Tettigoniidae of Australia Volume 3: Listroscelidinae, Tympanophorinae, Meconematinae and Microtettigoniinae

INTRODUCTION

This third volume of the Monograph of Australian Tettigoniidae covers four subfamilies. These are the Listroscelidinae, Meconematinae, Microtettigoniinae and the Tympanophorinae. The latter two are endemic to Australia. The largest taxon in this volume is the Listroscelidinae, which may actually comprise a number of subfamilies. This disparity will be attended to when colleague Piotr Naskrecki completes a major generic revision of the Tettigoniidae. He is undertaking a cladistic analysis of all known tettigoniid genera and his classification will help to settle some of the anomalies with which the tettigoniid specialist has had to work for decades. Therefore, the classification of the Australian Listroscelidinae should be considered as tentative. The Australian component of ‘listroscelidine’ genera will be important, however, in determining the composition and limits of the subfamily as a whole since the group is represented in Australia by eight genera and some forty-two species. These are segregated into five tribes; two of which are already described but three of the five are new and reflect the results of the generic analysis. All known Australian species are primarily carnivorous, feeding on a wide range of insects with some preferring other tettigoniids.

Format

This volume follows a format similar to that of Volume 2. This has proved successful as evidenced by reviews of the volume by Brown (1994), Otte (1994), Ramsay (1995) and Yamasaki (1994). At times, the author has had misgivings about the efficacy of the lengthy descriptions that provide the bulk of this monograph (and others by a variety of authors), especially when they are accompanied by copious illustrations. Indeed, if this series is to be completed in the author’s lifetime, some adjustment of the length of these descriptions will be necessary. These thoughts came after the author was well into this volume and it may be that a decision will be made, in the interest of time, to streamline the descriptions for future volumes.

Citation of full label information continues to irritate some colleagues, especially journal editors, but this is more than made up by the favourable comments by those who have to use it. Many people in non-taxonomic disciplines have spoken of the value in having the precise dates and localities and collectors listed in the text. Ecologists, cytologists, behaviourists and conservationists continually use the data which would otherwise require that they access the collection to obtain. Locality maps are often not adequate when precision is necessary.

It seems that with each volume in this series there is some major technological advance that should make the job of assembling a work of this size easier. In the era of Volume 1, it was word processing and the availability of the scanning electron microscope that made life easier. With succeeding volumes, it was the appearance of computerised analytical programs that could make sound and cladistic analysis more comprehensive and easier, and the use of Global Positioning Devices (GPS) that are commonplace now. Not only do these instruments provide instantaneous and precise locality placement, but also the software programmes that have accompanied them allow near precision with virtually any locality. In this Monograph where localities are interpreted by software programmes, the co-ordinates are preceded by ‘ca.’. These should be treated with caution. It was felt necessary to include such ‘approximate’ data to facilitate databasing. Where these coordinate refer to the reference point and not to the actual collecting locality, they follow the reference point.

More recently, digitisation programs have become available that should make the setting up of plates less time consuming. In future volumes electron micrographs will be digitised so that plates can be assembled and labeled without ever requiring a photograph to be made.

Common names

With the current interest in conservation biology and the use of tettigoniids in behavioural, ecological and acoustic studies, it is clear that common names are of utility and could now be deemed necessary. The author has, therefore, decided to provide many species with a common name at the point of its description to avoid the problems later when a common name is needed and left to others to decide on what it might be.

Habitats

In past volumes the introductory chapter contained photographs of relevant habitats. Since this assemblage of subfamilies is disparate, it was decided to discuss habitats under the respective taxa. Also, there was criticism that the habitat photographs were not in colour and did not convey the proper impression of the real appearance of the habitat. I hope that this has been reconciled here.

Collecting and Rearing

Techniques described for dealing with the species in this volume follow those described in both Volumes 1 and 2. Some difficulty was experienced in attempting to rear to maturity early instar nymphs of Terpandrus, various other listroscelidines and Tympanophora species. This is largely due to the predaceous habits of these species and the fact that an adequate diet has not been found that will provide the nutrition that will accomplish development to adulthood without malformed tegmina and wings legs etc. On the other hand, species thought to be equally predaceous like phisidines and meconematines and the excellent laboratory insect Requena, grow to maturity without any ill affects on the usual Orthoptera Food Mix (see Appendix III).

Preservation of Specimens

The longevity of liquid-preserved material is a continual cause for concern for museum curators, especially in these days of diminishing resources and the general reduction of time spent in curating collections. Will future generations be willing to attend to the matters of collection maintenance as has been so diligently done in the past? With this in mind, the author has commenced preserving more tettigoniids in the dry state than normally. This is one way of guaranteeing almost perpetual preservation. For many species, the best study specimens are those that have been first preserved in Pampels Fixative (see Appendix 4) then stored in 80% ethanol. These specimens reveal the critical taxonomic characters without distortion. They can be manipulated to achieve correct determination. Otherwise, dry-preserved specimens shrivel, decom-pose, and serve little taxonomic purpose. Specimens to be dry-preserved should first be gutted and the body cavity ‘dusted’ with the Gut Dust as outlined in Appendix 4. In tropical climes, the specimens should be dried at a low temperature, around 30 degrees C. In lieu of a source of heat, the drying boxes containing specimens should be placed in a thick plastic bag with plenty of colour-indicator silica gel. As the colour becomes pink, indicating moisture absorption, the crystals can be ‘cooked’ in a pot on the stove until they turn blue. They can then be reused. This process is continued until the crystals remain blue. The plastic bag should be sealed against further moisture absorption and stored until it is opened in the laboratory.

Morphological Characters Studied

The usual characters have been employed to distinguish species in this work. In addition, the author has embarked on a survey of several other morphological characters in order to test their utility in tettigoniid taxonomy. These characters are the mandibles and the proventricu-lus. They have been shown to be useful in gryllacridid classification and have potential in the Tettigoniidae.

Figure 1     Example of a typical tettigoniid proventriculus, Tympanophora uvarovi Zeuner. A, the anterior (cephalic) portion is to the bottom of the illustration. Five of the six longitudinal folds are shown. Note the reduced number of median teeth relative to the listroscelidines, for example. Could this be of taxonomic value? The less sclerotised, but densely hirsute portions near the base are also of taxonomic value but being weakly sclerotised, are subject to decomposition in poorly preserved material. B, several median teeth illustrating surface armature, original 430x. C, a few teeth at the anterior (bottom) of a longitudinal fold. See Figure 2 for terminology

Figure 2     A single longitudinal fold as illustrated by the phisidine, Phisis jinae, sp. nov. Major differences from the Gryllacrididae are in the absence of distinctly separate lateral lobes and the general lack of armature of the median tooth. [Compare with Fig. 1, Bland and Rentz, 1991.] SP, sclerotised partition; SL, sclerotised lobe; MT, median tooth; LL, lateral denticle; MD, median denticle.

Bland and Rentz (1991) examined a number of Australian gryllacridid genera to discover whether the proventriculus could be used as a taxonomic character. They examined the proventricular structure of seventeen species. Results suggested that those species with long lateral denticles tended to be predominantly plant feeders rather than predators (Bland 8c Rentz, 1991a). In contrast, specimens with a proventriculus with very short lateral denticles usually had moderate to large amounts of insect fragments. Co-incidentally, it was discovered that median and lateral denticles are not very well developed in the Tettigoniidae, at least in the species that have been examined in conjunction with this present study, for example (Figs. 1, 2,). This could be an important character distinguishing the two families.

With regard to mandibles, Rentz and John (1990) examined mandibles of the Gryllacrid-idae for taxonomic content. They concluded that although useful in taxonomy, the features were subtler than those found in the proventriculus and offered much less in the way of useful taxonomic characters. The author also decided it was important to check mandibles in tettigoniids. With these studies in mind, at least one proventriculus and mandible from each taxon in this monograph where a specimen could be so sacrificed was examined. The results suggest that there is potential taxonomic value in the use of these structures but care must be taken to use rather ‘young’ adult material. Both mandibles and proventriculi wear with age and one could err in viewing aged, misshapen and worn structures. This author would certainly recommend examination of these structures, especially when searching for characters at the higher levels. Preparation of the proventriculus for study is detailed in Bland and Rentz (1991). Mandibles need only be removed and cleaned for examination. Both were studied using the scanning electron microscope and the micrographs are presented here for critical evaluation. Terminology for the structures observed (Fig. 2) in the proventriculus follows that of Bland and Rentz (1991) and for that of the mandible (Fig. 3) from Rentz and John (1990).

Figure 3     Two very different kinds of tettigoniid mandible. A, B, the phaneropterine, Caedicia simplex(Walker), a mandible typical of a species that feeds on plant material. It is robust and with three almost equal dentes (D), the Incisor Dens is not much differentiated from the others, originals 30x. C, Terpandrus burragah, sp. nov., a mandible typical of a predator, original 40x. It is elongate with the Incisor Dens (I) hooked for prey capture. One of the Dentes is large and wedge-shaped. The Molar region (M) is distinctive in its cup shape, especially when viewed ventrally (B).

Other Taxonomic Characters

In the majority of examples, the left tegmina was removed and cleaned prior to gold coating and examination under the electron microscope. The tegmina of many genera, such as Terpandrus, Yullandria gen. nov. and Yutjuwalia gen. nov. are too large to be placed in the confined space of an electron microscope. In order examine these, the file would have to be cut from the tegmen. To avoid this, replicas of the stridulatory file were made after it was cleaned using a mild detergent and/or chloroform. Stubborn dirt particles and other substances can be coaxed to be removed by gentle brushing of the file with a camel hair brush. Replicas can be suitably and quickly made using Imprint™ vinyl polysiloxane impression material. This sub-stance is used by dentists in making molds or imprints of teeth. The replicas were trimmed to size, point-mounted, housed alongside the appropriate specimen, and labelled with the pho-tonegative number for future reference.

Materials and Methods

Measurements

Measurements of specimens were made using an ocular micrometer except when provided from electron micrographs. Because of the various potential sources of error in the enlargement of photographs and the juxtaposition of specimens, scale lines have been largely eliminated. A magnification is quoted which was the magnification at the time the specimen was photographed. Measurements that are more precise can be found in the Tables of Measurements that are provided for the taxa. In various tables, localities listed are often abbreviated. Full locality data, of course, can be found in the ‘Specimens Examined’ sections.

Song

Analysis of calling songs was accomplished using two techniques. Some of the material that was analysed early in the study followed the procedure described in Volume 2, p. 5. More recently, the Canary™ software program has been utilised to analyse and produce graphic representations of Calling Songs. In future, this will make analysis simpler and readily comparable. For clarification the following definitions are followed:

Calling Song—a sequence of chirps, usually uttered repeatedly by a single male, rarely in synchrony or in response to others;

Chirp—the smallest singing unit that can be resolved by the human ear;

Syllable—each chirp consists of a number of syllables;

Pulses—one simple wave train—each syllable consists of a number of pulses;

Maps

Maps have been listed separately to aid in locating them in the book.

Specimens Examined

Although this study is based largely on specimens in the Australian National Insect Collection, CSIRO Entomology, Canberra, I have examined all relevant material in all of the major collections of Australia and overseas. All types have been examined. Museums and collections from which specimens are cited in the ‘Specimens Examined’ sections are abbreviated according to the following code. Material of some species collected by the author will be deposited in other collections in Australia as well as in those overseas. Specimens cited in this monograph can be traced by the following abbreviations:

AMUS—Australian Museum, Sydney

ANIC—Australian National Insect Collection, Canberra

ANSP—Academy of Natural Sciences of Philadelphia, Philadelphia

BPBM—Berniece P. Bishop Museum, Honolulu, Hawaii.

BMNH—British Museum (Natural History), The Natural History Museum, London

MUSV—Museum of Victoria, Melbourne, Vic.

NTMU—Northern Territory Museum, Darwin, N. T.

SAMA—South Australian Museum, Adelaide, S. A.

QMUS—Queensland Museum, South Brisbane.

UQEN—Queensland University, Insect Collection, St. Lucia, Q.

WAMP—Western Australian Museum, Perth, W. A.

CHECKLIST OF SPECIES INCLUDED IN THIS VOLUME

LlSTROSCELIDINAE

Terpandrini

Terpandrus Stål

Horridus Group

Terpandrus horridus (Burmeister)

Terpandrus burragah Rentz

Terpandrus eucla Rentz

Tauwa Group

Terpandrus tauwa Rentz

Terpandrus bundawoodgera Rentz

Terpandrus woodgeri Rentz

Terpandrus illamurta Rentz

Terpandrus weema Rentz

Endota Group

Terpandrus endota Rentz

Terpandrus calperum Rentz

Terpandrus borral Rentz

Terpandrus paruna Rentz

Terpandrus splendidus Hebard

Jumbunna Group

Terpandrus jumbunna Rentz

Terpandrus jimiramira Rentz

Terpandrus norabeetya Rentz

Terpandrus moonga Rentz

Terpandrus cabon Rentz

Itye Group

Terpandrus itye Rentz Yullandria

Rentz Yullandria lawagimana Rentz

Yullandria kakadu Rentz

Chlorobalius Tepper Chlorobalius

leucoviridis Tepper

Yutjuwalia Rentz Yutjuwalia

nyalma Rentz Yutjuwalia

sallyae Rentz

Burnuia Rentz Burnuia

mirru Rentz

Hexacentrini

Hexacentrus

Serville Hexacentrus mundurra Rentz

Alison Rentz

Alison roachae Rentz

Alison sedlacekorum

Alison thamyris Rentz*

Alison garaina Rentz*

Alison nishidai Rentz*

Requenini

Requena Walker

Requena verticalis Walker

Requena rotto Rentz

Requena baraya Rentz

Requena pipa Rentz

Requena victoriae Rentz

Requena kangaroo Rentz

Requena kerla Rentz

Requena kadyakia Rentz

Requena kimi Rentz

Requena minya Rentz

Requena kurarda Rentz

Requena dajurta Rentz

Requena brolga Rentz

Requena helleri Rentz

Requena winstoni Rentz

Thumelinia Rentz

Thumelinia waminda Rentz

Thumeliniawinstoni Rentz

Xingbaoia Rentz

Xingbaoia karakara Rentz

Conocephalomlmini

Conocephalomima Rentz

Conocephalomima barameda Rentz

Phisidini

Paraphisis Karny

Subgenus HeterophisisJin

Paraphisis (Heterophisis) turnar Rentz

Paraphisis (Heterophisis) wonnewarra Rentz

Paraphisis (Heterophisis) kurnkuni Rentz

Paraphisis (Heterophisis) nurragi Rentz

Paraphisis (Heterophisis) lynae Rentz

Paraphisis (Heterophisis) modla Rentz

Paraphisis (Heterophisis) wirreecoo Rentz

[Paraphisis (Heterophisis) listen (Kirby)] Subgenus Tapangiphisis Rentz

Paraphisis (Tapangiphisis) chopardi Jin, Kevan &Hsu

Paraphisis (Tapangiphisis) alumba Rentz

Paraphisis (Tapangiphisis) leawillia Rentz

Paraphisis (Tapangiphisis) Jin

Beiericolya Kaltenbach

Beiericolya howensis Rentz

Beiericolya tardipes (Rentz)

Meiophisis Jin

Meiophisis likkaldin Rentz

Neophisis Jin

Subgenus Neophisis Jin

Neophisis (N.) ecmurra Rentz

Phisis Stål

Phisis jinae Rentz

Estrinia Karny Estrinia

dauanensis Rentz

TYMPANOPHORINAE

Tympanophora White

Andreae Group

Tympanophora andreae Rentz

Tympanophora insolita Riek

Tympanophora houstoni Rentz

Tympanophora rotto Rentz

Tympanophora uvarovi Zeuner

Pellucida Group

Tympanophora pellucida (White)

Tympanophora similis Riek

Tympanophora pinnaroo Rentz

Tympanophora kalbarri Rentz

Tympanophora splendida Riek

Diminuta Group

Tympanophora diminuta Rentz

Tympanophora picta Rentz

Tympanophora ourapilla Rentz

Aka Group

Tympanophora aka Rentz

MECONEMATINAE

Phlugidini

Austrophlugis Rentz

Austrophlugis debaari Rentz

Austrophlugis manya Rentz

Austrophlugis malidupa Rentz

Austrophlugis quaringa Rentz

Austrophlugis kununurra Rentz

Austrophlugis kumbumbana Rentz

Austrophlugis orumbera Rentz

Austrophlugis sp. 1

Tenuiphlugis Kevan

Tenuiphlugis brittoni Jin

Tenuiphlugis pitti Rentz

Tenuiphlugis tiwiwarrina Rentz

Indiamba Rentz

Indiamba quamara Rentz

Indiamba wirrawilla Rentz

Indiamba malkini (Jin)

MICROTETTIGONIINAE

Microtettigonia Rentz

Microtettigonia kangaroo Rentz

Microtettigonia tachys Rentz

Microtettigonia illcha Rentz

Microtettigonia whippoo Rentz

Microtettigonia alleni Rentz

Microtettigonia tunte Rentz

Microtettigonia kutyeri Rentz

Microtettigonia sp. 1

Microtettigonia sp. 2

Microtettigonia sp. 3

*Non-Allstralian species.

SYNOPSIS OF SUBFAMILIES INCLUDED IN THIS VOLUME

Listroscelidinae Kirby 1906

The Australian component of the Listroscelidinae as here constituted consists of genera which share a series of incurved spines on the ventral surface of the fore tibiae. These probably aid, along with similar spines on the middle tibiae, in capturing prey. On warm summer or autumn nights in areas where certain listroscelidine genera are common, one can observe with a bit of luck the ‘clap-trap’ juxtaposition of legs and spines that catch unwary insects. The katydids hang from shrubby vegetation by the hind legs, with at least one middle leg, and sometimes both, depending upon the situation, agape, forming a net-like trap should a flying insect happen by. This prey-capture technique has been observed in Saginae and Tettigoniinae and in such disparate-looking Australian listroscelidines as Terpandrus Stål and Paraphisis Karny species. However, such behaviour has not been observed in all genera but seems likely to occur. When the ‘clap-trap is not in use and the katydids capture prey by pouncing and subduing, the spines seem to have little or no function’, Rentz (1995).

The Australian Listroscelidinae range in size and shape from the large, robust Gumleaf Katydids of Terpandrus which occur throughout Australia to delicate phisidine katydids which occur on vegetation primarily in the rainforests of far north Queensland. Fully winged and micropterous species are known and included in the subfamily. It is interesting to note that no listroscelidine has been recorded from Tasmania. The classification adopted here is tentative and is mostly likely to change when the aforementioned classification appears. For the present I am following the ‘indications’ derived from the numerical treatment by Don Colless which follows, p. 14. Certain elements of his results are very explainable, others less so. For example, he suggests many plausible groupings, the most notable of which is the clustering of the phisidine genera and the terpandrines. This is further confirmed in both the Furthest Neighbour and Nearest Neighbour analyses. As to be discussed later, Conocephalomima, gen. nov. seems misplaced in the subfamily. The characterisation and components of the Listroscelidinae tribes follow.

The Terpandrini, trib. nov. are generally large katydids with those of the nominate genus, Terpandrus Stål, the most widespread of listroscelidine genera in the continent with species occurring in the far northern tropics on both coasts and in the south along the southern coast. They are diurnal or nocturnal depending upon the species and the song is very loud. They are often heard and recognised by an entomologically aware public but most have never seen one because they are usually at the tops of eucalypts or phyllode-type acacias. These katydids generally resemble eucalypt leaves and have been called Gum-leaf Katydids. The male calling song is loud and continuous and could be mistaken for that of a cicada. Other genera placed in this tribe are Burnuia, Yutjuwalia, Yullandria, genera nova, along with Chlorobalius Tepper which is widely known. This latter genus has had a chequered taxonomic history first being described as a decticine c 1, p. 1). It is monotypic and its widespread species is commonly encountered in arid parts of central Australia where there is natural vegetation. The other terpandrine genera are more restricted in their distributions and have peculiar life styles that are discussed more fully in appropriate sections of this monograph. It is worth noting here that Yutjuwalia sallyae, gen., sp. nov. has the lowest known chromosome number of any tettigoniid. This is further discussed on page 506. The North American listroscelidine,Neobarrettia Rehn (see Cohn, 1965) and the very aberrant South American genus Megatympa-non Toledo-Piza (see Riek, 1976) are also belong to this tribe.

The Hexacentrini Gorochov is represented in Australia by two genera. The nominate genus Hexacentrus Serville is a pan-tropical genus with 25 species as listed by Naskrecki & Otte(1997). Australia is known to harbour a single species that is widespread in the tropics. It is loud and common and, to considerable wonderment, has not been described until now. The second generic representative of the tribe is Alison gen. nov. with a single known species in northern, tropical Queensland and several from New Guinea. The two genera share a similar appearance and have the stridulatory file similarly peculiarly modified. The pairing of these genera is not reflected by the Colless analysis for reasons not yet understood.

The Requenini, trib. nov. comprises three genera. Requena Walker which, like Chlorobalius, was originally described as a decticine but is more appropriately considered a listroscelidine. The genus comprises at least 15 species occurring in heath habitats on the east and west coasts of the southern portion of the continent. Because the species are common and one, R. verticalis Walker, occurs in Perth, W. A., they have been used in a variety of behavioural and acoustic studies by colleagues at the University of Western Australia. Full details of these investigations are discussed in the appropriate section of this monograph. Also described from Western Australia is Thumelinia gen. nov. which lives on monocotyledonous plants, especially Lomandra species, where it and Requena species can be frequently found together. Xingbaoia gen. nov. is an aberrant-looking long-legged tettigoniid that lives in low shrubs in a restricted portion of the rainforests of tropical north Queensland.

The Conocephalomimi, trib. nov. is proposed to include a single known species, Conocephalomima baramedagen., sp. nov. As can be seen in the analyses (p. 15), Conocephalomima, gen. nov. may even be misplaced in the Listroscelidinae. It is included here provisionally and will be included in the generic analysis of P. Naskrecki noted earlier.

The extent of the Australian representation of the Phisidini Jin is one of the most surprising discoveries of this project. The 1992 monograph of the group by Jin & Kevan recorded only one genus from mainland Australia but now we know that six of the 14 genera are represented here, some 43% of the world fauna of this widespread tribe. The group is tropical or subtropical in distribution. Paraphisis Karny is the dominant genus with 12 species in two sub-genera represented on both sides of the continent. They are most often associated with rain-forest vegetation but species have been found on eucalypts in grassland habitats as well. Beiericolya Kaltenbach is known from two species, one each occurring on Norfolk and Lord Howe Islands respectively. Elsewhere the genus is known from New Caledonia. Meiophisis Jin, Neophisis Jin, Phisis Stål, and Estrinia Karny each contribute a single species to the Australian eastern rainforest fauna but all are better represented on islands in the Indo-Pacific region (Jin & Kevan, 1992).

Tympanophorinae

The Tympanophorinae is an endemic subfamily comprising a single genus, Tympanophora White. This is a most peculiar group in many respects. Males are fully winged and capable of flight, females are wingless. Both sexes have relatively short legs that appear to be modified for grasping twigs and branches. They are predaceous. Males have a peculiar stridulatory file that consists of two sections. In addition, it appears that they have a ‘reservoir’ beneath the file with spongy openings at two points along the file. This has been examined in cross-section and is discussed further in the section on the subfamily. There are other morphological oddities in this group which are discussed in detail later. Among these are the extraordinary development of the tracheal system in males and the greatly reduced size of the testes. Six species are added to the seven tympanophorine species already known. All but two are from the sand-heath areas of Western Australia. Several are of conservation significance because they have extremely limited distributions and their habitats are under threat due to residential development in the Perth suburbs. A single species occurs in heath habitats in coastal southern and central New South Wales. The most peculiar species of the genus is a very small example from heath habitats on the Eyre Peninsula of South Australia. This species is unique in that males are micropterous and cannot fly.

Meconematinae

The Meconematinae is considered as a separate family by some authors and a tribe or offshoot of the Listroscelidinae by others. Remarkable convergence is illustrated by both the appearance and behaviour of the phisidine listroscelidine genera and the phlugidine meconema-tines (compare Fig 153 and CP 37 with CP 44-50). The taxonomic literature has been confused due to the lack of appreciation of convergence as illustrated with these two groups. The Australian Meconematinae are represented by a single tribe, the Phlugidini, which comprises three genera, two of which are endemic and new. Austrophlugis gen. nov. includes seven new species from Queensland, two that are common in Brisbane gardens where their continuous diurnal calling songs are easily revealed on the Mini Bat Detector. Other species occur north along the coast and on several adjacent islands. The arboreal habits of these species preclude their being encountered by general collectors. Tenuiphlugis Kevan and Indiamba, gen. nov.are micropterous katydids both with unusual development of the male genitalia. Each of these last two genera is known from three species occurring in the northern areas of Australia on both sides of the continent.

Microtettigoniinae

The Microtettigoniinae is an endemic subfamily comprising a single genus, Microtettigonia Rentz. Previously known from two described species, one from Western Australia, the other from south coastal South Australia, the genus now contains five additional species, all from Western Australia.

Subfamily Listroscelidinae

Predatory Katydids

Key to Australian Listroscelidinae

1.         Body form very slender (Fig. 153), size small, not large robust katydids. Ovipositor falcate (Figs. 158, 159).............................................................................................................. 2

            Body form very robust (Figs. 44, 66, 73), large to very large, robust katydids;ovipositor ensiform................................................................................................................................ 4

2(1).    Fore and middle legs very slender, elongate recurved spines (Fig. 152A, B); tibial auditory structure often swollen (Fig. 170F)........................................ Phisidini, p. 290

            Fore and middle legs more normal in appearance, the recurved spines not as slender or elongate as above...................................................................................................................... 3

3(2).    Having a ‘ Conocephalus appearance’ (CP 35 ). Overall colour greenish; both sexesfully winged. Male cercus slender, strongly incurved (Fig. 149B, C); female with minute hooks at base of subgenital plate (Fig. 149D, E)

            ...............................Conocephalomima, gen. nov., p. 280

            Not appearing like Conocephalus. Overall colour brown (Fig. 139A, B); both sexes micropterous. Male cercus short, not as   above (Fig. 146A, B); female without hooks at base of subgenital....................................................... plate Xingbaoia, gen. nov., p.

4(1).    Sexually dimorphic. Males fully-winged, capable of short flights (CP 29 ); females mesopterous to fully winged but tegmina differently shaped than those of male........................................... 5

            Not sexually dimorphic; wings micropterous and fully hidden beneath pronotum or normal with both sexes capable of flight............................................................................................................. 7

5(4).    Base of fore and middle tibiae with minute black markings (Figs. 95F, G; 99D, E; 100D, E; 101C, D). Dorsal surface of middle tibia unarmed................... Alison, gen. nov., p. 177

            aBase of fore and middle tibiae without any markings. Dorsal surface of middle tibia armed with spines............................................................................................................................................. 6

6(5).   Dorsal surface of pronotum with a dark hourglass figure (Fig. 85A). Species tropical in distribution (Map 12).............................. Hexacentrus Burmeister, p. 164

            Dorsal surface of pronotum lacking hourglass figure (Fig. 77), Species confined to Western Australia (Map 11).................................. Burnuia, gen. nov., p. 153

7(4).   Large fully winged species capable of flight (CP 25 )....................................................................................................................................... 8

            Small micropterous species (Figs. 118; CP 31, 33), the male tegmina generally hidden underneath the pronotum, not capable of flight............................ 11

8(7).   Tegmina and body with many small, white spots (Figs. 65, 66). Tegmina slender....................................Chlorobalius Tepper, p. 131

            Tegmina and body lacking many small spots, although it may be striped or bear opalescent markings on the thorax. Tegmina usually quite broad, especially basally............................................................................ 9

9(8).   Ventral surface of fore and middle femora and tibiae armed with short spines or teeth........................................................................... 10

            Ventral surface of fore and middle femora and tibiae unarmed. [Very long-legged appearance]........................................ Yutjuwalia, gen. nov., p. 145

10(9).  Thoracic armature reduced. Pro- and mesozona forwardly produced. Males lacking a sclerotised epiphallus. [Body and tegmina usually glistening]............................................Yullandria, gen. nov., p. 118

            Thoracic armature well developed. Pro- and mesozona not projecting forward.

            [Body and tegmina not generally glistening]........................................ Terpandrus Stal, p. 21

11(7).  Overall colour brown or grey, but rarely with turquoise markings on sides of abdomen (Figs. 116, 118). Male pronotum flaring caudally, the tegmina often not wholly concealed. Female with ovipositor length and curvature varied as to species and rarely armed...................................................... Requena Walker, p. 197

            Overall colour green. Male pronotum nearly cylindrical (Fig. 136A), not flaring caudally and wholly concealing the tegmina. Female ovipositor long, straight, unarmed............................................................. Thumelinia, gen. nov., p. 259

CLASSIFICATION AND NUMERICAL ANALYSIS OF AUSTRALIAN LISTROSCELIDINE KATYDIDS

By D. H. Colless

The data were presented as a matrix of 22 OTU’s (genera) scored for 67 characters. Of those, 14 were clearly unordered; the remainder were treated as ordered, but not directed (e.g., orders 0>1>2, 2>1>0, and 0<1>2 were equivalent. The data appeared ‘robust’, in that 60% of characters had no more then 9 OTU’s sharing the same state. Dr. Rentz had some residual qualms about possible subjective scoring of some characters. This was tested by running preliminary phenetic analyses, first with all characters, then with 26 doubtful ones deleted, and then with the doubtful ones alone. The results were so similar (even in the last case) that it was decided to proceed using all 67.

Figure 4       Principal Coordinate Analysis of distance matrix: plot of axis 1 against axis 2.

Figure 5       Minimum Spanning Tree from distance matrix.

The analyses that follow used a battery of programmes of my own devising, with the obvious exception of the cladogram computed using PAUP 2.4. Genera or subgenera are referred to throughout by the first 6 letters of their names.

Phenetics

A distance matrix was first computed using Manhattan Metric on range-coded data, and a Principal Coordinate Analysis of these distances, squared for clearer separation of groups, yielded Fig. 4 as a plot on the first 2 coordinates (the results using raw distances were very similar). This picture may be significantly distorted, since only 53% of variability was explained and there were several negative roots. Therefore, to further clarify relationships, a Nearest-Neighbour Table (showing 6 nearest neighbours to each OUT); (not shown) and a Minimum-Spanning Tree (MST; Fig. 5) were also computed.

For phenograms, 4 standard sorting strategies were applied to the distance matrix: Nearest-Neighbour (NNB), Furthest-Neighbour (FNB), UPGMA, and WPGMA. Using a criterion mentioned elsewhere (e.g., Colless 1994: 521), coefficients of variation in internode length were computed for each phenogram. These ranged from 0.66 to 0.86, and that with the lowest coefficient (FNB) was chosen as the ‘best’. This is shown as Fig. 6. With the exception of NNB, which was very poorly structured, differences between the other 3 phenograms were not great (a strict consensus showed 80% of subtrees held in common). The FNB phenogram was then ‘cleaned’ of minor (and therefore more dubious) internodes by deleting all those less than 0.75 of mean internode length. The result is shown in Fig. 7.

Figure 6       Furthest Neighbour phenogram.

In all analyses the outgoups Clonia and Hemisa are sharply separated from the study group, providing useful support for later cladistic studies. Likewise, within the study group there are 3 clearcut groupings, each with some possibly significant internal structure; as follows:

Group 1. (Terpan, Bumui, Yutjuw, (Chloro, Yullan, Neobar)) Group 2. ((Thumel, Xingba, Requen), Alison, Megaty, Hexace). Group 3. ((Neophi, Meioph), (Phisis, Estrin), (Hetero, Tapang) Beieri).

Conoce is not clearly placed, but could be considered an outlier of group 3. Also, the place-ment of Alison, Megate, and Hexace is rather weak, as shown by the MST (Fig. 5), where they intrude into group 1. Higher level relationships of these groups remain unclear.

For classificatory purposes, the above 3 groups could be recognised, if required, with Conoce left unplaced or attached to group 3.

Figure 7       As for Fig. 3 but with shorter internodes deleted (see text).

Cladistics

Fig. 8 shows the cladogram generated by the Neighbour-joining strategy of Saitou and Nei (1987), applied to the distance matrix described above, while the standard PAUP 2.4 pro-gramme yielded the tree shown in Fig. 9. The Consistency Index of the latter was 0.38, indicat-ing a considerable level of homoplasy in the data. The two trees are very similar, sharing 80% of their subtrees, as can be seen in the Strict Consensus Tree (Fig. 10). The second tree is rather less symmetrical than the first, with Isym tha (Colless 1995) of 0.38 as compared with 0.24; and, as usually found, both are less symmetrical than the phenograms (Isym for FNB was 0.19).

For classificatory purposes, these trees yield much the same results as discussed for the phenetic analyses. Groups 1 and 3 are strongly supported, as is the (Thumel, Xingba, Requen) subgroup of group 2; but the (Megaty, Hexace, Alison) subgroup is rejected, with its members as paraphyletic on group 2 (the first 2 could perhaps be attached to group 1). The cladograms do show a fairly clear association of groups 1 and 2, but Conoce remains once more unplaced. This relates perhaps to the very long branch that connects it to the tree: 19 steps, as compared with a mean length of 8 steps for other branches and a maximum of 17 (for Yutjuw).

As indicators of evolutionary history, I would regard the fine detail of the cladograms as highly suspect: some 60% of steps on the PAUP tree must be homoplasious (convergent) and

Figure 8       Neighbour-Joining cladogram.

most characters show a convoluted history of convergence and reversal. Also, the considerable asymmetry of that tree, while not damning, is nevertheless suspicious for reasons discussed in Colless (1996). However, there are a few ‘clean’ apomorphies (i.e., not involving homoplasy elsewhere on the tree): character 46, state 1, supports group 1 + Megaty and state 2 supports group 1 alone; character 35, state 1+, supports group 1; characters 17 and 21, state 2 in both, support the Requen, Xingbao, Thumel subgroup; character 27, state 3, supports group 3; character 62, state 0, supports Yullan, Chloro, Neobar; and character 56, state 2, supports Hetero, Tapang, Beieri. All in all, the considerable congruence between both phenetics and cladistics suggests strongly that the broad details, of the groupings described, credibly reflect the phylogeny of the study group.

References

Colless, D.H., 1994. A new family of muscoid diptera from Australasia, with sixteen new species in four new genera (Diptera: Axiniidae). Invertebrate Taxonomy 8:471 -534.

Figure 9       PAUP cladogram (see text).

Colless, D.H., 1995. Relative symmetry of cladograms and phenograms: an experimental

study. Systematic Biology 44:102-108. Colless, D.H., 1996. A further note on symmetry of taxonomic trees. Systematic Biology

45:385-390.

Figure 10      Strict Consensus Tree Of Fig 5 And 6.

CP 1, Terpandrus horridus, Tea Gardens, N. S. W., female. CP 2, T. , sp. nov., holotype male.

CP 3, 7 endota, male, Queanbeyan, N. S. W. CP 4, T. splendidus, female, 85 km E. Menzies, W. A.

CP 5, same species, male, Parachilna, S. A. CP 6, T. jumbunna, sp. nov., holotype male.

CP 7, Terpandrus jimiramira, sp. nov., holotype male. CP 8, T. moonga, sp. nov., holotype male.

CP 9, T. tauwa, sp. nov., male, Davies Creek, Qld. CP 10, T. burragah, sp. nov., female, Coonabarabran, N. S. W. CP 11, T. eucla, sp. nov., nymph, note tuberculate pronotum, nr Eucla, W. A. CP 12, T. woodgeri, sp. nov., male, Malcolm, W. A. CP 13, Terpandrus woodgeri, sp. nov, paratopotype male. CP 14, T. bundawoodgera, sp. nov., male, nr Augathella, Qld. CP 15, T. illamurta, male, Upper Running Waters, N. T. CP 16, T. endota, sp. nov., female, Cooloola Nat. Pk, Qld.

CP 17, T. borral, sp. nov., male, Eyre Bird Observatory, W. A. CP 18, same species, defensive posture, note markings on underside of fore femur. CP 19, Terpandrus calperum, sp. nov., male paratopotype. CP 20, T. splendidus, nymph, Erldunda, N. T. CP 21, T. splendidus, female, Flinders Ranges, S. A. CP 22, T. jumbunna, sp. nov., male, paratopotype. CP 23, same species, female with fresh spermatophore. CP 24, T. norabeetya, sp. nov., female, nr W. Wallsend, N. S. W.

CP 25, Chlorobalius leucoviridis, male, Kalgoorlie, W. A. CP 26, same species, nymph in ambush, nr Alice Springs, N. T. CP27, Hexacentrus mundurra, sp. nov., male stridulating, Mt Webb, Qld. CP28, same species, long-winged female, Capt Billy’s Landing, Qld. CP 29, Alison roachae, gen. sp. nov., holotype, note shape of tegmina and colour pattern. CP 30, same species, note colour pattern and leg markings.

CP 31, Requena rotto, sp. nov., male, paratopotype. CP 32, R. victoriae, sp. nov., male paratopotype. CP 33, Thumelinia waminda, gen. sp. nov., male, paratopotype, note colour and pronotum. CP 34, same species, paratopotype, female. CP 35, Conocephalomima barameda, gen. sp. nov. female, Cooloola Nat. Pk., Qld. CP 36, Paraphisis (T.) alumba, sp. nov., Atherton, Qld.

CP 37, Phisis jinae, sp. nov., male, Green I., Qld. CP 38, Tympanophora andreae, sp. nov., male, paratopotype. CP 39, same species, female. CP 40, T. uvarovi, male, nr Tabourie Lakes, N. S. W. CP 41, T. similis, male, Kings Park, Perth, W. A. CP 42, T. diminuta, male, Cape Le Grand Nat. Pk, W. A.

CP 43, Tympanophora aka, holotype male. CP 44, Austrophlugis debaari, gen. sp. nov., paratopo-type female, note eye colour. CP 45, Austrophlugis malidupa, gen. sp. nov., paratopotype male, note eye colour. CP 46, Tenuiuphlugis pitti, sp. nov., female, Iron Range, QlcL, note eye colour. CP 47, 7. pitti, male, Heathlands, Qld. CP 48, T. tiwiwarrina, sp. nov., amale, Mitchell Plateau, W. A.

CP 49, Indiamba malkini, comb, nov., male, Berrimah, N. T. CP 50, /. quamara, gen. sp. nov., male paratopotype. CP 51, /. wirrawilla, sp. nov., female, Kapalga, N. T. CP 52, Yullandria lawagiamana, gen., sp. nov., female, Kings Canyon, N. T. CP 53, Bumuia mirru, gen., sp. nov., male, Balladonia, W. A. CP 54, Meiophisis likkaldin, sp. nov., adult male, note abbreviate tegmina, Atherton Tableland, Qld.

TRIBE TERPANDRINI, RENTZ, TRIB. NOV.

Known from five Australian genera. Characterised by similar appearance of both sexes, fully winged, capable of flight except in Burnuia, gen. nov. where gliding may be the principal form of flight. Stridulatory file with normal appearance, generally straight to slightly arching, the teeth lamellar, the broadest teeth near the middle (Figs. 30-32,) in Terpandrus but with a peculiar spacing of proximal teeth in Chlorobalius (Figs. 68, 69). Legs with spines of fore and middle tibiae long and incurved, femora also often heavily spined. Male cercus highly species distinctive (for example, compare Figs. 12-14 and 60, 67A, 75). Phallic complex with sclerotised portions (Terpandrus, Chlorobalius (feebly), Yutjuwalia, Burnuia), or with no sclerotised portions, (Yullandria). Ovipositor very elongate, straight to feebly curved upwards.

Gumleaf Katydids

Genus Terpandrus Stål

Terpandrus Stål, 1874, p. 117. Type species: Hexacentrus horridus Burmeister, 1838, by monotypy.

Terpandrus is one of Australia’s most distinctive katydid genera. One of its described species is among the oldest of Australian tettigoniid taxons. Since it was described, the name T. horridus (Burmeister) has been wrongly attributed to the species correctly known as T. splendidus Hebard. Terpandrus is among the most widely distributed of Australian tettigoniid genera with species occurring in most parts of the country except for Tasmania. The 19 species range in size from relatively small to the largest known species of Australian Tettigoniidae. They are characterised by their strong resemblance to ‘gum leaves’ (Eucalyptus spp.) even though not all individuals in any one species are to be found on eucalypts. Gum-leaf Katydids frequently inhabit plants other than eucalypts with Acacia spp. being the most common. The resemblance of Terpandrus species to eucalypt foliage is so perfect that the legs of several species are reddish brown, simulating the petioles of the leaves and stems of the trees they inhabit. McCoy (1886) was the first to christen one of these species with a common name, The Great Green Gum-tree Grasshopper’ which he identified as Terpandrus vigentissima Serville. However, the species he illustrated and discussed was most likely T. endota sp. nov., p. 66.

Terpandrus is a Gondwanan genus. It is related, though distantly (see p. ), to Megatympanon Toledo-Piza from South America which was discussed in detail by Riek (1976) and later by Rentz (1979). The similarities and differences between the two taxa are well worth noting here. In general appearance, Megatympanon is very similar to a number of Terpandrus species (Fig. 11). The colour of the body, tegmina and wings, head and facial features are very similar between the two genera. The fore and middle femora are armed in Megatympanon as they are in Terpandrus but there are 9-10 spines on both sides of ventral surface of both tibiae of Mega-tympanon whereas in Terpandrus there are 6-7. The tarsi are also similar in the two taxa. Where Terpandrus species bear a pair of apical spurs on the dorsal surface of the hind tibia, Megatympanon bears only a single spur on the internal margin. The ventral surface bears two pairs of spurs where in Terpandrus there is but a single pair. The most important difference is that in Megatympanon the tegmina are identical. That is the stridulatory areas are identically developed on both the right and left tegmina. The file, mirror and scraper are identical on each side, whereas, in Terpandrus, and all other tettigoniids with which I am familiar, these structures are not bilaterally symmetrical. This suggests that the male Megatympanon can stridulate

Figure 11   Megatympanon speculatum Toledo-Piza, male, a distant relative of Australian listroscelidine species from South America. Note fully functional stridulatory veins at the base of each tegmen, a characteristic unique in the Tettigoniidae.

with either tegmen. Riek (1976) suggested that this condition was primitive and, except for Megatympanon, was found only in Haglidae, for example, Cyphoderris Uhler, and not known elsewhere in extant Tettigoniidae.

Terpandrus species are predators. They are often found in flowering trees where insects are abundant. Some species are nocturnal, others diurnal. Frequently at a given locality more than one species may be found, and, generally, one will be nocturnal, the other diurnal. Most species mature in summer with adults found well into autumn. In tropical climes, the regime is roughly parallel with maturation occurring in the wet season and adults commonly singing beyond the middle of the dry season. In many tropical localities adults can be found long after they have succumbed in more southerly climes.

Females oviposit in the ground near the end of their lives and can often be found in late summer on roads after dark. In temperate areas eggs hatch from late winter to mid spring. The nymphs can be easily found amongst rank annual and perennial herbs where there is an abundance of small insects upon which they feed. As they mature, and become larger, they are the targets of other predatory organisms. The spring grasses and herbs die and their prey dwindle. In the last instar stages, they can still be found in shrubbery, at night and moulting individuals are not uncommonly encountered. They then tend to ascend taller vegetation as adults residing at the tops of high trees and shrubs especially if there is an abundance of food.

Singing individuals can be heard calling from the same position in a given tree or shrub for many nights in succession. T. splendidus Hebard, an unusual species in many respects (CP 4, 5), is different in that it is highly mobile and may move to several different sites in a single evening. In this respect, it is similar to the wholly sympatric and related terpandriine Chlorobahius leucoviridis Tepper (p. 124).

Rearing Terpandrus species in the laboratory is a very difficult task, especially if attempting to raise nymphs that are very young. Perhaps, there is some element of food that is not provided in the laboratory even if live food (bush flies) is provided in addition to the artificial diet described in Volume 1 of this series. Under such conditions the katydids become distorted with each succeeding moult and mature with shrivelled tegmina and wings and bent limbs. Individuals collected in the later instars have a greater chance of maturing properly than do young nymphs in the laboratory.

Because the genus is so widespread (Maps 1-6), and many species appear to be localised, it is likely that only a portion of the total number of species are represented here. Indeed, several taxa are represented by only a few specimens. One could spend a lifetime listening, recording and documenting the life histories of this interesting and characteristic genus.

The vagility of many of the species beckons further intensive study to determine the degree, if any, of hybridisation. This may explain some of the anomalies encountered in this taxonomic investigation which is, by necessity, based on dead museum specimens. Even with the availability of a limited number of tape-recorded songs, cytology, and modern observational equipment such as the electron microscope, it seems that we only have part of the story here. Like most other genera in a study of this nature, further directed fieldwork and laboratory studies could prove most rewarding.

Terpandrus, as presently understood, comprises 5 species groups. These groups are based on the shape of the male cercus and secondarily on the morphology of the phallic complex and colour. Females are less easy to characterise because their features are more subtle and do not exhibit obvious differences. An additional problem with colour as a character is that many museum specimens are faded and the colours are not true resulting in incomplete data for many species.

The five species groups are defined as follows:

The Horridus Group

Male cercus very elongate, cylindrical or subcylindrical with a small internal tooth (Fig. 12). Tegmina coriaceous. Male phallus with epiphallus represented as a single, unpaired lobe either largely unsclerotised or sclerotised and shagreened. Three species known.

The Tauwa Group

Surface of pronotum with a black transverse stripe, remainder reddish brown; lateral lobe bearing an oblique pearly stripe. Ventral surface of fore and middle femora without dark marks. Male cercus strongly dorsoventrally flattened, bearing one or two internal teeth (Fig. 13). Male epiphallus little more than a pair of weakly sclerotised, divided lobes. Five species are known.

The Endota Group

Male cercus elongate, dorsoventrally flattened, serrated along a considerable length of internal margin (Fig. 14). Caudal margin of lateral pronotal lobe white or creamish yellow. Male phallus with epiphallus with a well sclerotised central, transverse bridge with a rod-like median shaft apically sagittate. Female subgenital plate with a strong or with a faint median carina. Five species known.

Figure 12    Left male cerci, dorsal, internal and ventral views of some members of the Horridus and Itye Groups of Terpandrus. A-C, T. horridus, Sutherland (Sydney), N. S. W. D-F, same species, Port Hacking