features as reported for previously published numbers (see Ball,

1983, Quaestiones Entomologicae, 19 (3-4): 487-488). Below, a citation is given

to each number published in 1988 and 1989, with a few notes that draw attention

to generalizing or other interesting statements in the text. Such statements by the

authors extend coverage of these publications beyond that of identification

manuals.

Each publication contains keys to the taxa treated, and each is illustrated

extensively, principally with line drawings of high quality that are interpreted

easily. In only two publications, however, are scales provided (Numbers 15 and

16), so that one can judge size of the parts illustrated.

It is gratifying to note that five of these publications treat relatively obscure

groups (microlepidoptera and microhymenoptera). It is high time that such taxa

receive the attention that is their due, for each has its own distinctive brand of

biological importance.

These, and previous Numbers, may be ordered through: The Bookshop, DSIR

Publishing, PO Box 9741, Wellington, New Zealand.

NOYES, J. S. 1988. Encyrtidae (Insecta: Hymenoptera). FNZ, Number 13, 187

pp. Price, US $44.95.

Encyrtid adults are small (0.3-3.0 mm. in length) chalcidoid wasps. Most of

the species as larvae are endoparasitoids or hyperparasitoids of coccoid bugs or

arachnids. Most of the New Zealand species for which life history data are

available are parasitoids of coccoids.

Thirty five genera and 70 species of Encyrtidae are recorded from New

Zealand and the adjacent subantarctic islands. In this publication, four genera and 32

species are described as new.

Although a classification to subtribe is provided, the taxonomic treatments

are alphabetical in sequence, based on generic names, and within each genus, on

specific epithets. That is, the classification is not used in a meaningful way.

The author accounts for the marked intraspecific variation exhibited by the

encyrtids of New Zealand by suggesting recent occupation of many ecological

niches that have become available as a result of post-glacial speciation of the

coccoid hosts.

Nearly half of the known encyrtid species of New Zealand probably are man-

introduced, mostly from Europe and Australia. Of the 39 native taxa, eight occur in

Australia also, and are postulated to be recent overseas arrivals in New Zealand.

The detailed treatment of taxa is useful not only for identification of New

Zealand encyrtids, but also as a basis for comparison with the encyrtid fauna of

other parts of the world. Another generally useful feature is the description of

collecting methods, with its emphasis on yellow pan traps and Malaise traps.

Book Review

DUGDALE, J. S. 1988. Lepidoptera— annotated catalogue, and keys to

family-group taxa. FNZ, Number 14, 262 pp. Price, US $49.95.

Appropriately, this publication begins with a dedication to "the memory of

three amateur lepidopterists" whose combined efforts contributed so much to

knowledge of the lepidopteran fauna of New Zealand: "George Vernon Hudson,

1867-1946, whose life work this publication attempts to keep evergreen"; "Alfred

Philpott, 1871-1933, whose pioneering studies in Lepidoptera morphology are

now assuming their true significance"; and "Kenneth John Fox, FRCOG, 1936-

1986, who died before this catalogue reached full term but whose enthusiasm and

insistence ensured its completion". A quotation from John Tyndall's 'Fragments of

Science', published in 1876, reminds readers of the empirical basis of science: "I

would advise you to get a knowledge of the facts from actual observation. Facts

looked at directly are vital; when they pass into words half the sap is taken out of

them". To that comment, I note that one living in the present age of published pap

and propaganda must be concerned not only about the sap that is taken out when

"facts" are published, but also about the polluted sap that may be interjected as

facts by authors who are more concerned with self-aggrandizement than with

honest exposition and appraisal.

This volume is a valuable historical account of knowledge of basic taxonomic

aspects of the Lepidoptera of New Zealand. The 1761 species (1582 endemic)

from New Zealand are assigned to family and the families are grouped into

superfamilies. For genera, type species are indicated. For species, type specimens

are indicated by label data, sex, and institution where housed. An appendix

provides a list of the taxa of Lepidoptera recorded from the Kermadec Islands.

The excellent illustrations include habitus figures of representative specimens

placed at the beginning of the text for each family.

An incisive discussion of the taxa of Lepidoptera above the level of

superfamily outlines major classificatory problems, and provides an entrance to the

literature about this subject. The author points out that "most species belong to

"one division (Ditrysia) of relatively uniform structural organization", whereas

"the remaining 1 -2% show a great variety of structural and genital

organization, and often profound differences between groups". The evolution of

Lepidoptera is characterized by the Hennigian term "additive typogenesis",

implying a gradual acquisition of group characters, from the primitive

micropterygoids onward, culminating in the endoporian ditrysian suite of

characters.

A discussion of the composition of the lepidopteran fauna of New Zealand

emphasizes the marked endemicity of the fauna and the good representation of the

non-ditrysian groups which comprise at the species level 5 percent of the total

fauna. However, a number of non-ditrysian families are absent, even though their

food plants are present. More generally, the faunal relationships of the New Zealand

taxa are varied, but such relationships have been determined for few groups. Much

zoogeographical work remains to be done.

This scholarly work will serve well the development of study of the

lepidopteran fauna of New Zealand. The discussion of relationships of the higher

taxa will be of interest to systematic entomologists, generally. Those to whose

memory this volume is dedicated are indeed honored.

NAUMANN, I. D. 1989. Ambositrinae (Insecta: Hymenoptera: Diapriidae).

FNZ, Number 15, 165 pp. Price, US $39.95.

The Ambositrinae are one of the major hymenopterous components of the

forest fauna of New Zealand. Probably, as larvae, all are parasitoids of the

Quaest. Ent., 1990, 26(1)

■

Book Review

immature stages of nematocerous Diptera (Mycetophilidae and Keroplatidae, for

example), the members of which are also abundant in the litter and low vegetation of

New Zealand forests. Wing reduction, a phenomenon that is common among

islandic taxa, is marked in the Ambositrinae, with 89 percent of the New Zealand

species being brachypterous or apterous.

The New Zealand component of this proctotrupoid subfamily comprises

seven genera (three described as new) and 46 species. Thirty four species are

described as new, and 11, known only from inadequate material, are not named

formally.

Although keys are provided to the subfamilies of Diapriidae in New Zealand,

to the genera of Ambositrinae and to the species of each genus, no information is

offered about how one distinguishes the Diapriidae from other proctotrupoid

families.

A reconstructed phylogeny, based on analysis of 45 characters, is provided

for 13 groups of Ambositrinae (12 genera and the Dissoxylabis genus-group of

seven genera). Within this assemblage, the genera of the New Zealand fauna range

from the most archaic to the most highly derived. As the author indicates, this

reconstruction must be viewed with considerable caution, because many of the

branches are based on symplesiotypy, only.

The present-day austral disjunct distribution of Ambositrinae reflects a

Gondwana radiation no later than the Cretaceous, and the marked structural

divergence in the subfamily is evidence of a long evolutionary history.

The most appealing feature of this admirable study is the author's attempt to

go beyond provision of a clear report about a group of very small and

taxonomically difficult creatures. The reconstructed phylogeny and general

observations about the geographical history of the Ambositrinae will serve well

future workers in their endeavors to understand relationships and classification of

this taxon.

DONNER, H. and C. WILKINSON. 1989. Nepticulidae (Insecta:

Lepidoptera). FNZ, Number 16, 88 pp. Price, US $22.95.

Nepticulid larvae are leaf miners. The species in New Zealand (28 in total, 14

described as new in this publication) are variously recorded from plants of eight

families, with the majority of species mining leaves of the family Asteraceae.

All of the New Zealand species are assigned to the genus Stigmella Schrank.

One species, S. microtheriella (Stainton) was introduced from western Europe. The

j remaining 27 species are endemic.

The illustrations of genitalia are especially well done, and are laid out in such a

way that comparisons are made easily.

Unfortunately, no attempt is made to treat the New Zealand fauna by means of

an evolutionary analysis. Comparisons, made in the text to ease the task of

identifying these small moths (wing span 2-9 mm.), could have served as well as the

basis for a reconstructed phylogeny.

NOYES, J. S. and E. W. VALENTINE. 1989. Mymaridae (Insecta:

Hymenoptera)— introduction and review of genera. FNZ, Number 17, 95 pp. Price,

US $24.95.

The family Mymaridae is represented in New Zealand by 160 species

distributed among 42 genera. Known as "fairy flies", adult mymarids range in body

size from less than 0.4 to about 4.0 mm. They are so small that effectively they

swim through the air rather than fly. They can be carried for great distances as part

of the aerial plankton.

Quae st. Ent., 1990, 26(1)

Book Review

Most mymarids as larvae are egg parasitoids. Most mymarid species develop

from eggs of sternorrhynchous Homoptera, but various species have been reared

from the eggs of other Hemiptera, Coleoptera and Psocoptera.

The proportion of flightless species is high: at least 17 genera have species

with flightless adults. The largest number of flightless species lives in leaf litter.

Of the 42 genera treated, 20 are known from New Zealand, only; four are

shared with Australia, only; one is shared with South America, only; one is

distributed through Australasia and ranges to South America; three reach the Indian

sub-continent; and 13 are cosmopolitan.

NOYES, J. S. and E. W. VALENTINE. 1989. Chalcidoidea (Insecta:

Hymenoptera)— introduction and review of genera in smaller families. FNZ,

Number 18, 91 pp. Price, $22.95.

Treated taxonomically are the New Zealand genera of 15 families of

Chalcidoidea. Also included is the Mymarommatidae (two genera, one in New

Zealand), which is the putative sister group of the Chalcidoidea. A table provides

data about hosts for this important group of wasps. Most chalcidoid parasitoids

seek hosts among members of the suborder Homoptera. However, there are

phytophagous taxa, such as members of the family Agaonidae. Also, phytophagous

species appear in genera in five additional families.

The number of taxa is instructive. Described are 93 species in 75 genera.

However, represented in collections in New Zealand are a total of 636 species in

202 genera. These latter figures represent a 2.7 fold increase for genera, and a 6-

fold increase for species. One can conclude that much remains to be done, before the

chalcidoid fauna of New Zealand is even tolerably well known. This publication is

an important step toward attaining a more complete knowledge of New Zealand

chalcidoids.

In conclusion, I am pleased to report that the Editors and authors continue to

serve admirably, with their contributions to this outstanding series of

publications, the entomological community not only of New Zealand but of the

world. The individual volumes provide full value at their asking prices.

George E. Ball

Department of Entomology

University of Alberta

Quaest. Ent., 1990, 26(1)

Quaest

lones

Entomolog

icae

t/tCZ

library

|\UG 0 6 1990

HARVARD

UNlVBR=>iTY

A periodical record of entomological investigations

published at the Department of Entomology,

University of Alberta, Edmonton, Canada.

VOLUME 26

NUMBER 2

SPRING 1990

Publication of Quaestiones Entomologicae was started in 1965 as part of a

memorial project for Professor E.H. Strickland, the founder of the Department

of Entomology at the University of Alberta in 1922.

It is intended to provide prompt relatively low cost publication for compre¬

hensive accounts of entomological research of greater than average length.

However, shorter papers about insects in the Prairie Provinces of Canada are

acceptable. Page charges are normally levied, the rate determined by printer’s

charges. For information about current page charges, consult the Editor.

Copy for all types of papers should conform to the Style Manual for

Biological Journals, published by the American Institute of Biological Sci¬

ences, Second Edition, 1964, except that title of periodicals should be given in

full. For style of taxonomic papers, the Editor should be consulted. Two copies

of a manuscript are requested. All manuscripts will be reviewed by referees.

Abstracts are required, one in English and one in another language, prefer¬

ably French.

Copy for illustrations must accompany the manuscript, and be of such char¬

acter as to give satisfactory reproduction at page size (less 1/2 inch, or 1.2 cm

on plates of full page size [7X4 1/4 inches or 17.8 X 10.7 cm]). Reprints must

be ordered when proofs are returned, and will be supplied at cost.

Subscription rates are the same for institutions, libraries and individuals,

$20.00* per volume of four issues, normally appearing at quarterly intervals;

single issues $5.00**. Back volumes and issues are available at the same cost.

These prices supersede those previously indicated, and are subject to change as

required by inflationary pressure on the value of money.

Communications regarding subscriptions should be addressed to the

Subscription Manager, and regarding manuscripts to the Editor.

* Same rate in US$ for non-Canadian subscriptions.

** Single issues of more than 100 pages: $10.00; exception: volume 21(4) - $20.00.

Published quarterly by:

Department of Entomology

University of Alberta

Edmonton, Alberta, CANADA

T6G 2E3

Second Class Mail Registration Number 5222

Return Undeliverable mail to the address above. Return Postage Guaranteed

Issued July 1990

THIRD INTERNATIONAL CONFERENCE ON CLASSIFICATION,

PHYLOGENY, AND NATURAL HISTORY OF HYDRADEPHAGA

(COLEOPTERA)

Proceedings

Organized and Edited by

R. E. Roughley and R.B. Aiken

°°,'Elt-JVVi V<i'

Convened at the XVIIIth International Congress of Entomology

Vancouver, British Columbia, Canada

3 - 9 July 1988

iOLOGV

QUAESTIONES ENTOM OLOGICAE

ISSN 0033-5037

A periodical record of entomological investigation published at the

Department of Entomology, University of Alberta, Edmonton, Alberta.

Volume 26 Number 2 1990

CONTENTS

Roughley and Aiken — Introduction . 137

NATURAL HISTORY OF HYDRADEPHAGA

Garcia, Hagen and Voigt — Life History, Termination of Summer Diapause,

and other Seasonal Adaptations of Agabus disintegratus (Crotch)

(Coleoptera: Dytiscidae) in the Central Valley of California . 139

Larson — Odonate predation as a factor influencing dytiscid beetle

distribution and community structure . 151

SYSTEMATICS OF HYDRADEPHAGA

Beutel — Phylogenetic Analysis of the Family Gyrinidae (Coleoptera) Based

on Meso- and Metathoracic Characters . 163

Hilsenhoff — Use of Gonocoxae and the Sternal Apex to Identify Adult

Females of North American Gyrinus Geoffroy (Coleoptera: Gyrinidae). .1 93

Alarie, Harper and Maire — Primary Setae and Pores on Legs of Larvae of

Nearctic Hydroporinae (Coleoptera: Dytiscidae) . 199

Bistrom — Revision of the Genus Queda Sharp (Coleoptera: Dytiscidae) . 2 1 1

Burmeister — The Systematic Position of the Genus Agabetes Crotch within

Dytiscidae (Coleoptera: Adephaga) . 221

Brancucci — A New Species of Platambus, Subgenus Agraphis, from Nepal

and notes on P. guttulus (Regimbart) (Coleoptera: Dytiscidae) . 239

INTRODUCTION

In 1982, Rob Roughley and Bill Wolfe organized the First International

Conference on the Classification, Phylogeny and Natural History of the

Hydradephaga in Toronto. The proceedings of the meetings were published in the

Proceedings of the Academy of Natural Sciences of Philadelphia, 137(1) in 1985.

The second conference was organized by Michel Brancucci and Konrad Dettner

and was convened at the XVII International Congress of Entomology in Hamburg in

1984. The results of this symposium were published in Entomologica Basiliensia,

11 in 1986. The next two issues of Quaestiones Entomologicae (Volume 26,

Numbers 2 and 3) mark the culmination of the third such conference held in

conjunction with the XVIII International Congress of Entomology in Vancouver in

1988.

The rationale for these International Conferences has been and continues to be

to stimulate international collaboration. This is the preferred method for advancing

science in general and the study of Hydradephaga in particular. Nevertheless,

science is a human endeavour: its advancement depends on the enthusiasm and ideas

of the participants. We hope that the latest proceedings will lead to even greater

collaboration and further advance the study of Hydradephaga. From this

viewpoint, it is encouraging to note that, in the present proceedings, European

authors discuss Nearctic taxa and North American authors discuss Palearctic taxa.

As with the other conferences, the majority of the papers in this volume deal

with hydradephagan systematics. As such, they represent significant contributions

rarely found in one proceeding. Several of these papers treat the systematics of

families (Beutel - Gyrinidae; Burmeister - Amphizoidae) and genera (Wolfe and

Roughley - Laccornis; Roughley -Dytiscus). These studies lay a solid foundation

for phylogenetic and evolutionary analyses of this important group of aquatic

insects.

The five families which comprise the Hydradephaga include 5,000 - 6,000

species in the world fauna and encompass a wide range of structural diversity. An

analysis of the literature on Hydradephaga over the last five years suggests that

about 200 papers per year are being published by about 200 authors. This

literature is primarily systematic-taxonomic-faunistic, suggesting that much more

work is needed in these areas. Nevertheless, there are encouraging signs that other

studies about natural history, behaviour and ecology of Hydradephaga are

becoming more common. This melding of a variety of research efforts holds great

promise for both the interpretation of relationships within the Hydradephaga and

an understanding of evolutionary changes occasioned by invasions of fresh water.

Several colleagues helped in the preparation of this volume by reviewing

mansucripts for us. We offer our sincere thanks to Anders Nilsson, Yves Bousquet,

Gary Gibson, Rolf Beutel, Phil Perkins, Dave Larson, Ales Smetana, Valerie Behan-

Pelletier, Olof Bistrom, Sule Oygur, F. Merv Atton, Bill Wolfe, Richard Garcia,

Ingolf Askevold and Terry Galloway. We also extend a vote of thanks to George

Ball, Editor, Quaestiones Entomologicae and Mrs. Suseela Subbarao for their

efforts in getting these proceedings into print.

R. E. Roughley, Winnipeg, Manitoba

R. B. Aiken, Sackville, New Brunswick

Publication of Quaestiones Entomologicae was started in 1965 as part of a

memorial project for Professor E.H. Strickland, the founder of the Department

of Entomology at the University of Alberta in 1922.

It is intended to provide prompt relatively low cost publication for compre¬

hensive accounts of entomological research of greater than average length.

However, shorter papers about insects in the Prairie Provinces of Canada are

acceptable. Page charges are normally levied, the rate determined by printer’s

charges. For information about current page charges, consult the Editor.

Copy for all types of papers should conform to the Style Manual for

Biological Journals, published by the American Institute of Biological Sci¬

ences, Second Edition, 1964, except that title of periodicals should be given in

full. For style of taxonomic papers, the Editor should be consulted. Two copies

of a manuscript are requested. All manuscripts will be reviewed by referees.

Abstracts are required, one in English and one in another language, prefer¬

ably French.

Copy for illustrations must accompany the manuscript, and be of such char¬

acter as to give satisfactory reproduction at page size (less 1/2 inch, or 1.2 cm

on plates of full page size [7X4 1/4 inches or 17.8 X 10.7 cm]). Reprints must

be ordered when proofs are returned, and will be supplied at cost.

Subscription rates are the same for institutions, libraries and individuals,

$20.00* per volume of four issues, normally appearing at quarterly intervals;

single issues $5.00**. Back volumes and issues are available at the same cost.

These prices supersede those previously indicated, and are subject to change as

required by inflationary pressure on the value of money.

Communications regarding subscriptions should be addressed to the

Subscription Manager, and regarding manuscripts to the Editor.

* Same rate in US$ for non-Canadian subscriptions.

** Single issues of more than 100 pages: $10.00; exception: volume 21(4) - $20.00.

Published quarterly by:

Department of Entomology

University of Alberta

Edmonton, Alberta, CANADA

T6G 2E3

Second Class Mail Registration Number 5222

Return Undeliverable mail to the address above. Return Postage Guaranteed

Issued July 1990

THIRD INTERNATIONAL CONFERENCE ON CLASSIFICATION,

PHYLOGENY, AND NATURAL HISTORY OF HYDRADEPHAGA

(COLEOPTERA)

Proceedings

Organized and Edited by

R. E. Roughley and R.B. Aiken

Convened at the XVIIIth International Congress of Entomology

Vancouver, British Columbia, Canada

3 - 9 July 1988

iOLOGV

QUAESTIONES ENTOMOLOGICAE

ISSN 0033-5037

A periodical record of entomological investigation published at the

Department of Entomology, University of Alberta, Edmonton, Alberta.

Volume 26 Number 2 1990

CONTENTS

Roughley and Aiken — Introduction . 137

NATURAL HISTORY OF HYDRADEPHAGA

Garcia, Hagen and Voigt — Life History, Termination of Summer Diapause,

and other Seasonal Adaptations of Agabus disintegratus (Crotch)

(Coleoptera: Dytiscidae) in the Central Valley of California . 139

Larson — Odonate predation as a factor influencing dytiscid beetle

distribution and community structure . 151

SYSTEMATICS OF HYDRADEPHAGA

Beutel — Phylogenetic Analysis of the Family Gyrinidae (Coleoptera) Based

on Meso- and Metathoracic Characters . 163

Hilsenhoff — Use of Gonocoxae and the Sternal Apex to Identify Adult

Females of North American Gyrinus Geoffroy (Coleoptera: Gyrinidae). .1 93

Alarie, Harper and Maire — Primary Setae and Pores on Legs of Larvae of

Nearctic Hydroporinae (Coleoptera: Dytiscidae) . 199

Bistrom — Revision of the Genus Queda Sharp (Coleoptera: Dytiscidae) . 2 1 1

Burmeister — The Systematic Position of the Genus Agabetes Crotch within

Dytiscidae (Coleoptera: Adephaga) . 221

Brancucci — A New Species of Platambus, Subgenus Agraphis, from Nepal

and notes on P. guttulus (Regimbart) (Coleoptera: Dytiscidae) . 239

INTRODUCTION

In 1982, Rob Roughley and Bill Wolfe organized the First International

Conference on the Classification, Phytogeny and Natural History of the

Hydradephaga in Toronto. The proceedings of the meetings were published in the

Proceedings of the Academy of Natural Sciences of Philadelphia, 137(1) in 1985.

The second conference was organized by Michel Brancucci and Konrad Dettner

and was convened at the XVII International Congress of Entomology in Hamburg in

1984. The results of this symposium were published in Entomologica Basiliensia,

11 in 1986. The next two issues of Quaestiones Entomologicae (Volume 26,

Numbers 2 and 3) mark the culmination of the third such conference held in

conjunction with the XVIII International Congress of Entomology in Vancouver in

1988.

The rationale for these International Conferences has been and continues to be

to stimulate international collaboration. This is the preferred method for advancing

science in general and the study of Hydradephaga in particular. Nevertheless,

science is a human endeavour: its advancement depends on the enthusiasm and ideas

of the participants. We hope that the latest proceedings will lead to even greater

collaboration and further advance the study of Hydradephaga. From this

viewpoint, it is encouraging to note that, in the present proceedings, European

authors discuss Nearctic taxa and North American authors discuss Palearctic taxa.

As with the other conferences, the majority of the papers in this volume deal

with hydradephagan systematics. As such, they represent significant contributions

rarely found in one proceeding. Several of these papers treat the systematics of

families (Beutel - Gyrinidae; Burmeister - Amphizoidae) and genera (Wolfe and

Roughley - Laccornis; Roughley -Dytiscus). These studies lay a solid foundation

for phylogenetic and evolutionary analyses of this important group of aquatic

insects.

The five families which comprise the Hydradephaga include 5,000 - 6,000

species in the world fauna and encompass a wide range of structural diversity. An

analysis of the literature on Hydradephaga over the last five years suggests that

about 200 papers per year are being published by about 200 authors. This

literature is primarily systematic-taxonomic-faunistic, suggesting that much more

work is needed in these areas. Nevertheless, there are encouraging signs that other

studies about natural history, behaviour and ecology of Hydradephaga are

becoming more common. This melding of a variety of research efforts holds great

promise for both the interpretation of relationships within the Hydradephaga and

an understanding of evolutionary changes occasioned by invasions of fresh water.

Several colleagues helped in the preparation of this volume by reviewing

mansucripts for us. We offer our sincere thanks to Anders Nilsson, Yves Bousquet,

Gary Gibson, Rolf Beutel, Phil Perkins, Dave Larson, Ales Smetana, Valerie Behan-

Pelletier, Olof Bistrom, Sule Oygur, F. Merv Atton, Bill Wolfe, Richard Garcia,

Ingolf Askevold and Terry Galloway. We also extend a vote of thanks to George

Ball, Editor, Quaestiones Entomologicae and Mrs. Suseela Subbarao for their

efforts in getting these proceedings into print.

R. E. Roughley, Winnipeg, Manitoba

R. B. Aiken, Sackville, New Brunswick

■

. , .. ■

• '

LIFE HISTORY, TERMINATION OF SUMMER DIAPAUSE, AND

OTHER SEASONAL ADAPTATIONS OF AGABUS

DISINTEGRATUS (CROTCH) (COLEOPTERA: DYTISCIDAE) IN

THE CENTRAL VALLEY OF CALIFORNIA

R. Garcia

K.S. Hagen

W.G. Voigt

Quaestiones Entomologicae

26: 139-149 1990

Division of Biological Control

University of California

Berkeley, CA 94720

U. S.A.

ABSTRACT

The predaceous diving beetle, Agabus disintegratus, breeds in temporary

water sources, and adults pass the summer dry period in a state of diapause.

Diapausing beetles bury themselves at the base of wetland plants until the pond is

reflooded. Diapause is terminated in the laboratory by a short photoperiod, and

weight changes and reproductive activity of field collected adults indicate that

termination of diapause begins in the early fall before the ponds refill with water.

Laboratory experiments indicate that microhabitats must maintain 100% relative

humidity for long term survial of aestivating adults. Beetles could not be induced

to fly during diapause. Flight activity corresponds to the most optimal period for

finding natural water sources in California's Mediterranean climate.

INTRODUCTION

Adults of the predaceous diving beetle, Agabus disintegratus (Crotch), pass

the dry summer months in diapause (Garcia and Hagen 1985, Garcia and Hagen

1987). A. disintegratus is a univoltine species. Larval development takes place

through the winter, and in early March teneral adults emerge, feed, and synthesize

fat reserves through the spring prior to summer aestivation. As the water source

dries, adults bury into the root layer and debris of plants in the basin of the pond,

where they pass the summer. Although Young (1960) reported that A. disintegratus

flew from drying ponds in Indiana, Garcia and Hagen (1987) could find no evidence

for flight in adult A. disintegratus collected in the summer, even though the wings

appeared to be normal.

The conditions and habitats in which A. disintegratus aestivates were

described in Garcia and Hagen (1987). However, little is known about the

adaptations of this beetle to the absence of water, its dispersal ability, or about

conditions which regulate the termination of diapause. This report investigates

these aspects of the biology of this aquatic beetle.

MATERIALS AND METHODS

Study Site. — Field observations and collections of A. disintegratus were

made at approximately one month intervals from June 1986 through August 1988.

All experimental materials were collected from pond #78 at the Gray Lodge

140

Garcia, Hagen and Voigt

Wildlife Area, Butte County, California, which has been described in detail by

Garcia and Hagen (1987). The basin of this pond is now almost completely covered

with bermudagrass ( Cynodon dactylon L.), in contrast to only partial coverage in

1986. Pond #78 was filled with water during the first week of October 1986, and

remained full until the pond was drained in mid-May 1987. The pond was flood-

irrigated for 3-6 days in June and again in July, 1987. In 1987, pond #78 was

filled in early October and remained so until it dried in mid- April 1988. There were

no summer irrigations in 1988.

Immatures. — Immature beetles were collected with a fine-mesh aquatic sweep

net. Mature larvae, pupae and post-eclosion adults of A. disintegratus were

collected from the soil on the north-western embankment above the water line of

pond #78. A 5 cm diameter soil corer was used to sample the first 6 cm of soil,

which was later sorted in the laboratory. These samples were taken at

approximately 2 week intervals from March 2, 1988 through May 5, 1988, at

which time no pupae or adults remained. Fifteen mature larvae were removed from

their pupal cells, weighed and transferred to 2 cc vials with moist cotton on the

bottom. The resulting pupae were reweighed within 1 day of pupation. Larval

instar determinations were based on body length and head capsule width.

Adults. — Adult beetles were collected either from aquatic sweep net samples

when standing water was present (October through April), or from the sod beneath

the bermudagrass (late April through late September) (see Garcia and Hagen 1987).

Beetles collected in the summer were sieved from bermudagrass sod collected with

a shovel as described in Garcia and Hagen (1987). Other adults were reared in the

laboratory from larvae and pupae collected from soil samples taken along the

exposed banks of the pond.

Adult A. disintegratus collected from each field visit were weighed to ± 0.1

mg on an analytical balance and 5-10 specimens were dissected to determine

reproductive activity and fat development as described by Garcia and Hagen

(1987). Adults collected from water were blotted lightly with dry tissue paper

before weighing. Beetles were weighed within 24 hrs after field collection or

laboratory emergence. Teneral adult weights were determined by weighing 33

individuals within 1 day of eclosion. Beetles were held alive for up to one week

until they could be dissected.

Weight Gain. — The rate of weight increase of adults was determined in the

laboratory using newly emerged A. disintegratus adults. One to three-day old

adults were placed individually in a 250 cc jars of water with a small piece of

Elodea sp. for a resting substrate. The beetles were fed 20 late instar Culex

pipiens L. larvae every other day and weighed at 7-10 day intervals until their

weights stabilized.

Weight Loss. — Adult A. disintegratus in diapause were subjected to various

levels of water moisture to estimate their ability to withstand desiccation.

Eighteen adults were collected from sod samples on July 29, 1988, and weighed the

following day. All beetles were held overnight (18 hrs) in 200 cc of water to

equalize hydration levels and then reweighed . Half of the beetles (5 females, 4

males) were placed in 250 cc jars with moistened sand (100 g sand with 20 cc

water), while the other group (5 females, 4 males) was placed in sand which had

previously been dried to constant weight. Both groups were held at room

temperature (20 ± 5°C). Beetles were weighed at 1-2 day intervals until no

survivors remained in the dry sand group.

Weight loss of adult beetles at various relative humidities was measured as

follows: relative humidities estimated at 100%, 90%, 69%, and 50% were attained

by mixing 0, 10, 37.5, and 62.5 g of potassium hydroxide, respectively with 100

Agabus disintegratus (Crotch)

141

cc water in 0.94 1 jars (Peterson 1959). Beetles were placed in refugia made from

tubes of black cotton fabric (15x6 mm diameter) and closed at one end. The

refugia provided a dark, constricted resting place which most beetles voluntarily

entered. Two beetles (1 male & 1 female) in separate refugia were placed in a

plastic cup (3 cm diameter), which was supported above the solution by a 2.8 cm

diameter PVC pipe which had several holes drilled in it to facilitate equilibration

of water vapor. A lid with 1.2 mm mesh screen was placed on the cups to prevent

beetles from escaping. All jars were covered with a lid and placed in a sealed box

to exclude light, and the box was then placed in a constant temperature cabinet at

21.5 ± 1°C. The beetles were weighed at 4-6 day intervals for 25 days. Body

weights were transformed to proportion of original body weight at day 0.

Regression lines based on these values over time were tested for parallelism, and

post hoc comparisons of the slopes of the regression lines were made using

Scheffe's multiple comparison (Marascuilo & Levin 1983).

Flight activity. — To determine time of year when A. disintegratus individuals

could be induced to fly, field collected adult beetles were tested for flight within 2-

3 days after collection and laboratory reared animals within 3 days of emergence.

A 250 cc jar with a mouth opening of 6 cm in diameter was filled with 5 cm of sand

and 150 cc of water. A round wooden stick about 10 cm in length and 0.3 cm in

diameter was inserted vertically in the sand so that its apex extended about 1 cm

above and in the center of the mouth of the jar. Ten jars were then placed in a water

bath.

An individual beetle was placed into each jar and allowed to acclimate for at

least 15 minutes before observations were started. Flight tests were usually

conducted in full sun at mid-day, and run until water temperatures exceeded 30°C

unless otherwise noted. Flight was determined from direct observations or from

the presence of beetles in the surrounding water bath. The beetles were unable to

escape the jar without flight; therefore any beetles which were missing or found in

the water bath were assumed to have flown.

A single flight test was conducted on an overcast day (March 31, 1987) in

which two randomly selected groups of 12 adults each (6 males, 6 females) were

placed in water artificially heated from 23 to 36°C or in water kept at ambient

temperatures (23 to 25°C).

Occasionally, adults were placed directly on loose dry soil exposed to full sun,

and gross behavioral reactions recorded. Beetles observed to fly from any of the

above conditions were recaptured when possible and dissected to determine

reproductive condition and fat reserve.

Termination of diapause. — Adults used in this experiment (three groups of 5

pairs each) were collected from sod samples in May and June 1987. On June 29,

each male/female pair was held in moistened sand (100 g sand and 15 cc water) in a

250 cc jar. A lid was placed on each jar to reduce evaporation. A 2 mm hole in the

lid allowed for gas exchange. Each group of 5 jars was placed in one of three

environmental chambers under the following conditions: "short light" ( 1 2L: 1 2D hr)

at 13°C and 18°C and "long light" (16L:8D ) at 15°C. All temperatures were

controlled to within ± 1°C. On September 9, 1987, the 14 surviving pairs from the

three groups were placed in jars with 200 cc of tap water and one or two pieces of

Elodea as an oviposition substrate under 16L:8D, 15°C conditions. Ten to 20

Culex pipiens L. larvae (3rd-4th instar) were given to the beetles every one or two

days, and the plants were inspected daily for eggs from September 10 to

September 18, 1987, at which time the beetles were dissected and examined for

reproductive condition.

Quaest. Ent. . 1990, 26(2)

142

Garcia, Hagen and Voigt

Another group of adults collected from sod on June 22, 1987 was paired by

sex and placed directly in jars with 250 cc of water and one or two pieces of

Elodea on June 25, 1987, and given 10 late instar Cx. pipiens larvae every three to

five days. Three groups of five male/female pairs were placed under the same light

regimes as the groups placed in moist sand. The Elodea was examined each week

for eggs until the beetles were dissected September 2-4, 1987.

RESULTS

Immatures. — First instar A. disintegratus larvae were observed from October

26 through April 16, indicating continuous egg laying through the winter and early

spring. Second instars were observed from November 23 through April 16, and

third instars were observed from January 14 through April 16.

Mature larvae were observed leaving the pond to pupate from January 23

through April 12, 1988. Larvae crawled up the slope (a 1.5 m levee) to a vertical

height of about 1 m above the water line and buried about 1-2 cm deep into the

moist soil of the slope. Once under the soil, the larva shaped a spherical mud

pupation cell about 6 mm in diameter. Several mature larvae but no pupae were

observed inside their cells on February 17, 1988. Pupae were first observed on

the following visit (March 2), and the first adults were recovered from pupal cells

on March 15. In 1986 and 1987, teneral adults were first observed in the water on

March 19, and March 26, respectively, but in 1988, they were not seen until April

12.

Mature larvae collected from the field inside their pupal cells weighed 23.1 ±

2.0 mg S.D. and the resulting pupae weighed 28.8 ± 1.8 mg (N = 15). The apparent

increase in weight must have resulted from absorption of water, as they were held

in a 100% humidity chamber on moist cotton.

Adults. — Figure 1 shows seasonal changes in adult body weight and fat

reserves from eclosion to termination of diapause in the fall. A sharp increase in

body weight and fat reserves was noted in the spring. Body weight and fat

reserves remained relatively constant until mid-August when a decline was noted in

both categories. After the pond was filled in the fall, body weights increased, but

fat reserves declined.

No evidence of reproductive activity was noted in dissected adults collected

from sod samples in the summer. Reproductive activity was observed in beetles

collected from water in early October 1987, only five days after the field had been

filled with water. Of five females dissected, all contained sperm in their

spermathecae and 10-19 large developing eggs. Evidence of reproductive activity

( i.e ., mature eggs in the ovarioles, sperm in the spermathecae, etc.) was observed

from October through April.

In March and April of 1986 and 1987, both teneral and older generation adults

were found together in the pond. In 1988, however, no older generation adults were

seen in the pond after mid February. On February 15, 1987 more than a hundred

adults (approx. 30 were collected within 30 m^) were observed under plant debris

along the levee of the pond. These were all reproductively active, older generation

adults, as determined by dissection. On March 2, adults were not found on the

bank, and could not be found in the pond by sweeping with an aquatic net. Only a

small percent of the mature larvae in the soil samples had pupated by March 2, and

teneral adults were not observed in pupal chambers until March 15. All adults had

emerged from pupal cells and returned to the water by the end of April.

Agabus disintegratus (Crotch)

143

X

<5

□

5

45 mg

40 mg

35 mg

30 mg

25 mg

20 mg

15 mg

Mar Apr May Jun Jul Aug Sap Oct Nov Dac

9

J_ 1_ I_ I_ I_ I_ I_ I_ L

Fig. 1. Seasonal changes in average body weights and fat body reserves of field collected Agabus

disintegratus Crotch from 1 986- 1 988 1 . Midline horizontal line indicates mean. Vertical bars of

weights indicate ± 2 standard error of the mean. The numbers indicate the number of beetles

measured.

1 Fatbody stages: I, layer of fat bound to integument; II, free fat bodies; III, "filled" with free fat

bodies.

2 Values for March represent weights from teneral adults reared from field collected pupae.

Adults were weighed within 2 days of eclosion.

3 Only non-reproductive adults were included in the April and May averages, since they

represented the current generation.

Quaest. Ent. . 1990, 26(2)

144

Garcia, Hagen and Voigt

Number of Days

Fig. 2. Body weights of aestivating Agabus disintegratus Crotch under various conditions and

relative humidities. The groups at various relative humidities were held in the dark in a 21 ± 1°C

chamber.

* Held under room conditions (15-25°C).

2 Held in dark at 21±1°C.

Agabus disintegratus (Crotch)

145

Table 1. Post hoc comparisons of slopes of weight loss vs. time for adult A.

disintegratus under various conditions. Body weights were transformed to

proportions of original body weights. Comparisons were of the slopes of

1 I = Reproductively inactive, A = Reproductively active. From beetles which

were recovered after flight.

2 In artificially heated water bath under overcast skies.

3 Unheated water bath under overcast skies. Conducted concurrently with heated

group.

4 Reared from mature larvae collected February 2, 1988.

5 Tested in shallow soil exposed to bright sun.

Quaest. Ent. , 1990, 26(2)

146

Garcia, Hagen and Voigt

Table 3. Effects of photoperiod and substrate on reproductive conditions of field

collected female Agahus disintegrate.

1 Oocytes conspicuous in ovarioles at 10X

2 Sperm present in spermathecae

3 Held in soil from 29 Jun through 4 Sept, 1987, then placed in water and fed

on mosquito larvae under 16L:8D photoperiod until dissected on 18 Sept.

4 Held in water with mosquito larvae from 25 Jun until dissected on 2-4

September.

Weight Gain. — Teneral females averaged 21.3 ± 2.8 mg (N=20) and males

averaged 22.4 ± 3.0 mg (N=13). Newly emerged adults fed 10 late instar

mosquito larvae/day gained an average of 48% of their body weight in the first 10

days of observation.

Weight Loss. — Beetles placed in dry sand were found consistently on the

surface, whereas those under moist sand remained buried. Results of the weight

loss experiments are summarized in Figure 2. Body weights of the 100% R.H. and

moist sand groups remained constant, while the beetles exposed to dry sand or

lower humidities lost weight. The 50% R.H. group lost weight at about the same

rate as the dry sand group, and rate of weight loss was inversely proportional to

relative humidity. Regardless of the conditions of the experiment, all beetles that

died had an average body weight of 19.8 ± 2.4 mg. There were no significant

differences between the 100% R.H. and wet sand groups, nor between the dry sand

and 50% R.H. groups. Otherwise, all other groups had significantly different slopes

(Table 1).

Flight activity. — Flight from the test containers was observed in the

laboratory during the fall, winter, and spring (Table 2). Beetles which flew during

the flight tests typically climbed to the tip of the vertical stick, oriented with its

back to the sun for up to 15 minutes, and then initiated flight. Some of the newly

emerged, laboratory reared beetles were observed to attempt flight in indoor

containers covered with clear plastic. This occurred in early March, after several

days under dry but cool conditions (20-24°C). Beetles exposed to the sun on bare

soil (March & October) attempted to fly sooner (i.e. in less than 15 min.) than

beetles in water filled jars. In October 1987, 4 out of 6 beetles flew from soil, and

2 out of 4 beetles flew from soil in March 1988. No beetles were observed to fly

from the surface of the water in any of the flight tests.

In the March 1987 test under overcast skies, four beetles flew from the heated

water bath (at >31°C), while none of the beetles in the unheated water (maximum

Agabus disintegratus (Crotch)

147

22°C) attempted flight although three were observed to move up the stick and out

of the water. All the beetles in the heated group moved up and down the stick or

swam about their container as the water temperatures rose above 28°C. After

water temperatures exceeded 32°C, the beetles spent the majority of time above the

water on the stick. The unheated group usually remained below the surface

grasping the stick near the sand except for occasional surfacing to breathe, but

seldom climbed on the stick.

In April 1987, both reproductive (2 males, 3 females) and teneral adults (8

males, 4 females) flew. None of the beetles collected from summer sod samples were

observed to fly, in spite of being exposed to high water temperatures.

Termination of Diapause. — Females collected in June and held with a male in

either soil or water under a photoperiod of 12L:12D for 10 weeks exhibited

reproductive activity upon dissection as indicated by conspicuous oocyte

development and the presence of sperm in the spermathecae (Table 3). No eggs

were observed on the Elodea in the holding containers under any of the conditions

tested. Beetles under 12L:12D at both 13 and 18°C consumed 33.0 ± 4.2 S.D. and

33.2 ± 5.6 mosquito larvae, respectively, while those held at 16L:8D (15°C)

consumed 28.6 ± 6.4 larvae. Differences in number of larvae consumed were not

significant (Student's t-test, P > 0.05).

DISCUSSION

In the Central Valley of California, A. disintegratus is adapted to breeding and

developing during the fall and winter months. Artificial inundation of pond 78 has

occurred near the first week of October for the last two years, and adults have been

observed mating by mid-October. In addition, dissection of adult females a few

days after flooding in early October revealed developing oocytes and sperm in the

spermathecae. Thus, reproductive activity begins in the early fall in these

artificially flooded ponds. Under natural conditions, however, standing water, such

as temporary ponds and vernal pools formed from precipitation and runoff, usually

does not accumulate in the Central Valley until late November or December. Thus

artificial flooding at the refuge allows for earlier development for A. disintegratus.

Whether this is advantageous or not is unclear at this time, because despite

observations of early reproductive development and the appearance of 1st instar

larvae in late October, third instar larvae have not been seen until January. Since

water temperatures are relatively mild through the fall, the occurrence of 3rd instar

larvae should be expected much earlier in the season. We have no evidence at this

time whether mortality or some physiological factor in larval development accounts

for the lack of later instars.

The artificially manipulated water systems in waterfowl management areas

may be an advantage for this species by providing more optimal conditions for

adults during summer diapause, particularly in bermudagrass. Body weights of

beetles collected through the dry summer remained relatively constant, which is

apparently related to high humidities in the sod microhabitat. Maintenance of

body weight through the summer may be partly influenced by the summer

irrigations of the field. Summer irrigations not only provide standing water for a

short period but more importantly, remoisten the rhizome debris layer where the

beetles reside. In addition, growth and transpiration by the plants moderate the

high temperature extremes that are common during the summer months in this part of

California. Total rainfall for Colusa (15 km SW from the study site) during June,

July, August and September, averaged 3.1, 1.0, 1.3, and 3.8 cm, (U.S. Weather

Bureau summaries, 1980-1985). During 1987, however, no precipitation was

Quaest. Ent. , 1990, 26(2)

148

Garcia, Hagen and Voigt

recorded for the entire summer. Such small amounts of summer precipitation are

normally insufficient to moisten the soil, especially under a thick canopy of

bermudagrass. Consequently, moisture levels where these beetles reside would be

dependent mostly on water retained in the soil from the spring or from water added

by summer irrigations.

In the laboratory, adults placed in moist conditions maintained a constant

body weight, but under totally dry conditions lost weight rather quickly and died

within 15 days. This indicates that the microhabitat where the beetles reside

cannot become completely dry even for relatively short periods during aestivation.

Natural habitats with wetland macrophytes such as cattails ( Typha spp.) and

bulrushes ( Scirpus spp.), from which A. disintegratus has been recovered (Garcia

and Hagen 1987), may provide a more suitable microhabitat than artificial habitats

with bermudagrass. The soil around these more robust plants more readily forms

cracks at the base of the stems which would allow beetles access into the deeper

zones of the soil-root interface.

Beetles were able to fly throughout the fall, winter and spring, which

corresponds to periods when standing water is likely to be available. During the

summer, beetles did not fly, even after exposure to harsh conditions. The increased

activity of beetles subjected to dehydration indicates that although beetles are

unable to fly from drying conditions, they probably do crawl about in the field

seeking more favorable microhabitats. This is highly adaptive in that other natural

water sources are likely to be unavailable, and flight dispersal at that time would

deplete fat and water reserves with little possibility of achieving success. By

seeking out relatively insulated refugia for diapause, A. disintegratus avoids the

extreme heat and dryness of the Central Valley in summer.

Female A. disintegratus dissected after being held in the laboratory with males

in either soil or water under a photoperiod of 12L:12D were reproductively active,

whereas the females held at 16L:8D were not, suggesting that diapause is

terminated by a short-day photoperiod. Termination of summer diapause by short

photoperiod has been demonstrated in several other insect species, including four

different orders (see reviews in Tauber et al , 1986; Brown and Hodek, 1980). The

decline in average body weight and fat reserves in late summer prior to inundation

of the fields further suggests that termination of diapause begins while beetles are

still in the dry state.

Several species of dytiscids exploit more permanent water sources, and

through aestivation, A. disintegratus avoids competing with these species. In turn,

A. disintegratus is able to exploit immediately temporary aquatic habitats as soon

as they become available, without having to expend energy to seek them out

through dispersal. Seasonal flight activity allows dispersal of reproductive and

teneral adults during periods when natural water sources are more likely to occur.

ACKNOWLEDGEMENTS

We thank Kim Des Rochers and Aaron Goldberg for laboratory assistance for

some of the experiments. R.B. Reno and Lee Ashford of the California Department

of Fish and Game at the Gray Lodge Waterfowl Area provided logistical

assistance. This study was supported in part by special funds for mosquito

research in California.

Agabus disintegratus (Crotch)

149

REFERENCES CITED

Brown, V.K., and I. Hodek (eds.) 1983. Diapause and life cycle strategies in

insects. Dr W. Junk Publishers. The Hague, Boston, London.

Garcia, R. and K.S. Hagen 1985. The occurrence of adult dytiscids in dry

waterfowl refuge ponds in California (Coleoptera:Dytiscidae). Coleopterists

Bulletin, 39: 391-392.

Garcia, R. and K.S. Hagen 1987. Summer dormancy in adult Agabus disintegratus

(Crotch)(Coleoptera:Dytiscidae) in dried ponds in California. Annals of the

Entomological Society of America, 80: 267-271.

Marasuilo, L.A., and J.R. Levin 1983. Multivariate statistics in the social

sciences. Brooks/Cole Publ., Monterey.

Peterson, A. 1959. Entomological techniques. How to work with insects.

Edwards Brothers, Inc., Ann Arbor.

Tauber, M.J., C.A. Tauber, and S. Masaki 1986. Seasonal adaptations of insects.

Oxford University Press, New York and Oxford.

Young, F.N. 1960. The water beetles of a temporary pond in southern Indiana.

Proceedings of the Indiana Academy of Sciences, 169: 154-169.

Quaest. Ent. , 1990, 26(2)

ODONATE PREDATION AS A FACTOR INFLUENCING

DYTISCID BEETLE DISTRIBUTION AND COMMUNITY

STRUCTURE

David J. Larson

Dept, of Biology,

Memorial University of Newfoundland

St. Johns, Nfld.

A1B 3X9

CANADA

Quaestiones Entomologicae

26: 151-162 1990

ABSTRACT

Dragonfly larvae and predacious water beetles (Dytiscidae) are abundant

predators in many shallow lentic habitats. The distributions of members of these

two groups differ somewhat with odonates dominating in more open and

permanent sites while dytiscids are more abundant in habitats of less stability and

denser vegetation. It is postulated that predation of odonate larvae on dytiscids,

especially the larval stages, is at least a contributory factor to this partitioning.

Evidence in support of this hypothesis is drawn from general considerations of the

biology and behaviour of the two groups, literature records, collecting experiences

and a study that measured odonate density and the prevalence of dytiscids as food

items in their guts. In certain Newfoundland bog pools, the density of odonate

larvae is adequate to eliminate vulnerable dytiscids in a matter of days.

Mechanisms by which dytiscids can avoid odonate predation are discussed.

INTRODUCTION

What governs the nature of natural communities? This question has generated

much interest among biologists. The major conclusion to come out of the

considerable research conducted on the question seems to be that there is no simple

answer. Historical factors determine the suite of species present in a fauna that can

interact potentially in communities. Abiotic tolerances determine which set of

species can occur in a given physical arena. Within this arena, biotic interactions

such as predation and competition further affect species densities and dispersion.

Add to this niche and trophic specialization and temporal and behavioral

similarities or differences among species, and the complexities of community

organization are readily apparent.

Water beetle workers have recognized characteristic associations of water

beetles species, at least within regional faunas. There is a long history of these

associations being described and related to habitat characteristics. Recently,

several authors (e.g. Larson 1985, Ranta 1985, Flechtner 1986, and Cuppen 1986)

have used numerical techniques to define communities and relate species

distributions to habitat parameters. Most studies of water beetle communities

have emphasized the importance of physical, chemical and vegetal features of

habitats as determiners of beetle distribution, although some authors have

considered also predation and competition effects (Nilsson 1986, 1988).

There are two aspects to the problem of whether predation pressures have had

a role in shaping dytiscid communities. First, past predation pressures may have

been responsible for shaping aspects of the ecology, behaviour and morphology of

152

Larson

dytiscids, for example the development of defensive glands that produce a complex

array of defensive substances (Dettner 1985), or the evolution of protective

coloration by species in certain types of habitat (Young 1960). However, if these

traits have become fixed because of this pressure, no direct evidence is left with

which to demonstrate the relationship so that selection pressures for the evolution

of the feature must be inferred based on concepts of its function. The second

aspect involves current and continuing interactions where predation effects and

outcomes are not stabilized and vary depending upon the conditions under which

they occur. These sorts of interactions are amenable to observation and this aspect

of predation interaction is considered here.

It is necessary, first, to establish if predators in aquatic systems are capable of

exerting enough pressure on prey populations to modify the prey species mix,

population structure, and/or their morphological traits. There is abundant evidence

to indicate that many littoral habitats are predator dominated systems. When

present, fish generally have a major impact on invertebrate communities, occupying

'the role of top predator by virtue of their large size and activity (Gilinsky 1984).

Fish predation on invertebrates may result in elimination of species from a system,

change the population densities, or change the morphology and behaviour of prey

(Stenson 1978, Nilsson 1981, Morin 1984). Wilson (1923) was concerned about

the impact of water beetle predation on fish in fish-culture ponds. However, he

came to the conclusion that beetle larvae and adults are eaten freely by many fish,

and that all beetle larvae and adults of smaller species constitute a very important

item of fish food. Wilson reviewed and supported the observation made by many

authors that dytiscids tend to be much less abundant and diverse in large ponds and

lakes than in smaller bodies of water. With qualification, this observation is still

generally valid (Larson 1985, Ranta 1985). It is probable that fish predation in

larger bodies of water has a bearing on this distribution.

While fish are undisputably important predators, certain factors limit their

distribution, e.g., drying of habitat, oxygen depletion, freezing, and dense debris

or plant structure in the habitat. Fish are absent from many northern lentic habitats.

In such habitats, can other groups of predators exert the same type of impact on

invertebrate communities on which they prey? If so, which groups of predators are

likely to do this? Within the Insecta, Odonata, Hemiptera (which will not be

considered here) and Coleoptera are especially diverse and abundant predators in

lentic habitats. The relative abundance and success of each group varies from site

to site. The purpose of this paper is to evaluate the evidence for predation by

odonate larvae as a factor affecting the distribution of dytiscid beetles.

ODONATES AS SIGNIFICANT PREDATORS OF DYTISCIDS?

Evidence from literature and observation

Wissinger (1988) found large odonates increased in numbers to become top

predators in the absence of fish. Benke (1976) recorded dragonfly larvae in very

high densities in a South Carolina pond, where he concluded they were capable of

rapidly annihilating their prey (most animals of suitable size [Pritchard 1964])

which survived only because they found refuges. The high density of odonate

larvae observed by Benke is not unusual: for example Ball and Hayne (1952),

Beatty and Hooper (1958) and Macan (1964) reported high densities of odonate

larvae in shallow lentic habitats. In fact, the standing crop of odonates commonly

may exceed that of their prey (Benke 1976). Thorp and Cothran (1984) showed

that dragonfly predation can influence significantly a benthic community, primarily

by changing prey density rather than community diversity. Density dependent

Odonate Predation

153

Table 1 . Comparison of predation strategies of odonate and dytiscid larvae

effects may be major factors controlling odonate communities (c.g., Johnson et. al. ,

1985) and may result in both inter- and intraspecific asynchrony in growth so that

odonates of a range of sizes are present to crop prey of a range of sizes.

Larson and Colbo (1983) suggested that odonates are significant predators

of dytiscids. This idea derives from consideration of population densities,

feeding methods and life history patterns of both groups. The most important

interactions are probably between the larval stages because this is the only active

stage of odonates to be in the water. Adult beetles appear to be fairly well

protected from predation by size, hard cuticle and perhaps defensive secretions;

larvae are apparently more vulnerable (Pritchard 1964, Griffiths 1973). Table 1

summarizes major differences in the predation strategies of odonate and dytiscid

larvae.

Several lines of observation lend support to the idea that odonates negatively

affect dytiscids. A few examples will illustrate this.

A. Numerous authors made the general observation that dytiscids are scarce in

large lakes and are most numerous with the greatest diversity in seasonal habitats,

newly formed ponds and the flooded margins or zone of dense emergent vegetation

of larger ponds (Galewski 1971, Nilsson 1984, Larson 1985). Stability does not

seem to favour many species. On the other hand, dragonfly larvae are generally not

numerous or diverse in highly variable habitats, possibly due to their inability to

cope with habitat drying (Fischer 1961) or freezing. In other words, there is

somewhat of a habitat segregation between odonates and dytiscids. Historical and

physical characteristics may limit the range of habitats occupied by odonates, but

why are dytiscids not more successful in the habitat types occupied by odonates?

B. In 1986, John Carr and I sampled a series of small moraine ponds on the

western slope of the Nahanni Mountains, Yukon Territories, along the Cantung

Quaest. Ent.. 1990, 26(2)

154

Larson

Table 2. Frequencies of co-occurrence of major taxa in habitats from which

dytiscid beetles have been collected (after Flechtner 1986, Fig. 5).

Road. The ponds ranged in elevation from the coniferous forest zone, through

birch-lichen habitat to alpine habitats. Ponds within and near the coniferous forest

zone had abundant odonates but few beetles which were largely confined to

emergent vegetation at the very edges. Higher elevation ponds were superficially

, similar to those of lower elevations but odonate density was dramatically less

1 while beetles were much more numerous. These ponds reflect a common faunal

pattern for in general, with increasing elevation or latitude (at the highest latitudes),

and deteriorating climates, beetles tend to be found more widely distributed in

lentic habitats and occupy more open waters. For example, in barren alpine or

arctic pools agabines and hydroporines occupy a wider range of depths and

habitat types than is generally observed for lower elevation or lower latitude

populations. Odonates are usually absent or in very low densities in such pools.

C. Some types of dytiscids occur regularly in habitats with dense populations

of odonate larvae. These include: species of Dytiscus and Cybister which, because

of their very large size, probably enjoy a switch in predator advantage;

thermonectines, the larvae of which are pelagic and occupy a different zone than the

dragonfly larvae; and very small species of dytiscids ( e.g ., bidessines) which

generally occur among very dense detritus, in moss or algal mats, or in very shallow

water right at the water's edge - zones in which odonate populations are low.

Larson (1985) pointed out that the size distribution profile within the dytiscid

Odonate Predation

155

faunas of Alberta and Florida differed, with very small species comprising a

proportionally much larger element in the Florida fauna. A partial explanation could

be that the rich odonate (as well as fish) fauna of Florida selects for dytiscid forms

that escape predation through adaptation to microhabitats which provide refuge

from predators.

D. Flechtner (1986) challenged the suggestion of Larson and Colbo (1983) of

a negative correlation between dytiscids and odonates. However, the data of

Flechtner's Fig. 5 (reproduced in Table 2) which give percentage occurrence of

various major taxa in collections with dytiscids, provide support for Larson and

Colbo. It is significant that fish have the lowest co-occurrence (ca 30%) with

dytiscids, followed by odonates (ca 38%). These data do not indicate how many

samples contained either fish or odonates but not dytiscids, which would even

further lower the co-occurrence rates.

Dytiscid beetles and larvae are subject to different predation pressures, with

the larvae more vulnerable to predators. Life history theory predicts that selection

acts to reduce the duration of the stage with the higher mortality rate (Wilbur

1980). This appears applicable to dytiscids, for in general, the larval stage is

relatively short compared with the life span of adults. For example, even for the

many species the life histories of which are not known, adults can be collected

throughout much of the year while larvae appear to occur for a shorter and more

specific period. This pattern of life history probably has more to do with habitat

seasonality than predation (Larson 1985), but also could be reinforced by

predation pressures.

A QUANTITATIVE ESTIMATE OF PREDATION

Diverse groups of insects, such as dytiscids and odonates, can be expected to

interact in a variety of ways depending upon the taxa and habitats involved. A

recent study on insect communities in a series of ombrotrophic bog pools (Larson

and House 1990) provided an opportunity for a quantitative assessment of

odonate predation on dytiscids in this habitat. The primary objective of the study

was to determine abundance and distribution patterns of macroscopic animals

within the pool system, and to interpret these in relation to habitat features and

interaction patterns between taxa. For the purposes of this discussion, only the

patterns observed for odonates and dytiscids will be discussed. Full details of the

habitat and arthropod community structure are published elsewhere (Larson &

House 1990).

The study was carried out on an ombrotrophic, domed bog located on the

Avalon Peninsula 20 km south of St. John's, Newfoundland. The bog was treeless

with the principal vegetation being sphagnum mosses, ericaceous shrubs, rushes

and sedges. The bog contained in excess of 200 pools ranging in surface area from

less than one to greater than 500 m2 The pools were divided into four size classes

based on their surface area, namely: Class A - > 100 m2; B - 10 to 100 m2: C - 1 to

10 m2; and D < 1 m2). In spite of the size differences, the pools were similar in

water quality and form. Pool depth was positively correlated with surface area.

Water level fluctuation was similar across all pools which meant that some of the

smaller, shallower pools lost visible water during dry periods.

Pools were sampled by collecting all insects within a quadrat of 1 m2. The

entire area of pools less than 1 m2 was sampled and the resulting counts transformed

to numbers per m2 Regardless of pool size, one edge of the quadrat was always

formed by the pool bank so that edge effects were standardized across

Quaest. Ent., 1990, 26(2)

156

Larson

Table 3. List of taxa collected and their density (number rcr2), (standard error) and

prevalence (%) in pools of the four size classes. The numbers for Dytiscidae

include pooled larvae and adults. * - mean density less than 0. 1 nr2.

POOL SIZE CLASS

DC BA

(continued on next page)

Odonate Predation

157

Table 3 (continued)

POOL SIZE CLASS

C B

COLEOPTERA

Dytiscidae

Quaest. Ent., 1990, 26(2)

158

Larson

Table 4. Odonate density and predation rates on dytiscids in bog pools

samples. Sampling was conducted by repeatedly sweeping the quadrat with an

aquatic net of 1 mm mesh, then visually picking insects from the sweepings. A

quadrat was repeatedly swept until no further specimens were found.

A list of the species of Odonata and Dytiscidae and the density of each in

pools of the four size classes is presented in Table 3. Beetle and odonate densities

(logten (number nr2+l)) were inversely correlated (r = -0.39, p < .01). Beetle

populations were densest in the smallest pools and decreased rapidly with

increasing pool size, while odonate populations were more than ten times as dense

in the A pools as in the D pools.

Size of adult beetles was positively correlated to pool size (r = 0.56, p < .01).

The correlation was calculated between mean adult size of each species represented

in each pool and pool surface area. Thus, the occurrence of a species in a pool was

treated as a single observation. If the size of each individual and the pool size in

which it was found were correlated, the relationship would be even stronger.

Generally, small species occurred in small pools: however, Hydroporus badiellus

Fall and H. obscurus Schaum also occurred in low frequencies in the A and B pools.

This was a result of including a length of bank in each sample, because these species

occur in the moss at the water's edge. If the samples were taken farther from the

bank so as to exclude these peripheral species, the pool size - beetle size

correlation would strengthen.

To determine if the odonates actually were preying upon dytiscids, gut

contents of 500 odonate larvae, representing the five most abundant species

collected from pools of all sizes from May through October, were examined. Some

specimens of all species were found with dytiscid larvae in their guts and the

Aeshna species also contained hydrophilid and hydroporine adults. Prevalence of

dytiscid remains in the gut contents of these species ranged from 0.7 % in

Leucorrhinia hudsonica Selys to 15% in Aeshna sitchensis Hagen. If prevalence of

Odonate Predation

159

dytiscid remains in the gut of each species of dragonfly is multiplied by the

density of the respective species in pools of each size class and these values then

summed, an estimate of the predation rate of dragonflies on dytiscids can be

obtained (Table 4). If it is assumed that residence time of material in a dragonfly

gut is one day (indicated by Pritchard 1964) then this figure represents the daily

predation rate of dragonflies on dytiscids averaged over the ice-free season - the

period over which odonate larvae were collected for gut content analysis.

Dividing the mean density of dytiscids by this predation rate gives an estimate of

the length of time it would take for odonates to eliminate the dytiscid population

from each habitat type (Prey clearance rate, Table 4). This is highly simplified

with many possible sources of error such as the fact that "dytiscids" includes both

adults and larvae and that adults may not in fact be at risk to predation.

Nevertheless, the figures still indicate that odonates exert a powerful predation

pressure on dytiscids and that in the presence of a dense odonate population, low

dytiscid numbers may be explained by predation.

In this study, the most abundant dytiscid in the larger pools was Ilybius

pleuriticus LeConte (Table 3). This is the largest North American Ilybius (Larson

1987), and occurs in deeper and more open water habitats than other members of

the genus. Its larvae crawl around on the substrates, generally in habitats with dense

odonate populations. Why are they not annihilated by dragonflies? I. pleuriticus

larvae are distinctive among known North American Ilybius larvae (unpublished

data) in that they have a very bold pattern of longitudinal stripes extended the

length of the dorsal surface of the body. Perhaps this striped pattern makes the

larvae more difficult for an odonate larva to hit with a visually aimed labial strike.

By itself, this explanation is not compelling. However, within the same habitat are

two species of Aeshna, A. eremita Scudder and A. subarctica Walker. Aeshna larvae

probably behave more like dytiscid larvae than do other dragonfly larvae, i.e. they

are rather active and move around considerably. As cannibalism occurs amongst

odonate larvae, the small Aeshna larvae should be at risk to odonate predators due

to their movement. Larvae of both Aeshna species are very boldly coloured: A.

eremita larvae are black with the middle third of the body pale yellow; A. subarctica

larvae are longitudinally striped, similar to that of larval I. pleuriticus. The color

pattern of both Aeshna species is very disruptive to the human eye. These

markings are strongest on the smallest larvae, tending to obliterate in larvae of 10

to 15 mm length, with the larger specimens more uniformly dark.

The most probable function of color pattern is to provide protection from

visually hunting predators. If the predators were vertebrates, it would seem most

likely that the predation pressure would become more intense as larvae became

larger thus large larvae should possess protective coloration. But it is the small

larvae that have the disruptive color pattern and this pattern disappears at about

the size that larvae become large enough to escape odonate predation. Thus it is

probable that strikingly disruptive color patterns are a defense against odonate

predation.

EXPERIMENTS AND CAVEATS

If dragonflies are important predators on dytiscids, removal of odonates from

a habitat should result in an increase in beetle density and also perhaps in a

broadening of range of habitat occupied by at least the larvae. Such an experiment

has not been conducted specifically to test the impact of odonates on beetles.

However, Benke (1978) did a removal experiment in which odonate populations

Quaest. Ent., 1990, 26(2)

160

Larson

were reduced within pond enclosures and he noted an increase in prey abundance

and survivorship of remaining odonates.

Laboratory rearing confirms the finding from gut content analysis: odonates

prey on smaller dytiscid larvae and strike at adults although they may have

difficulty in handling them (Pritchard 1964, Griffiths 1973). However, laboratory

studies should be done to determine patterns of interaction occurring across a

range of sizes of dytiscids and odonate larvae. Is there a size of dytiscid adult or

larvae which is too large or active for particular odonate species to capture or

handle? Are the tables turned by large dytiscids? Large larvae of species of

Dytiscus and Cybister can prey upon large odonate larvae. For example, third

instar larvae of D. alaskanus Balfour-Browne successfully attacked Aeshna larvae

that had been thrown back into a pond after being collected in a dip net (B.C.,

Cassiar Road km 723, July 18, 1987). R. Trottier (pers. comm., 1987), studying

the large odonatid, Anax junius Drury, in southern Ontario thought that Dytiscus

larvae were preying upon the dragonfly larvae. Compared to dytiscids, odonates

hatch at a small size and grow slowly through a large number of instars. Probably

these small odonates are suitable prey for many dytiscids.

The extent of dytiscid larval predation on odonates will be difficult to

determine. Because dytiscid larvae feed on prey fluids, predation could be

determined only through direct observation or by serological analysis of gut

contents. In any event, high predation rates on odonate larvae may not have a major

effect on trophic patterns. There is evidence that many habitats are overstocked

with dragonfly larvae such that cannibalism and competition are severe and

limiting. In such situations, mutual predation between dytiscids and odonates

probably would favour the odonates for the lower starting populations of

dytiscids put them at a disadvantage. Based on consideration of predation

characteristics (Table 1), it is predicted that under conditions of low prey density

odonates will out-compete dytiscids. This could be tested in the lab.

Although it has not been proven that odonate larvae have a major impact on

dytiscid distribution and abundance, there is much evidence to suggest this is so.

Odonate predation must be less important than physical suitability and trophic

opportunities as factors structuring lentic dytiscid communities. However,

predation and competition pressures within shallow lentic habitats are significant

factors shaping resident communities and dytiscids can not be immune to these

forces. Many of their behaviours and adaptations are likely to be responses to

such biotic pressures.

ACKNOWLEDGEMENTS

I thank Peter Genge, Nancy House and Marie McCarthy for their assistance in

field sampling and odonate gut content analysis, and Margaret Larson and Dr. J.

Pickavance for reviewing the manuscript. This study was supported by an

operating grant from the Natural Sciences and Engineering Research Council of

Canada and by the Canadian Forestry Service.

REFERENCES CITED

Ball, R. C. and D. W. Hayne. 1952. Effects of the removal of the fish population on

the fish-food organisms of a lake. Ecology, 33: 41 - 48.

Beatty, L. D. and F. F. Hooper. 1958. Benthic associations of Sugarloaf Lake.

Papers of the Michigan Academy of Sciences, 43: 23 - 106.

Odonate Predation

161

Benke, A. C. 1976. Dragonfly production and prey turnover. Ecology, 57: 915 -

927.

Benke, A. C. 1978. Interactions among coexisting predators - a field experiment

with dragonfly larvae. Journal of Animal Ecology, 47: 335 - 350.

Cuppen, J. G. M. 1986. The influence of acidity and chlorinity on the distribution

of Hydroporus species (Coleoptera, Dytiscidae) in the Netherlands.

Entomologica Basiliensia 11: 327 - 336.

Dettner, K. 1985. Ecological and phylogenetic significance of defensive

compounds from pygidial glands of Hydradephaga (Coleoptera).

Proceedings of the Academy of Natural Sciences of Philadelphia, 137: 1 56 -

171.

Fischer, Z. 1961. Some data on the Odonata larvae of small pools. Internationale

Revue des Gesamten Hydrobiologie und Hydrographie, 46: 269 - 275.

Flechtner, G. 1986. Association analysis of water beetle communities (Coleoptera:

Dytiscidae et Haliplidae). Entomologica Basiliensia 1 1: 297 - 308.

Galewski, K. 1971. A study on the morphobiotic adaptations of European

species of the Dytiscidae (Coleoptera). Polskie Pismo Entomologiczne, 41:

488 - 702.

Gilinsky, E. 1984. The role of fish predation and spatial heterogeneity in

determining benthic community structure. Ecology 65: 455 - 468.

Griffiths, D. 1973. The food of animals in an acid moorland pond. Journal of

Animal Ecology, 42: 285 - 293.

Johnson, D. M., P. H. Crowley, R. E. Bohanan, C. N. Watson, and T. H. Martin.

1985. Competition among larval dragonflies: a field experiment. Ecology,

66: 119 - 128.

Larson, D. J. 1985. Structure in temperate predacious diving beetle communities

(Coleoptera: Dytiscidae). Holarctic Ecology 8: 18 - 32.

Larson, D. J. 1987. Revision of North American species of Ilybius Erichson

(Coleoptera: Dytiscidae). with systematic notes on Palaearctic species.

Journal of the New York Entomological Society, 95: 341- 413.

Larson, D. J. and M. Colbo. 1983. The aquatic insects: biogeographic

considerations, pp. 593 - 677 in South, G. R. (Ed.), Biogeography and

Ecology of the Island of Newfoundland. Monographiae Biologicae 48. Junk,

The Hague.

Larson, D. J. and N. House. 1990. Insect communities of Newfoundland bog pools

with special reference to Odonata. Canadian Entomologist (in press).

Macan, T. T. 1964. The Odonata of a moorland fishpond. Internationale Revue der

gesamten Hydrobiologie und Hydrographie, 49: 325 - 360.

Morin, P. J. 1984. The impact of fish exclusion on the abundance and species

composition of larval odonates: results of a short-term experiment in a North

Carolina farm pond. Ecology, 65: 53 - 60.

Nilsson, A. 1984. Species richness and succession of aquatic beetles in some

kettlehole ponds in northern Sweden. Holarctic Ecology, 7: 149 - 156.

Quaest. Ent., 1990, 26(2)

162

Larson

Nilsson, A. 1986. Community structure in the Dytiscidae (Coleoptera) of a

northern Swedish seasonal pond. Annales Zoologica Fennici, 23: 39 - 47.

Nilsson, A. 1988. Larval consumption rates, interspecific predation, and local

guild composition of egg-overwintering Agabus (Coleoptera, Dytiscidae)

species in vernal ponds. Oecologia, 76: 131 - 137.

Nilsson, B.-I. 1981. Susceptibility of some Odonata larvae to fish predation.

Verhandeln Internationale Verein Limnology, 21: 1612-1615.

Pritchard, G. 1964. The prey of dragonfly- larvae (Odonata; Anisoptera) in ponds

in northern Alberta. Canadian Journal of Zoology, 42: 785 - 800.

Ranta, E. 1985. Communities of water-beetles in different kinds of waters in

Finland. Transactions of the Philadelphia Academy of Science, 137: 33 - 45.

Stenson, J. A. E. 1978. Differential predation by fish in two species of Chaoborus

(Diptera, Chaoboridae). Oikos, 31: 98-101.

Thorp, J. H. and M. L. Cothran, 1984. Regulation of freshwater community

structure at multiple intensities of dragonfly predation. Ecology, 65: 1546 -

1555.

Wilbur, H. M. 1980. Complex life cycles. Annual Review of Ecology and

Systematics, 11: 67 - 93.

Wilson, C. B. 1923. Water beetles in relation to pondfish culture, with life

histories of those found in fishponds at Fairport, Iowa. Bulletin of the United

States Bureau of Fisheries, Washington, 39: 231 - 345.

Wissinger, S. A. 1988. Life history and size structure of larval dragonfly

populations. Journal of the North American Benthological Society 7: 13 - 28.

Young, F. N. 1960. The colors of desert water beetles - environmental effect or

protective coloration? Annals of the Entomological Society of America, 53:

422 - 425.

PHYLOGENETIC ANALYSIS OF THE FAMILY GYRINIDAE

(COLEOPTERA) BASED ON MESO- AND METATHORACIC

CHARACTERS

ROLF G. BEUTEL

Institut fur Biologie II (Zool.)

RWTH Aachen

D-5 100 Aachen Quaestiones Entomologicae

West Germany 26: 163-191 1990

ABSTRACT

Thirty six characters of the meso- and metathorax of adults of

Spanglerogyrus albiventris Folkerts and other members of Gyrinidae were

examined and analyzed phylo genetically. The acquired data suggest that

Spanglerogyrinae are the sister-group to the remainder of Gyrinidae ; oar-like tibial

processes , feather-like swimming hairs, and the presence of one tibial spur only are

autapomorphies of Spanglerogyrus. Members of Gyrininae are characterized by a

large number of synapomorphic character states. Some of these are: anepisternal-

elytral opening, excavations for the prolegs in repose, paddle-like middle- and

hind legs, swimming lamellae, metanotum extended laterally, metapostnotum

inflected below the scute llum, metasternal transverse-ridge completely reduced,

metafurca arising from the fused medial metacoxal walls, lateral metafur cal

projections reduced, medial metacoxal walls fused, loss of several flight muscles,

loss of Mm. furca-coxalis anterior and lateralis (M 81 and M 82), presence of M.

noto-trochanteralis (M 84). The absence of M. sterno-episternalis (M 72) is

considered as a possible synapomorphy of Gyrinus and Aulonogyrus (+

Metagyrinus, Heterogyrus ?). Orectochilini and the enhydrine genera seem to form a

well-founded monophyletic unit. The following characters are interpreted as

synapomorphies of this assemblage: anterior and posterior walls of tibiae and

proximal tarsomeres connected by cuticular columnae, markedly developed elytral

glossula, median metanotal area only half as broad as lateral parts, metanotum

without membranous area. The modified shape and position of metatarsomeres 4 +

5 is considered as a synapomorphy of the genera Dineutus, Porrorhynchus,

Macrogyrus, Andogyrus and Orectochilini. Consequently, Enhydrini are not

monophyletic. The concealed mesoscutellar lobe in members of Dineutus and

Porrorhynchus is a possible synapomorphy of both genera. The modified shape

and position of mesotarsomeres 4 + 5 is considered as a synapomorphy of a

monophyletic group comprising Macrogyrus, Andogyrus and Orectochilini. The

attachment of the anterior metacoxal wall to the hind margin of the ventral sclerite of

the metathorax is another apomorphic character state that suggests a close

relationship between Orectochilini and Andogyrus (+ Macrogyrus ?). Orectochilini

are characterized by distinctive synapomorphies, some of which are: opening

between elytra and mesothoracic anepisternum narrow, anterior median ridge of the

mesothoracic preepisternum present, lateral internal process of the mesocoxae

trilobed, tendons of M. noto-coxalis (M 40) and M. coxa-subalaris (M 43) arise

from the lateral internal process of the mesocoxa, basalar disc absent, loss of further

flight muscles. A sister group relationship between Orectogyrus and Orectochilus is

indicated by two apomorphic character states: anterior walls of mesocoxae

164

Beutel

attached to the ventral sclerite of the mesothorax, lateral internal process of the

mesocoxae fused with the anepisternum. Whether Heterogyrus is more closely

related to the genera Gyrinus, Metagyrinus and Aulonogyrus, or to the enhydrine-

orectochiline lineage remains open to question.

Zusammenfassung

Sechs und dreizig Merkmale des Meso- und Metathorax von adulten Vertretern der

Gyrinidae, inshesondere von Spanglerogyrus albiventris Folkerts wurden untersucht und

phylogenetisch ausgewertet. Die vorliegenden Daten legen den Schlufi nahe, daft die

Spanglerogyrinae den iibrigen Gyrinidae als Schwestergruppe gegenuberstehen. Ruderartige

Tibialfortsatze, gefiederte Schwimmhaare, und das Vorhandensein von nur einem Tibialsporn sind

autapomorphe Merkmale von Spanglerogyrus. Die Gyrininae sind dutch eine groftere Anzahl

von Synapomorphien gekenzeichnet. Einige dieser Merkmale werden im Folgenden aufgefuhrt:

Offnung zwischen dem mesothorakalen Anepisternum und der Elytrenbasis, Vertiefungen zum

Anlegen der Vorderbeine in Ruhestellung, paddelartige Mittel- und Hinterbeine ,

Schwimmblattchen, Metanotum lateral verbreitert. Metapostnotum unter das Scutellum

eingeschlagen, metasternale Transversalleiste vollig reduziert, Ursprung der Metafurca von den

verwachsenen medialen Hinterhiiftswanden, seitliche Metafurcalarme reduziert, mediate

Hinterhiiftswande verwachsen, Reduktion einiger Flugmuskeln, Mm. furca-coxalis anterior und

lateralis (M 81 und 82) fehlen, M. furca-trochanteralis (M 84) ist vorhanden. Das Fehlen von

M. sterno-episternalis (M 72) wild als mogliche Synapomorphie der Gattungen Gyrinus und

Aulonogyrus (+ Metagyrinus, Heterogyrus?) angesehen. Die Orectochilini scheinen zusammen

mit den Gattungen der Enhydrini eine wohlbegriindete monophyletischeEinheit zu bilden.

Folgende Merkmale werden als Synapomorphien dieser Gruppierung interpretiert : die

vorderen und hinteren Wande der Tibiae und der proximalen Tarsomeren sind durch kutikulare

Verstrebungen fest miteinander verbunden, die Glossula der Elytren ist stark ausgepragt, das

Metanotum ist median nur etwa halb so breit wie lateral, die mediane membranose Zone fehlt.

Die abgewandelte Form und Stellung der Metatarsomeren 4+5 wird als Synapomorphie der

Gattungen Dineutus, Porrorhynchus, Macrogyrus, Andogyrus und der Orectochilini gedeutet.

Daraus folgt, daft die Enhydrini nicht monophyletisch sind. Das verdeckte Schildchen des

Mesoscutellum ist eine mogliche Synapomorphie der Gattungen Dineutus und Porrorhynchus.

Die abgewandelte Form und Position der Mesotarsomeren 4+5 wird als Synapomorphie einer

monophyletischen Gruppe gewertet, die die Gattungen Macrogyrus, Andogyrus, sowie die

Orectochilini umfaftt. Die Verwachsung der vorderen Wand der Metacoxae mit dem Hinterrand

des ventralen Sklerit des Metathorax ist ein weiteres apomorphes Merkmal, das eine nahere

Verwandtschaft zwischen Andogyrus (+ Macrogyrus ?) und den Orectochilini nahelegt. Die

Orectochilini sind durch aussagekraftige Synapomorphien gekennzeichnet: die offnung zwischen

dem mesothorakalen Anepisternum und der Elytrenbasis ist verengt, das Praeepisternum des

Mesothorax ist mit einer anteromedianen Leiste verse hen, der laterale, innere Fortsatz der

Mittelhiifte ist in drei Sektionen aufgefachert, die Sehnen von M. noto-coxalis (M 40) und M.

coxa-subalaris (M 43) entspringen am later alen Fortsatz der Mittelhiifte, die Basalarscheibe fehlt,

weitere Flugmuskeln sind reduziert. Ein Schwestergruppenverhaltnis zwischen Orectochilus und

Orectogyrus wird durch zwei apomorphe Merkmale nahegelegt: die vordere Mittelhiiftwande sind

mit dem ventralen Sklerit des Mesothorax verwachsen, der lateralen Fortsatzes der Mittelhiifte

ist mit dem Anepisternum verwachsen. Ob Heterogyrus naher mit den Gattungen Gyrinus,

Aulonogyrus und Metagyrinus, oder naher mit den Gattungen der Enhydrini und Orectochilini

verwandt ist bleibt ungeklart.

TABLE OF CONTENTS

Introduction

Material and Methods

Characters

Notes about the evolutionary history of Gyrinidae and phylogenetic

165

165

168

conclusions

180

Phylogenetic Analysis of Gyrinidae

165

Acknowledgements 189

References cited 1 90

INTRODUCTION

The purpose of this paper is to reconstruct the evolution of gyrinid meso- and

metathoracic structures, and to use the acquired data to analyse the phylogenetic

relationships within the family. Special emphasize is placed upon the study of

Spanglerogyrus albiventris Folkerts, which was described by Folkerts in 1979,

and placed in a newly erected subfamily Spanglerogyrinae Folkerts.

An outstanding and comprehensive study of the locomotor organs of

Gyrinidae was made by Larsen (1966), and it should be emphasized that Larsen's

work was an indispensable prerequisite for this study. However, Spanglerogyrus ,

which is substantially different from other gyrinids in many features (Folkerts,

1979; Steiner & Anderson, 1981; Beutel, in press, in prep.), was not known at that

time. Moreover, Larsen's purpose was a comparative study of structure and

function, and he did not subject his data to rigorous phylogenetic analysis. Larsen

(1966) did not use cladistic methods (Hennig, 1966), and many of his phylogenetic

statements remain vague.

Despite the great interest in Gyrinidae over a long period, and many brilliant

taxonomic studies (e.g., by Georg Ochs and Per Brinck), a stringent, cladistic

analysis of Gyrinidae is lacking. This study proposes to reconstruct the

evolutionary history of Gyrinidae through examination and phylogenetic

interpretation of meso- and metathoracic characters of adult gyrinids. Emphasis is

placed upon determination of the polarity of character states and on functional

considerations.

MATERIAL AND METHODS

All specimens of Spanglerogyrus albiventris used for this study were

collected by R. E. Roughley and R. G. Beutel at a shaded stream near Evergreen

(Conecuh County, Alabama). The specimens were fixed in Kahle's fluid and

preserved in alcohol. Araldite was used as an embedding medium for microtome

sections. The sections were cut with a glass knife at 2-5 pm and stained in

methylene blue. Drawings were made with an ocular reticule (stereo microscope)

and with the help of SEM micrographs (Cambridge Stereoscan 250 Mk 2).

Representatives of the genera Gyrinus L., Aulonogyrus Motschulsky,

Orectochilus Lac., Orectogyrus Reg., Gyretes Brulle, Andogyrus Ochs, Dineutus

Macleay, Enhydrus Laporte were examined for external skeletal structures.

Specimens of Dineutus assimilis Kirby, Andogyrus colombicus Reg., Gyretes

tricolor Young, Orectochilus villosus Mull., Aulonogyrus coccinus Klug, and

Gyrinus marginellus Fall were examined for both external and internal structures.

Furthermore representatives of all the remaining adephagan groups, and specimens

of Priacma serrata LeConte and Tetraphalerus (undescribed species) (Cupedidae)

were examined.

Out-group comparison is used for the determination of the polarity of

character states whenever possible. A flexible and comprehensive out-group,

comprising terrestrial and aquatic families of Adephaga, and Cupedidae, was

chosen, as recommended by Beutel (in press). The In-group or character-correlation

criterion is used for structures of adults of Gyrinidae that either are not present or

are radically different in the remaining Adephaga. As strong evidence is given for

the monophyly of Gyrininae (sensu Folkerts) and a sistergroup relationship

Quae st. Ent., 1990, 26(2)

Table I : Distribution of character states among genera of Gyrinidae. See text for explanations of character states.

166

Beutel

O — — o- o* — — 1 — 1 <>•

O — — O- 0s- O- —I O* O* — — — —

O' — o • o • o • — < o • O' ' — (N (N (N

O — — O- O- O- fN O' O' (N (N (N <N

O — — — O- — — — OI(NCN(NC\|

O O O O o-

— OOOo-OOOOOOOO

*** ********

-M(N(NC"'fNrJM(N(N(NM(N

— OOOo-OOOOOOOO

O - • o-

O O O o- o- o- — o- o- O 1 — ' — —

O O O o* o- o- O o- o- cn m m m

O O O o* O' o- O o- o- o — — O

O O O 0* o- o- O o- o- O — — 1 <— 1

o

O — — o-

O' O' o- -i (N M M

OOOOOO — — OOOO —

6*3

-s: OJ 03

sj S co 3

s: s ^

>•> ^ ^ -3

-5: °c

P -2 ^ x £ jo s

o 2 ^ ^ ^ ^ *2 ^

$ *2 s: ~ C >

3 S

^ 3

T-

Oc -

Oc i3

O

00 Co

2 2

t3

£

&■

O ’

2,3 = apomorphic, steps 2, 3;

3a, 3b = substates of apomorphic character state 3;

1*, 2* = apomorphic character states which have evolved independently from 0 or 1.

Genera Character and Character States

Phylogenetic Analysis of Gyrinidae

167

O — ' o- o- ^ c^- o- — — —

O O O o- c^- o- O o- O *— i — —

O — — < 1 o- o- c^- <— i <>• o- i—> — — —

O ^ o- c^-

O ^ o- o- c^- O o- <>• O O O O

O ■ — ' o* <>• o* ’— ' <>• o* > — (N (N (N

O' — ' — — < C^* ’ — (N (N (N (N N (N M

O O O c^- c^- c^- O O — 1 (N <N <N

O — — ' o- <>• (N O- C^- <N <N CN 04

O — 1 <— i c^- <>• o- — c^- o-

O — 1 — 1 o- c^- c^- o* <>•

O o* o- <>• i—i <>• o- — ' — < — —

O O O o- C''* O C“* c^* O ’— 1 ■ '— '

O — — 1 o- o- —

05 g

^ s;

«5; <o> oj >)

O ^ 3

? S ^ hr

• -sr C'CPno sc ^

S.^aSfcs-sfe.allsi:^

to O -iSaztijQ&.S'sa.a.O

a.

il

o

OJ

1 = apomorphic, step 1 ;

2,3 = apomorphic, steps 2, 3;

3a, 3b = substates of apomorphic character state 3;

1 *, 2* = apomorphic character states which have evolved independently from 0 or 1 .

168

Beutel

between Gyrininae and Spanglerogyrus (Beutel, in press, in prep.), character states

that are shared by Spanglerogyrus and some members of Gyrininae are most likely

primitive, whereas the corresponding character state is derived. Spanglerogyrus is

used as an out-group for character evaluation within Gyrininae. Missing data for

some characters resulted because no specimens of Heterogyrus , and no internal

structures of members of the genera Metagyrinus, Enhydrus, Porrorhynchus, and

Macrogyrus were examined. My analysis of these latter taxa is based on

examination of dried specimens on loan from the British Museum (Natural History).

Some character states which are hypothesized as synapomorphies of

Orectochilini in this study, may be found to be synapomorphies of a monophyletic

group comprising Orectochilini and a part of the enhydrine set of genera. There is,

however, little chance that future reevaluation of characters based on studies of

internal structures of members of the genera listed above will affect the positions of

the genera in the cladogram (Fig. 16).

Larsen's (1966) nomenclature is used for the muscles throughout the paper. The

characters below are discussed from anterior to posterior on the body. The

distribution of characters by taxon are shown in Table 1 and a cladogram of the

relationships of the genera is given in Fig. 16.

Plesiomorphic character states are indicated as 0, apomorphic character states

as 1, 2, 3 (transformation series). An asterisk designates an hypothesis of an

independently derived apomorphic character state.

CHARACTERS

Mesoscutellar lobe of mesothorax (Character 1)

Character state O. — The mesoscutellar lobe is triangular and visible externally.

In members of Spanglerogyrus (Fig. 1, 2), Heterogyrus (Brinck, 1955), Enhydrus,

Andogyrus, Macrogyrus, Orectochilus, Orectogyrus, Metagyrinus, Gyrinus and

Aulonogyrus (Hatch, 1926).

Character state 1 . — The mesoscutellar lobe is covered by the elytra at rest. In

members of Gyretes and the enhydrine genera Porrorhynchus and Dineutus (Hatch,

1926).

Polarity rationale. — An exposed mesoscutellar lobe is characteristic of most

adephagan adults and those of Cupedidae. The mesoscutellar lobe is concealed in

members of Hydroporinae (Dytiscidae), Noteridae excl. Phreatodytes (Ueno,

1957), and Haliplidae.

Analysis. — It appears most plausible to interpret character state 1 as a

possible synapomorphy of Dineutus and Porrorhynchus (char. 1. 1), and as

independently derived in members of Gyretes (char. 1. 1*). As an exposed

mesoscutellar lobe is present in members of Orectochilus and Orectogyrus, this

condition has to be assigned to the groundplan of Orectochilini. The monophyly of

Orectochilini is demonstrated by various synapomorphous character states in the

following. This character should not be overvalued, as a concealed mesoscutellar

lobe has evolved several times independently within Adephaga (see polarity

rationale).

Opening between anepisternum and elytron of mesothorax

(Character 2)

Character state O. — No opening between basal elytral margin and

anepisternum. In adults of Spanglerogyrus (Fig. 5).

Character state 1. — A conspicuous, fairly wide and triangular opening

between basal margin of elytron and anepisternum (Larsen, 1966; Fig. 7). In

members of Gyrinus, Aulonogyrus, Dineutus (Larsen, 1966), and Andogyrus.

Phylogenetic Analysis of Gyrinidae

169

Character state 2. — Opening between basal margin of elytron and

anepistemum, narrowed by bulged dorsal parts of anepisternum (Larsen, 1966, Fig.

25). In members of Orectochilini (Larsen, 1966).

Polarity rationale. — Character state 0 is plesiomorphic, as an opening

described above is found neither in members of Cupedidae, nor in members of any

other adephagan group. Conspicuously bulged dorsal parts of the anepistemum are

not found in Spanglerogyrus, and are not described for members of any other

adephagan group. Therefore it appears plausible to consider character state 2 as

derived from character state 1 .

Analysis. — The opening between the basal elytral margin and the anepistemum

allows compression of air into the subelytral space, as described by Larsen (1966).

This uncommon feature is probably correlated with the highly efficient meso- and

metathoracic locomotor system. Character state 1 is a derived groundplan feature

of Gyrininae (char. 2.1). Character state 2 is a synapomorphy of Orectochilini (char.

2.2).

Excavations for reception of the prolegs in repose (Character 3)

Character state O. — No excavations for reception of the forelegs in repose. In

members of Spanglerogyrus (Fig. 5).

Character state 1 . — A distinct concavity extended from hind margin of

prothorax over lateral part of mesothoracic preepisternum, mesepimeron,

metathoracic anepisternum, and elytral epipleuron. In members of Gyrinus,

Aulonogyrus , Metagyrinus, in members of the enhydrine genera, and in members of

Orectochilini (Hatch, 1927; Larsen, 1966; personal observation).

Polarity rationale. — A concavity for reception of the prolegs in repose as

described above is not found in members of Cupedidae or members of other

adephagan groups, and is thus plesiomorphic. Thus character state 1 is apomorphic.

Analysis. — These excavations improve the streamlining of the ventral body

surface when the prolegs are drawn up against the body. Character state 1 is the

result of a complex modification of the ventral body surface and a synapomorphy

of Gyrininae (char. 3.1).

Anteromedian ridge of the mesothoracic preepisternum (Character

4)

Character state 0. — No internal median ridge in anterior region of

preepisternum. In members of Spanglerogyrus , Gyrinus , Aulonogyrus , Dineutus , and

Ando gyrus.

Character state 1 . — A high internal ridge, not marked by external suture, in

anterior region of the preepisternum. In members of Orectogyrus (Larsen, 1966),

Orectochilus, and Gyretes (pers. obs.).

Polarity rationale. — An internal ridge of anterior region of mesothoracic

preepisternum is absent from members of Cupedidae ( Priacma serrata ; Baehr,

1975), and from members of other adephagan families. Thus character state 1 is

apomorphic.

Analysis. — By means of this ridge the area of attachment of M. sterno-

trochanteralis (M 51) is considerably enlarged. The ridge is a synapomorphy of

Orectochilini (char. 4.1).

Quaest. Ent., 1990, 26(2)

170

Beutel

Flexibility of the mesocoxae (Character 5)

Character state 0. — Mesocoxal motility restricted to abduction and

adduction. In members of Spanglerogyrus, Gyrinus, Aulonogyrus, Dineutus,

Andogyrus, and Gyretes.

Character state 1. — Anterior walls of mesocoxae are solidly attached to hind

margin of ventral sclerite of mesothorax. In members of Orectochilus and

Orectogyrus (Larsen, 1966).

Polarity rationale. — Immobilized mesocoxae are not known from members of

Cupedidae or members of any other adephagan group. Thus character state 1 is

apomorphic.

Analysis. — Character state 1 is a synapomorphy of Orectochilus and

Orectogyrus (char. 5.1).

Lateral process of the mesocoxae (Character 6)

Character state 0. — Internal, lateral process as attachment area of lateralmost

part of M. coxo-trochanteralis (M 54). The process is slender in its basal section,

and is extended distally. In members of Spanglerogyrus (Fig. 6), Gyrinus , and

Aulonogyrus.

Character state 1 . — The lateral process is short, fairly broad in the basal part

and extended distally. In members of Dineutus.

Character state 2. — Lateral process of mesocoxae extensive, broad in the

basal section, and extended distally. In members of Andogyrus.

Character state 3. — Lateral process trilobed, and strongly enlarged. In

members of Orectogyrus (Larsen, 1966), Orectochilus, and Gyretes (Fig. 1 1).

Suhstate 3 a. — Anterolateral lobe of process not fused with the anepistemum.

In members of Gyretes.

Suhstate 3 h. — Anterolateral lobe fused with anepisterum. In members of

Orectogyrus and Orectochilus (Larsen, 1966)

Polarity rationale. — Following the in-group criterion, character state 0 has to

be considered as plesiomorphic within Gyrinidae. Character states 1 and 2

represent similar conditions, although the difference in relative size of the process

is distinct. It is quite likely that character states 1 and 2 are intermediate stages

between character state 0 and character state 3. The latter character state is

undoubtedly highly derived. Nothing comparable is known from any member of

other adephagan groups. Substate 3 b is derived from 3 a in correlation with the

immobilization of the mesocoxae (5.1).

Analysis. — The complex trilobed internal process of the mesocoxae is a

synapomorphy of Orectochilini (6.3). Substate 3 b is a possible synapomorphy of

Orectochilus and Orectogyrus (6.3 b). For a phylogenetic interpretation of

character states 1 and 2, study of specimens of H eterogyrus , Enhydrus ,

Porrorhynchus, and Macrogyrus is needed.

Tendons of M. noto-coxalis (M 40) and M. coxa-subalaris (M

43) (Character 7)

Character state 0. — Tendons of both muscles arise from lateral region of

posterior wall of mesocoxae. In members of Spanglerogyrus (Fig. 6), Gyrinus,

Aulonogyrus, Dineutus (Larsen, 1966), and Andogyrus.

Character state 1. — Tendons of M 40 and M 43 (if this muscle present) arise

from median lobe of internal, mesocoxal process. In members of Orectochilini (Fig.

1 1) (Larsen, 1966; pers. obs.).

Phylogenetic Analysis of Gyrinidae

171

Polarity rationale. — A condition similar to character state 1 is not described

for any member of other adephagan groups or members of Cupedidae.

Consequently, character state 1 is apomorphic.

Analysis. — Character state 1 is interpreted as a synapomorphy of

Orectochilini (7.1).

Shape of femur and tibia (Character 8)

Character state 0. — Femora and tibiae not shortened, broadened, and only

very slightly flattened. In adults of Spanglerogyrus (Fig. 12, 13).

Character state 1. — Femora and tibiae markedly shortened, broadened, and

flattened. In members of Gyrininae examined (Bott, 1928; Larsen, 1966; Nachtigall,

1961; pers. obs.).

Polarity rationale. — Nothing similar to the highly specialized, paddle-like

legs is found in any other group of Coleoptera. Character state 1 is thus

apomorphic.

Analysis. — Character state 1 is a synapomorphy of Gyrininae (8.1).

Insertion of the tarsus (Character 9)

Character state 0. — Mesotarsus inserted at distal end of tibia. In members of

Gyrininae.

Character state 1. — Mesotarsus inserted close to base of tibia. Oar-like

tibial projection extended almost parallel to tarsus. In adults of Spanglerogyrus

(Fig. 12, 13).

Polarity rationale. — Character state 1 is apomorphic as a similar condition is

not described for members of any other adephagan group or for members of

Cupedidae.

Analysis. — Character state 1 is autapomorphic for Spanglerogyrus (9.1).

Swimming hairs (Character 10)

Character state 0. — Mesotibiae and mesotarsi with simple, unmodified

swimming hairs. Not found among extant members of Gyrinidae.

Character state 1 — Mesotibiae and mesotarsi with ctenoid or feather like

swimming-hairs, of stem with two rows of side branches. In members of

Spanglerogyrus (Fig. 15).

Character state 2. — Mesotibiae and mesotarsi with swimming blades or

lamellae. In members of Gyrininae (Bott, 1928; Nachtigall, 1962; Larsen, 1966;

pers. observation).

Polarity rationale.— Character states 1 and 2 are apomorphic as nothing

similar is known from members of any other adephagan group or from members of

Cupedidae. It is very unlikely that one of both character states is derived from the

other. Both structures are distinctly different in terms of structure and function.

There is good reason to assume that they have evolved independently from simple,

unmodified swimming hairs (character state 0).

Analysis. — The swimming blades or lamellae are an important part of the

extremely efficient locomotor apparatus of members of Gyrininae, as pointed out

by Nachtigall (1961). At the same time character state 2 is a significant

synapomorphy of this group. Character state 1 is an autapomorphy of

Spanglerogyrus.

Tibial spurs (Character II)

Character state 0. — Two mesotibial spurs. In members of Gyrininae examined.

Quaest. Enl., 1990, 26(2)

172