THE AUSTRALIAN

Entomologist

published by THE ENTOMOLOGICAL SOCIETY OF QUEENSLAND

<P

€

Volume 39, Part 3, 15 September 2012 Price: $8.00 per part

ISSN 1320 6133

THE AUSTRALIAN ENTOMOLOGIST

ABN#: 15 875 103 670

The Australian Entomologist is a non-profit journal published in four parts annually by the Entomological Society of Queensland and is devoted to entomology of the Australian Region, including New Zealand, Papua New Guinea and islands of the south-western Pacific. Articles are accepted from amateur and professional entomologists. The journal is produced independently and subscription to the journal is not included with membership of the society.

The Publications Committee

Editor: Dr D.L. Hancock Editorial Co-ordinator Assistant Editors: Dr C.J. Burwell Dr F. Turco

Queensland Museum Business Manager

Dr F. Turco Dr G.B. Monteith

Queensland Museum (geoff.monteith@bigpond.com) Subscriptions

Subscriptions are payable in advance to the Business Manager, The Australian Entomologist, P.O. Box 537, Indooroopilly, Qld, Australia, 4068.

For individuals: A$33.00 per annum in Australia. A$40.00 per annum in Asia-Pacific Region. A$45.00 per annum elsewhere. For institutions A$37.00 per annum in Australia. A$45.00 per annum in Asia-Pacific Region. A$50.00 per annum elsewhere. Please forward all overseas cheques/bank drafts in Australian currency.

GST is not payable on our publication.

ENTOMOLOGICAL SOCIETY OF QUEENSLAND

Membership is open to anyone interested in Entomology. Meetings are normally held at the Ecosciences Precinct, Dutton Park, at 1.00pm on the second Tuesday of March-June and August-December each year. Meetings are announced in the Society's News Bulletin which also contains reports of meetings, entomological notes, notices of other Society events and information on Members' activities. Enquiries relating to the Society should be sent to the Honorary Secretary, Entomological Society of Queensland, P.O. Box 537, Indooroopilly, Qld, 4068.

Cover: A male of Megacmonotus magnus (McLachlan 1871), one of the largest of the Australian members of the lacewing family Ascalaphidae. Ascalaphids are sometimes known as “owl flies” and many are partly daytime active. This species has a wing length of up to 45 mm and is very widespread in Australia, being recorded from all mainland states except Victoria and South Australia. The strange process jutting up from the base of the abdomen is found in many male ascalaphids and is of unknown function.

The illustration is reproduced by permission from CSIRO’s Insects of Australia and is by the late Mary Quick, one of the many talented artists who worked in the 1960s on the hundreds of new insect illustrations for the first edition of this classic work.

Australian Entomologist, 2012, 39 (3): 97-104 97

BIODIVERSITY DISCOVERY PROGRAM BUSH BLITZ SUPPLIES MISSING ANT SPIDER FEMALES (ARANEAE: ZODARIIDAE) FROM VICTORIA

BARBARA C. BAEHR' and ROBERT WHYTE Se

! Queensland Museum, PO Box 3300, South Brisbane, gla OL an bak ALY Environmental and Life Sciences, University of Nescastle7Callaghan, NSW 2308™ (Email: Barbara.Baehr reanfaly ‘GOV.aU) ¢ "ah Queensland Museum, PO Box 3300, S uth brisbark, iát 2012 (Email: robertw. hyteus@ & EEN om)

Abstract AN Ss Meee,

Bush Blitz 2011, the biodiversity discovery partnership program be between _the Australian Government, BHP Billiton and Earthwatch Australia, has yielded “key specimens of several

zodariid species from Ned’s Corner Station on Victoria’s far north-west desert fringe, some previously known only from male holotypes. Females of Pentasteron sordidum Baehr & Jocqué, 2001 and Pentasteron storosoides Baehr & Jocqué, 2001 are described for the first time.

Introduction

Ned’s Corner Station, a former sheep station on the fringe of the desert in Victoria’s far north-west, is managed for conservation by Trust for Nature as part of the National Reserve System. It is a 30,000 ha property purchased in 2002 because of its importance in Victoria's conservation landscape.

The property is bordered by National Park to the south and the Murray River to the north. It provides important habitats for native plants and wildlife

rarely, if ever, seen in other parts of the State. Plant species local to the region have been planted and nurtured.

The reserve is dominated by open chenopod scrublands (Fig. 2) covering 88% of the land. Open Red Gum floodplain woodlands (Fig. 3) occupy 1.5% and Black Box floodplain woodlands (Fig. 4) occupy 3%.

Bush Blitz 2011 at Ned’s Corner Station involved about 40 people, 30 of them leading Australian scientists. The 2011 team found many species new to science, including 14 new spider species.

Collection of zodariids yielded five species in four genera. All of them mimic ant behaviour and live with ants while feeding on them. Their mimicry extends in some cases to their ability to produce ant pheromones (Allan et al. 1996). One male and seven female specimens of Pentasteron sordidum Baehr & Jocqué, 2001 (Figs 5-12), previously known only from the male holotype, were collected. Large numbers of Pentasteron storosoides Baehr & Jocqué,

2001 (Figs 13-20), also known only from the male holotype, were found, 44 males and six females being collected.

Other zodariids collected include Habronestes raveni Baehr, 2003 (Fig. 1),

Holasteron spinosum Baehr, 2004 (Figs 21-23) and Zillimata scintillans (O.P.- Cambridge, 1869) (Figs 24-26).

98 Australian Entomologist, 2012, 39 (3)

This paper provides colour images of these species and the first description of the females of P. sordidum and P. storosoides.

Figs 1-4. (1), Habronestes raveni female (S91142, Photo by Mark Norman); inset at lower right = epigyne ventral view (Scale = 0.1 mm). (2-4), Ned’s Corner main habitats: (2) open chenopod scrubland, (3) open Red Gum woodland and (4) open Black Box woodland fringing the Murray River.

Australian Entomologist, 2012, 39 (3) 99

Material and methods

All zodariids were collected using pitfall traps and bark spraying. The latter technique involved thoroughly spraying the trunks of large trees using hand- held cans of Mortein Fast Knockdown insecticide, directing the jet of spray from the base to as far as possible up the trunk. Specimens were examined using a LEICA MZI16A microscope. Photo-micrographic images were produced using a Leica DFC 500 and the software program Auto-Montage Pro Version 5.02 (p). Female genitalia were cleared with pancreatin, as

described by Alvarez-Padilla and Hormiga (2008). All measurements are in millimeters.

Abbreviations are used in the text as follows: A — atrium; ALE — anterior lateral eyes; AME — anterior median eyes; CD — copulatory- duct; EA — embolar apophysis; E — embolus; LTA — lateral tegular apophysis; PLE — posterior lateral eyes; PME — posterior median eyes; S — spermathecae; VTA — ventral tegular apophysis. Institutional abbreviations used are: MV — Museum of Victoria, Melbourne; QM — Queensland Museum, Brisbane.

Systematics Family Zodariidae Thorell, 1881

Pentasteron Baehr & Jocqué, 2001 Type species: Pentasteron simplex Baehr & Jocqué, 2001.

Diagnosis. Members of this genus can be identified by the male palpal tibiae, which have a deep retrolateral concavity combined with a pronounced concavity on the base of the cymbium. The tegulum has a broad base traversed by the seminal duct. It ends in a typical median apophysis (VTA)

with a curved tip. Males of the following species were described by Baehr and Jocqué (2001).

Pentasteron sordidum Baehr & Jocqué, 2001 (Figs 5-12, 27)

Type material examined. Holotype 3, AUSTRALIA: New South Wales, Lake Wytchugga, 6 km W of Wilcannia, 31°30'S, 143°26'E, black box bark spray, 21- 22.xii.1998, M. Baehr, deposited in QM (S46889).

‘Other material examined. 1 9, Victoria, Ned’s Corner, 34°07’S, 141°17’E, pitfall, 22- 29.xi.2011, B. Baehr, deposited in MV (K-11542); 1 Q, same data as above (S91132); 1 ĝ, same data except 34°08’S, 141°16’E (S91133); 1 Q, same data except 34°07’S, 141°17’°E (S91136); 4 99, same data except 34°12’S, 141°31’E (S91134, S91135).

Diagnosis. Males and females resemble P. storosoides in having a shiny black abdomen with two pairs of white spots in the front half and 3 crescent- shaped spots in front of the spinnerets (Fig. 6). The male palp has a deep tibial concavity but can be easily separated from other species by the large longitudinal ventrolateral swelling (Figs 9-10). Females can be separated from P. storosoides by the long inverted u-shaped atrium (Figs 11-12).

100 Australian Entomologist, 2012. 39 (3)

Female. Total length 6.35; carapace 2.45 long, 1.53 wide. Colour: Carapace chestnut brown; chelicerae and sternum medium brown; coxae Pale; trochanter I-IV yellowish brown; femora I-IV white in proximal half, yellow overlaid with dark brown in distal half; other parts yellow. Abdomen dorsally shiny black with two pairs of white spots in the front half and 3 crescent- shaped spots in front of the spinnerets. Ventrally, yellowish in front of the epigastric fold and on lip in front of the tracheal spiracle. Carapace finely granulated; sternum smooth. Eyes: AME: 0.15; ALE: 0.13; PME: 0.14; PLE: 0.17; both eye rows strongly procurved. Epigyne (Figs 11-12) with long inverted u-shaped atrium, short curved copulatory ducts and laterally situated spermathecae (S).

Distribution. Western New South Wales and northwestern Victoria (Fig. 27).

Figs 5-12. Pentasteron sordidum: (5, 7, 9, 10) male (S91133); (6, 8, 11, 12) female _ (K-11542): (5-6) habitus dorsal view; (7-8) same ventral view; (9) right palp ventrolateral view; (10) same ventral view; (11) epigyne ventral view; (12) same dorsal view. Scale = habitus | mm, genitalia 0.1 mm.

Australian Entomologist, 2012, 39 (3) 101

Pentasteron storosoides Baehr & Jocqué, 2001 (Figs 13-20, 27) Type material examined. Holotype 6, AUSTRALIA: New South Wales, 30 km SW of Wilcannia, 32°25’S, 142°45’E, black box bark spray, 22.xii.1998, U. & M. Baehr, deposited in QM (846948).

Other material examined. | 2, Victoria, Ned’s Corner, 34°12’S, 141°31’E, pitfall, 22- 29.xi.2011, B. Baehr, deposited in MV (K-11544); 24 34, 2 99, same data as above (S91124, $91126); 3 3S, same data except 34°08’S, 141°19’E (S91125, S91130); 4 3d, same data except 34°08’S, 141°18’E (S91127); 3 Jd, 3 QF, same data except 34°07’S, 141°16’E (S91128); 5 GS, same data except 34°07’S, 141°17 E (S91129); 5 3, same data except 34°12’S, 141°31’E (S91131).

Diagnosis. Males resemble P. sordidum in having a palp with’ deep. tibial concavity delimited by the large longitudinal swollen ventrolateral swelling but can be separated by a dorsolateral apophysis with recurved tip (Figs 17- 18). Females can be separated by the small inverted v-shaped atrium and large coiled copulatory ducts ending in ventrally directed spermathecae (Figs 19-20).

eee ae t _ i a BUE :

ss ge ey

Figs 13-20. Pentasteron storosoides: (13, 15, 17, 18) male (S91124); (14, 16, 19, 20) female (K-11542): (13-14) habitus dorsal view; (15-16) same ventral view; (17) right palp ventrolateral view; (18) same ventral view; (19) epigyne ventral view; (20) same dorsal view. Scale = habitus 1 mm, genitalia 0.1 mm.

102 Australian Entomologist, 2012, 39 (3)

Female: Total length 5.56; carapace 2.52 long, 1.63 wide. Colour: Carapace chestnut brown; chelicerae and sternum medium brown; coxae white with dark brown rim; trochanter I - IV dark; femora I—IV white with dark patches at base in proximal half, dark brown in distal half; remainder of legs yellowish brown, posterior tibiae with blackish lateral streaks. Abdomen shiny black; dorsum with two pairs of small white spots, one pair near the anterior edge, the other pair roughly half way towards the rear. Three crescent-shaped. spots are in a line running lengthways immediately in front of spinnerets. The sides have one oblique white spot and pale mottling. Venter sepia, anterior lip of tracheal spiracle yellow brown. Carapace finely granulated; sternum smooth. Eyes: AME: 0.09; ALE: 0.13; PME: 0.14; PLE: 0. 14. both eye rows strongly procurved. Colulus a small swelling with 8 setae. Epigyne with small inverted v-shaped atrium, large coiled copulatory ducts ending in ventrally directed spermathecae (Figs 19-20).

Distribution. Western New South Wales and northwestern Victoria (Fig. 27).

Figs 21-26. (21-23) Holasteron spinosum male (S91139); (24-26) Zillimata scintillans male (S91137). (21, 24) habitus dorsal view; (22, 25) right palp retolateral view; (23, 26) same ventral view. Scale = habitus 1 mm, palps 0.1 mm.

Australian Entomologist, 2012, 39 (3) 103

Habronestes raveni Baehr, 2003 (Fig. 1) Material examined. 1 Q, Victoria, Ned’s Corner, 34°12’S, 141°31°E, pitfall, 22- 29.xi.2011, B. Baehr (S91142). Holasteron spinosum Baehr, 2004 (Figs 21-23) Material examined. 10 63, 4 2, Victoria, Ned’s Corner, 34°12’S, 141°32’E, pitfall, 22-29.xi.2011, B. Baehr (S91142, $91140); 1 ĝ, same data except 34°23’S, 141°20°E, pitfall, 23.xi.2011, P. Lillywhite (S91141). Zillimata scintillans (O.P.-Cambridge, 1869) (Figs 24-26) Material examined. 1 ĝ, Victoria, Ned’s Corner, 34°12’S, 141°31’E, pitfall, 22-

> oe

29.xi.2011, B. Baehr (S91138); 1 ĝ, same data except 34°07’S, 141°17°E, pitfall, 22- 29.xi.2011, B. Baehr (S91137).

iia eer -e Wak ers - sy ABE į VSA o f ea - 9) shy 7 yo ae don, am is, d y 5 g f j pf err \ ae ‘e w : ) “Gs, ay ; f \ ot Si AEN i t aX ens A Y co \ aa’ } A 2 pa A ESN poer S Af | Vey } \ ; | x { ~ PN | | Vy 4 | ® | — í \ % \ aes \ } \ ° ( | m : / } 7 È A i Ne cade cee iy iy ae Tin s- / 5 ATS NUD a c N aa { f > we) RE ster ce > A My

Fig. 27. Distribution map of P. sordidum (circle) and P. storosoides (star); Ned’s Corner, arrowed.

Acknowledgements

This paper would not have been completed without the support of the Bush Blitz program provided through ABRS. We would like to thank Jo Harding (Bush Blitz Manager), Kate Gillespie (Bush Blitz Senior Project Officer), Mim Jambrecina (Senior Project Officer - Bush Blitz Program) and the Bush Blitz team for their efficient support in the field.

104 Australian Entomologist, 2012, 39 (3)

References

ALLAN, R.A., ELGAR, M.A. and CAPON, R.J. 1996. Exploitation of an ant chemical alarm signal by the zodariid spider Habronestes bradley Walckenaer. Proceedings of the Royal Society of London 263: 69-73.

ALVAREZ-PADILLA, F. and HORMIGA, G. 2008. A protocol for digesting internal soft tissues and mounting spiders for scanning electron microscopy. Journal of Arachnology 35: 538- 542.

BAEHR, B.C. 2003. Revision of the Australian spider genus Habronestes (Araneae: Zodariidae). Species of New South Wales and Australian Capital Territory. Records of the Australian Museum 55: 343-376.

BAEHR, B.C. 2004. Revision of the new Australian genus Holasteron (Araneae: Zodariidae): taxonomy, phylogeny and biogeography. Memoirs of the Queensland Museum 49: 495-519. BAEHR, B.C. and JOCQUÉ, R. 2001. Revisions of the genera in the Asteron-complex (Araneae, Zodariidae). The new genera Pentasteron, Phenasteron, Leptasteron and Subasteron. Memoirs of the Queensland Museum 46: 359-385.

CAMBRIDGE, O.P. 1869. Descriptions and sketches of some new species of Araneidea, with characters of a new genus by O.P.- Cambridge. Annals and Magazine of Natural History 3: 52- 74.

Australian Entomologist, 2012, 39 (3): 105-108 105

FIRST RECORD OF GYNAIKOTHRIPS UZELI (ZIMMERMANN) (THYSANOPTERA: PHLAEOTHRIPIDAE) FROM AUSTRALIA

DESLEY J. TREE

Queensland Primary Industries Insect Collection (QDPC), Department of Agriculture, Fisheries and Forestry, Queensland, Ecosciences Precinct, GPO Box 267, Brisbane, Qld 4001

Abstract

Gynaikothrips uzeli (Zimmermann) is newly recorded from Queensland, Australia, causing leaf

galls on ornamental figs. Gynaikothrips uzeli is considered a pest of Ficus benjamina (Moraceae)

(Weeping fig) in southern Asia and America.

Introduction

Late in 2011, thrips specimens galling leaves of an unidentified ornamental fig near Cape York in northern Queensland were collected by Plant Biosecurity Queensland staff and sent to the author for identification. They were identified as Gynaikothrips uzeli (Zimmermann), a thrips not previously recorded from Australia (Fig. 1). These specimens have been lodged in the QDPC Insect Collection, Ecosciences Precinct, Brisbane, Queensland.

Gynaikothrips uzeli is native to Southern Asia and has been recorded from China, Hong Kong, Taiwan, India, Maldives, Singapore, USA, Mexico, Trinidad and Tobago, Costa Rica and Brazil (Anathakrishnan 1978, Mound et al. 1995, Mound and Marullo 1996, Held et al. 2005, Tree and Walter 2009, Cambero et al. 2010, Brito et al. 2012, D.J. Tree pers. obs. 2007, 2012). Leaf galls are induced by adults and larvae, which feed only on young leaves of Ficus benjamina — one of two common ornamental figs grown widely across Australia (Fig. 2), causing leaves to fold and/or curl (Fig. 3). The other common ornamental fig tree in Australia is Ficus microcarpa.

Discussion

The genus Gynaikothrips contains 41 species worldwide (Mound 2012). Prior to late 2011, only three Gynaikothrips species were recorded from Australia: G. ficorum (Marchal) - known as the primary leaf galler of Ficus microcarpa; G. australis Bagnall - the primary leaf galler of Ficus macrophylla, Ficus obliqua and Ficus rubiginosa; while G. additamentus (Karny) shares the leaf galls of G. australis (Mound and Minaei 2007, Tree and Walter 2009).

Gynaikothrips uzeli is closely related to G. ficorum. Mound et al. (1995) noted that the differences between the two species were the length of the posteroangular setae and the species of Ficus that host their galls. Female G. uzeli usually have the pronotal posteroangular setae 0.7 times as long as the epimeral setae and always longer than the pronotal discal setae (Fig. 4). In contrast, female G. ficorum have the pronotal posteroangular setae no more than 0.5 times as long as the epimeral setae and usually no longer than the pronotal discal setae. The length of the pronotal posteroangular setae in males of G. uzeli and G. ficorum is too variable to use as a character state to differentiate between the two species.

| |

106 Australian Entomologist, 2012, 39 (3)

Figs 1-3. Gynaikothrips uzeli. (1) adult female; (2) eggs and feeding life stages, larvae and adults, inside a leaf gall; (3) leaf galls on Ficus benjamina in Brisbane, Qld.

Despite the indicated differences between the females, variation in the length of the pronotal posteroangular setae of G. uzeli and G. ficorum can cause confusion in their identification (Mound et al. 2005, Mound and Marullo 1996, Goldarazena et al. 2008). Mound and Marullo (1996) suggested that G. ficorum could possibly be a ‘single, highly selected strain of G. uzeli which has been spread around the world by the horticultural trade’. Gynaikothrips

Australian Entomologist, 2012, 39 (3) 107

uzeli males have the pore plate on sternite VIII as a round central spot, whereas G. ficorum pore plates can be either the same as G. uzeli or a wide band across sternite VIII that continues around onto the lateral margins of tergite VIII as two round spots. However, these differences do not seem to be consistent, with some G. uzeli males having similar pore plates to those of G. ficorum. Further studies, such as molecular analysis and field work (including correct identification of hosts), are required to enable a clearer understanding of the relationships among the species of Gynaikothrips and, in particular, the relationship between G. ficorum and G. uzeli.

Fig. 4. Pronotum of Gynaikothrips uzeli, showing posteroangular setae (a) as long as the epimeral setae (b) and longer than the discal setae (c).

Since late 2011, G. uzeli has been recorded from near Cairns, Innisfail and Brisbane, all in Queensland. It is likely to spread further in Australia wherever Ficus benjamina grows. Prior to 2011 there are no records of any Gynaikothrips species inducing leaf galls on Ficus benjamina in Australia.

References

ANANTHAKRISHNAN, T.N. 1978. Thrips galls and gall thrips. Zoological Survey of India Technical Monograph 1: 1-95.

BRITO, R.O., ARTONI, R.F., VICARI, M.R., NOGAROTO, V., SILVA JR, J.C., MATIELLO, R.R. and ALMEIDA, M.C. 2012. Population structure and genetic diversity analysis in Gynaikothrips uzeli (Zimmermann, 1909) (Thysanoptera: Phlaeothripidae) by RAPD markers. Bulletin of Entomological Research 102: 345-351.

108 Australian Entomologist, 2012, 39 (3)

CAMBERO-CAMPOS, J., VALENZUELA-GARCIA, R., CARVAJAL-CAZOLA, C., RIOS- VELASCO, C., and GARCIA-MARTINEZ, O. 2010. New records for Mexico: Gynaikothrips uzeli, Androthrips ramachandrai (Thysanoptera: Phlaeothripidae) and Montandoniola confusa (Hemiptera: Anthocoridae). Florida Entomologist 93(3): 470-472.

GOLDARAZENA, A., MOUND, L.A., and ZUR STRASSEN, R. 2008. Nomenclatural problems among Thysanoptera (Insecta) of Costa Rica. Revista Biologia Tropical 56: 961-968. HELD, D.W., BOYD, D., LOCKLEY, T. and EDWARDS, G.B. 2005. Gynaikothrips uzeli (Thysanoptera: Phlaeothripidae) in the southeastern United States: distribution and review of biology. Florida Entomologist 88(4): 538-540.

KARNY, H. 1924. Results of Dr. E. Mjöberg’s Swedish scientific expeditions to Australia 1910- 1913. 38. Thysanoptera. Arkiv för Zoologi 17A(2): 1-56.

MARCHAL, P. 1908. Sur une nouvelle spèce de Thrips (Thysanoptera) nuisable aux Ficus en Algérie. Bulletin Société Entomologique de France 14: 251-253.

MOUND, L.A. 2012. Thysanoptera (Thrips) of the World — a checklist. {Accessed 28.iv.2012.] Available from URL: http:/www.ento.csiro.au/thysanoptera/worldthrips.html

MOUND, L.A. and MARULLO, R. 1996. The thrips of Central and South America: an introduction. Memoirs on Entomology. International; vi + 488 pp.

MOUND, L.A. and MINAEI, K. 2007. Australian insects of the Haplothrips lineage (Thysanoptera—Phlaeothripinae). Journal of Natural History 41; 2919-2978. http://pdfserve. informaworld.com/8693 19_751315335_789049544. pdf

MOUND, L.A., WANG, C-L. and OKAJIMA, S. 1995. Observations in Taiwan on the identity of the Cuban Laurel thrips (Thysanoptera, Phlaeothripidae). Journal of the New York Entomological Society 103(2): 185-190.

TREE, D.J. and WALTER, G.H. 2009. Diversity of host plant relationships and leaf galling behaviours within a genus of thrips — Gynaikothrips and Ficus in south east Queensland, Australia. Australian Journal of Entomology 48: 269-275.

ZIMMERMANN, A. 1900. Ueber einige javanische Thysanoptera. Bulletin de l'Institut Botanique de Buitenzorg 7: 6-19.

Australian Entomologist, 2012, 39 (3): 109-116 109

STUDIES OF AUSTRALIAN HYDROBIOSELLA TILLYARD (TRICHOPTERA: PHILOPOTAMIDAE): TWO NEW AUSTRALIAN SPECIES FROM NORTH QUEENSLAND

DAVID I. CARTWRIGHT 13 Brolga Crescent, Wandana Heights, Vic 3216 (Email: cartwright@ hotkey.net.au) Abstract

Two species of philopotamid caddis fly, Hydrobiosella eminentia sp. n. and H. ferrata sp. n. are newly described from Australia, based on features of the male genitalia. Both species are endemic to northeastern Queensland and share a unique feature in the genitalia, notably a pair of slender, elongate preanal processes situated basolaterally to segment X. On this basis they are assigned to a new species group within the genus Hydrobiosella Tillyard, the H. eminentia group. A key is provided for identification of all Australian Hydrobiosella species groups. Introduction

The first Australian species in the genus Hydrobiosella Tillyard were recognised only in 1953 with the transfer of H. michaelseni (Ulmer, 1908) from Dolophilus McLachlan and description of H. arcuata Kimmins, H. bispina Kimmins, H. cognata Kimmins, H. tasmanica Mosely and H. waddama Mosely (in Mosely and Kimmins 1953). Subsequently, additional species were described: H. letti Korboot (1964); H. armata Jacquemart (1965); H. anasina, H. cerula, H. corinna, H. orba and H. sagitta Neboiss (1977), H. amblyopia Neboiss (1982); H. anatolica, H. disrupta, H. otaria, H. propinqua, H. scalaris and H. tahunense Neboiss (2003), and, most recently, ten new species in the H. bispina group: Cartwright (2010). Forty species of Hydrobiosella are known worldwide: from Australia (30 species), New Zealand (4 species: Morse 1999) and New Caledonia (6 species: Espeland and Johannson 2007).

Neboiss (1977) separated the Tasmanian species into three groups based primarily on male genitalia — the H. corinna group, the H. tasmanica group and H. waddama. Cartwright (2010) expanded this to include a key to the Australian mainland species and H. bispina species group. The description of two new species here brings to 32 the total number of Australian species of Hydrobiosella. The Australian mainland species in the H. waddama group are currently being reviewed (Cartwright in prep.).

In this taxonomic paper a new species group, the H. eminentia group, is proposed to incorporate two new species from northern Queensland described below: H. eminentia and H. ferrata. Males of the two species in the Hydrobiosella eminentia group are the only Australian mainland species of Hydrobiosella known to have preanal appendages. These appendages in the H. eminentia group are more slender and elongate than similar ‘appendages’ reported for Tasmanian (notably the H.corinna and H. tasmanica groups), New Zealand (including the type species, H. stenocerca Tillyard) and New Caledonian species (including H. mouensis Espeland and Johanson). Hydrobiosella mouensis has a pair of elongate tubular processes attached

110 Australian Entomologist, 2012, 39 (3)

basally on segment ten with flat superior appendages present at lateral part of tubular processes (Espeland and Johanson 2007).

In this taxonomic revision of the Australian Hydrobiosella eminentia group only three male specimens were examined and referred to two species, H. eminentia and H. ferrata. Hydrobiosella eminentia was listed in the checklist of Walker et al. (1995) as Hydrobiosella sp. nov. PT-1039. The two new species in the Hydrobiosella eminentia group are from northeastern Queensland (latitudinal range 12°44'-17°16'S), within the Torresian region. Two species within the H. bispina group, H. unispina Cartwright and H. dugerang Cartwright, were also recorded from north Queensland (Cartwright 2010), in contrast to the more southern Bassian distributions of the other 28 described Hydrobiosella species (SE Queensland, New South Wales, Victoria, Tasmania and southwestern Australia). This mainly Bassian Australian distribution, together with a distribution of the genus that is otherwise restricted to New Zealand and New Caledonia, is generally suggestive of a ‘southern’ origin. Hydrobiosella ferrata is the most northerly species of Hydrobiosella known, recorded from a latitude of 12°44' S, H. eminentia from 17°16'S and the six New Caledonian species are reported further south at between 20°24'-25°S' S (Espeland and Johannson 2007).

Ross (1956) recognised Hydrobiosella as a subgenus of Sortosa Navas and postulated an original ancestral form that gave rise to two lines, a New Caledonian — New Zealand lineage (with small or reduced cerci (= preanal appendages)) and one in Australia (without cerci but with basal ridge or process of ninth tergite). The presence of preanal appendages in the two species described here from far northern Queensland, as well as in all New Caledonian, New Zealand and some Tasmanian species (H. corinna group), suggests other possibilities. When all Australian Hydrobiosella species groups have been revised then relationships of the Australian groups with species in New Zealand and New Caledonia can then be properly assessed.

Methods and abbreviations

Among Hydrobiosella species, size and body and wing colour can be useful taxonomic characters but are variable. Colour can be a useful character in freshly preserved material but, with time, it often fades in alcohol. The three H. eminentia group specimens examined in this study were stored in alcohol for 30 years or more. The material studied was on loan from Museum Victoria and made available by Dr Arturs Neboiss. All specimens, including types, mentioned in the text are lodged in the Museum Victoria, Melbourne (NMV).

Males of each species are most readily distinguished by genitalic features but often require clearing of the abdomen in potassium hydroxide.

Figured specimens are identified by the notebook numbers of Dr Arturs Neboiss (prefix PT-) or the author (prefix CT-). Terminology used generally

Australian Entomologist, 2012, 39 (3) 111

follows that of Neboiss (1977, 1982), Blahnik (2005) and Holzenthal et al. (2007). Abbreviations for genitalic parts are indicated on selected figures. Typically, setae or spines are illustrated only on the right side of the figure (as viewed) to enable a better view of the underlying structures. Length/width measurements generally refer to the maximum length divided by the maximum width.

Previous authors have used a confusing variety of names for the same or similar structures e.g. preanal appendages (homologous/or analogous structures to some or all of the following — cerci, shoulder-like projection or basal ridge or process of tenth tergite in Ross 1956; = superior appendages in Neboiss 1977, Henderson 1983; = superior appendages or tubular processes of Espeland and Johanson 2007; = preanal appendages in Holzenthal et al. 2007, Cartwright 2010).

Key to males of known Australian groups (or ungrouped species)

of Hydrobiosella Tillyard (updated after Cartwright 2010)

l Phallus without pair of parameres (Figs 2-3, 5-6; Neboiss 1986, figs pp 99, H. amblyopia; 101, H. tasmanica; 102, H. corinnd) .......ceoeeeene 2

— Phallus with pair of parameres (Cartwright 2010, figs 2-3; Neboiss 1986, figs pp 99, H. michaelseni, H. waddama; 101, H. letti, 102, H. bispina) PE KOAA N EA PE N EEE INE AEREN E ONNEEN SE 5

2 Preanal appendages present, usually small (Figs 2-3, 5-6; Neboiss 1977,

figs 204-205, 216-217; Neboiss 1986, figs pp 101, H. tasmanica; 102, H. corinna; Neboiss 2003, figs 8ah) E a A E N 3}

— Preanal appendages absent (Neboiss 1986, figs pp 99, H. amblyopia; 101; H. tasmanica)

3 Preanal appendages relatively slender, elongate and ‘unattached’ to segment IX (Figs 2-3, 5-6); NE-Qld ....... Hydrobiosella eminentia group — Preanal appendages often short and bulbous or ‘attached’ to segment IX (Neboiss, 1977, figs 204-211; Neboiss 1986, figs p. 102, H. corinna; Neboiss, 2003, figs 8A-H); Tas ............... Hydrobiosella corinna group 4 Phallus apically with downward projecting spine(s) (Neboiss 1977, figs 216-221, 225-226; Neboiss 1986, figs p. 101, H. armata, H. tasmanica; Neboiss 2003, figs 1OA-G, L1A-G, 12A-F); Tas ....ssserseresererseseerese

E A OE E A AS A bxdoddal OE T Hydrobiosella tasmanica group — Phallus apically without downward projecting spine(s) (Neboiss 1982, fig. 12; Neboiss 1986, figs p. 99 H. amblyopia); S-WA ............. cece eee BAA An Rent E A E A A EET AA H. amblyopia (ungrouped) 5 Inferior appendages with harpago with dark row of setae forming fringe along ventral margin (Cartwright 2010, figs 3, 6; Neboiss 1986, figs pp

112 Australian Entomologist, 2012, 39 (3)

102, H. bispina; 103, H. arcuata), E-Vic, E-NSW, E-Qld .............54++ Hydrobiosella bispina group

— Inferior appendages with harpago without dark row of setae forming fringe along ventral margin (Neboiss 1986, figs pp 99, H. michaelseni, H. WwaddamaO TRH Betti) erent: ene eet cin Moe et eT een teeth seats 6

6 Parameres elongate and sinusoidal, attached ventrally to base of phallus (Cartwright in prep., figs 2-3, 5-6; Neboiss 1977, fig. 233; Neboiss 1986, figs p. 99, H. waddama; Neboiss 2003, figs 12g-h); Tas, SE Aust.

Hydrobiosella waddama group

— Parameres not elongate and sinusoidal, not attached ventrally to base of phallus (Neboiss 1982, figs 9-10; Neboiss 1986, figs pp 99, H.

michaelseni al O LES letti) perros Pan eet ents te ee TET A 7 7 Parameres curved strongly and crossed (Neboiss 1982, figs 9-10; Neboiss 1986;ifigsipa99:20: michaelseni) oW A E T st ets AAE rst:

Hydrobiosella michaelseni (Ulmer) (unplaced to group)

— Parameres not curved strongly and crossed (Neboiss 1986,.figs p. 101, H. letti); CE-NSW ............ Hydrobiosella letti Korboot (unplaced to group)

Systematics Hydrobiosella Tillyard

Hydrobiosella Tillyard 1924: 288; Mosely and Kimmins 1953: 387; Neboiss 1977: 45; Neboiss 2003: 55; Espeland and Johanson 2007: 92.

Type species: Hydrobiosella stenocerca Tillyard, by monotypy.

Hydrobiosella eminentia group

Diagnosis. The diagnostic characters of the males of this group of two species are the obvious pair of slender and elongate preanal processes situated baso-laterally to segment X and the relatively simple phallus which lacks associated spines or parameres.

Description. Male. Wings light brown to brown, medium-sized. Forewing length, males: 4.3-5.2 mm; forewing length about 3 times width, wing venation (Fig. 1) similar to the type species H. stenocerca (Mosely and Kimmins 1953, fig 265a) and H. waddama (Mosely and Kimmins 1953, fig 269a), R1 simple, forks 1, 2, 3, 4 and 5 present; forks 1 and 2 sessile; fork 2 with nygma, length about 1.3-1.4 times length fork 1; fork 3 shorter, length 0.6-0.7 times length fork 2, fork length ranging from between |.7—1.8 times length footstalk, fork 4 similar in length to fork 3, length fork about three times length footstalk; fork 5 very long, length about 1.7 times length fork 4. Hind wing length about 2.3-2.7 times width, with forks 1, 2, 3 and 5 present; forks 1 and 2 sessile, fork 2 with nygma, length about 1.5 times length fork 1; fork 3 shorter, about 0.6-0.7 times length fork 2, fork 3 longer than footstalk,

Australian Entomologist, 2012, 39 (3) 113

length fork ranging between 1.7—1.8 times length footstalk; fork 5 very long, length between 1.9-2 times length fork 3; discoidal cell closed, length

between 3.7-4.8 times maximum width; with two or possibly three longer anal veins (Fig. 1).

Male. Sternite IX either with a small shallow notch (Fig. 4) or a medial knob on distal margin (Fig. 7). Segment X with a simple process, sclerotised dorsally; preanal appendages relatively elongate and slender, situated baso- laterally to segment X. Phallus generally tube-like, without any obvious spines or parameres. Inferior appendages 2-segmented, basal segment robust, slightly longer or similar in length to harpago, which is more slender and has a small field of dark spines apically (Figs 2, 3, 5, 6).

Female and larvae. Unknown.

Key to males of species of the Australian Hydrobiosella eminentia group

1 Segment X long and slender (Figs 2-3), in dorsal view, length about 3

times width (Fig. 2); sternite IX with a small, shallow notch medially on distalimarsin| (Riot) Perens csternn ai eee raya amen te H. eminentia sp. n.

Segment X not long and slender (Figs 5-6), in dorsal view, robust, length

about 1.5 times width (Fig. 6); sternite IX with a small knob medially on E EN A EA orae aumo iaa H. ferrata sp. n.

Hydrobiosella eminentia sp. n. (Figs 1-4)

Types. Holotype 3: QUEENSLAND, Mt Bartle Frere, 0.5 km N of S peak (about

17°16'S, 145°54'E), 1500 m, 6-8.xi.1981, Earthwatch-QM (NMV, T- 21250). Paratype & (specimen PT-1039 figured), collected with holotype (NMV).

Diagnosis. Hydrobiosella eminentia can be separated from H. ferrata by the

long and slender segment X and small shallow notch medially on distal margin sternite IX.

Description. Wings (Fig. 1), similar to H. stenocerca (Mosely and Kimmins

1953, fig. 265a) and H. waddama (Mosely and Kimmins 1953, fig. 269a). Length of forewing: male 5.2 mm.

Male. Sternite IX with a small shallow notch on ventromedial-distal margin (Fig. 4). Segment X mainly sclerotised, broadest basally, long and slender distally; in dorsal view, length about 3 times width (Fig. 2); in lateral view slender, slightly upcurved distally. Preanal appendages slender, elongate, situated baso-laterally to segment X; length about 0.6 times length of tergum X (Fig. 3). Phallus generally tube-like, slightly bulbous apically (Figs 2-3). Inferior appendages 2-segmented: in lateral view, basal segment sub- rectangular, length about 1.7 times width and about 1.3 times length of

harpago; harpago slightly more slender, length about twice width, tapered slightly distally (Fig. 3).

114 Australian Entomologist, 2012, 39 (3)

Female. Unknown.

Etymology. Eminentia - Latin for projection, prominence, in reference to the preanal appendages.

Remarks. Hydrobiosella eminentia is probably a rare and restricted species, since, despite considerable collecting in the Wet Tropics of northeastern Queensland, it is known only from the type locality at Mt Bartle Frere.

harpago

s inferior appendage

+, preanal appendages

preanal 3 2 6

pendaces segment x

Á __-phallus harpayo

Figs 1-7. Hydrobiosella eminentia group species. (1-4) Hydrobiosella eminentia sp. n.: (1) wings, apical section of forewing and hind wing; (2-4) male genitalia in dorsal, lateral and part ventral views; (2) dorsal; (3) lateral; (4) ventral, medioventral-distal margin of segment IX. (5-7) Hydrobiosella ferrata sp. n.: male genitalia in dorsal, lateral and part ventral views; (5) dorsal; (6) lateral; (7) ventral, medioventral-distal margin of segment IX.

Australian Entomologist, 2012, 39 (3) 115

Hydrobiosella ferrata sp. n. (Figs 5-7)

Type. Holotype 3 (specimen CT-562 figured): QUEENSLAND, Mt Tozer, Iron

Range (about 12°44'S, 143°12'E), 300 m, 30.1v.1973, S.R. Monteith (NMV, T- 21252).

Diagnosis. Hydrobiosella ferrata can be separated from H. eminentia by the robust segment X, dorsal view, and the ventromedial-distal margin of sternite IX produced in a small knob.

Description. Wings similar to H. eminentia (Fig. 1), length of forewing: male 4.3 mm.

Male. Sternite IX with a small knob on medio-distal margin (Fig. 7). Segment X mainly sclerotised, broadest basally, robust distally; in dorsal view, length about 1.5 times width; in lateral view, slender; preanal appendages slender, elongate, situated baso-laterally to segment X (Figs 5, 6). Phallus generally tube-like with a minute spine apically (Fig. 6). Inferior appendages 2- segmented: in lateral view, basal segment robust, length about twice width, sub-rectangular; harpago slightly more slender, sub-rectangular, inflexed apically (Fig. 6).

Female. Unknown.

Etymology. Ferrata — Latin for ‘relating to iron’, in reference to the type locality of Iron Range.

Remarks. Only a single male specimen of this probably rare and restricted species of Hydrobiosella has been collected from the Iron Range on Cape York Peninsula, northeastern Queensland (Latitude 12°44'S).

Acknowledgements

I thank the Department of the Environment and Water Resources, in particular the Australian Biological Resources Study (ABRS), for providing a grant to undertake this work. Thanks to the late Dr Arturs Neboiss who, whilst still active in research, provided access to the specimens and, together with Dr Alice Wells and John Dean, offered helpful advice on earlier drafts of this manuscript. The referees are thanked for their constructive comments. I am indebted to John Dean and Ros St Clair for technical assistance with scanning of the figures and for moral support during the project.

References

BLAHNIK, R.J. 2005. Alterosa, a new caddisfly genus from Brazil (Trichoptera: Philopotamidae). Zootaxa 991: 1-60.

CARTWRIGHT, D.I. 2010. Studies of Australian Hydrobiosella Tillyard: a review of the Australian species of the Hydrobiosella bispina Kimmins group.(Trichoptera: Philopotamidae). Memoirs of Museum Victoria 67: 1-13.

116 Australian Entomologist, 2012, 39 (3)

ESPELAND, M. and JOHANSON, K.A. 2007. Revision of the New Caledonian Hydrobiosella (Trichoptera: Philopotamidae) with description of five new species. Pp 91-102, In: Bueno-Soria. J., Barba-Alvarez, R. and Armitage, B. (eds), Proceedings of the XIIth International Symposium on Trichoptera. The Caddis Press.

HOLZENTHAL, R.W., BLAHNIK, R.J., PRATHER, A.L. and KJER, K.M. 2007. Order Trichoptera Kirby, 1813 (Insecta), Caddisflies. Zootaxa 1668: 639-698.

MORSE, J.C. (ed.). 1999. Trichoptera World Checklist. [Effective 27 March 1999, accessed 11 May 2011.] Available from URL: http:/entweb.clemson.edu/database/trichopt/index.htm MOSELY, M.E. and KIMMINS D.E. 1953. The Trichoptera (caddis-flies) of Australia and New Zealand. British Museum (Natural History), London; 550 pp.

NEBOISS, A. 1977. A taxonomic and zoogeographic study of Tasmanian caddis-flies (Insects: Trichoptera). Memoirs of the National Museum of Victoria 38: 1-208.

NEBOISS, A. 1982. The caddis-flies (Trichoptera) of south-western Australia. Australian Journal of Zoology 30: 271-325.

NEBOISS, A. 1986. Atlas of Trichoptera of the SW Pacific-Australian Region. Dr W. Junk, Dordrecht; 286 pp.

NEBOISS, A. 2003. New genera and species, and new records, of Tasmanian Trichoptera (Insecta). Papers and Proceedings of the Royal Society of Tasmania 136: 43-82.

TILLYARD, R.J. 1924. Studies of New Zealand Trichoptera or caddis flies no. 2. Descriptions of new genera and species. Transactions of the New Zealand Institute 55: 285-314.

WALKER, K., NEBOISS, A., DEAN, J. and CARTWRIGHT, D. 1995. A preliminary investigation of the Caddis-flies (Insecta: Trichoptera) of the Queensland Wet Tropics. Australian Entomologist 22: 19-31.

Australian Entomologist, 2012, 39 (3): 117-120 117

SCUTTLE FLIES (DIPTERA: PHORIDAE) FROM CORAL SEA ATOLLS

R. HENRY L. DISNEY’ and PENELOPE GREENSLADE”

'Department of Zoology, University of Cambridge, Downing Street, Cambridge CB2 3E]J, England. (E-mail: rhld2 @ hermes.cam.ac.uk)

Centre for Environmental Management, School of Science, Information Technology and Engineering, University of Ballarat, Mt Helen Campus, University Drive, Mt Helen, PO Box 663, Ballarat, Vic 3353. (E-mail: p.greenslade@ ballarat.edu.au) Abstract

Of 92 specimens of Phoridae collected from North East Herald Cay and Coringa Cay in the Coral Sea, there was | female Dohrniphora Dahl of a group only identifiable from males in our present state of knowledge. The rest were Megaselia spiracularis Schmitz, not previously recorded from the Australasian Region.

Introduction

The scuttle flies of Australia are poorly known. The keys of Borgmeier (1967a, b) provide a starting point. Keys to the species recorded from Tasmania (Disney 2003) and the most recent checklist for the Australian mainland (Disney 2008) update these works, along with Disney (201 1a).

PG asked RHLD to identify 14 samples of Phoridae, collected by herself and colleagues, from North East Herald Cay and Coringa Cay in the Coral Sea off

the north-east coast of Australia in 1995, 1997 and 2007 (Greenslade and Farrow 2007).

Methods

The specimens had been collected in pitfall traps and yellow pan traps. They were preserved in alcohol, deposited in the Australian Museum, Sydney and

sorted to family by Deborah Rich. RHLD mounted representative specimens on slides (Disney 2001).

Results

The samples represented 92 specimens belonging to the following two species.

Dohrniphora sp.

Material examined. CORAL SEA: 1 , North East Herald Cay, yellow pan trap, 15.v.2007, P. Greenslade & S. Donaldson (in Australian Museum, Sydney).

Males of the Australasian and Oriental species of Dohrniphora Dahl were keyed by Disney (1990), supplemented by Disney and Bartareau (1995) and Disney and Kistner (1997, 1999). Females can rarely be named when not associated with their males. The above female is not the widely distributed species D. cornuta (Bigot), although belonging to the same group of species (see the key to females in Disney and Bänziger 2009). It has a single pair of bristles on the scutellum but, unlike D. cornuta, the dorsal hair palisade of the mid tibia extends about 0.9 times its length and, in addition, it has an anterior

118 Australian Entomologist, 2012, 39 (3)

palisade extending about three quarters of its length. Until linked to its male it cannot be named.

Megaselia spiracularis Schmitz (Fig. 1) Megaselia spiracularis Schmitz, 1938: 81.

Material examined. CORAL SEA: 1 ĝ, North East Herald Cay (16.56°S, 149.11°E): 2.1i1.1995, S. Donaldson; 1 ĝ, same locality, pitfall trap, 5.iii.1995, S. Donaldson; 1 3, same locality, 1997, A. Anderson; 8 33, 2 99, same locality, 15.v.2007, pitfall traps, P. Greenslade; 49 33, 10 29, same data except yellow pan traps; 2 33, 1 Q, same locality; 15.v.2007 (43), 17.v.2007 (Q), yellow pan trap, P. Greenslade; 17 36, Coringa Cay (16.59°S, 149.53°E), pitfall traps, 17-19.iii.1995, S. Donaldson. (All in Australian Museum, Sydney).

The distinctive males of this species were included in a key by Borgmeier (1967a). The larval and pupal stages were described by Kaneko and Furukawa (1977), augmented by Liu et al. (2001).

These are the first records of this species for the Australasian Region. It has previously been recorded from the Eastern Palaearctic, the Oriental Region and New Zealand. It has been reared from dead snails in Japan (Schmitz 1938) and the larvae reported from human corpses in Malaysia (Thevan et al. 2010) and from cases of intestinal myiasis in Japan (Kaneko and Furukawa 1977). It has also been reported in a package of ‘sterile’ rodent feed imported into France from Japan (Disney 2011b). Being a saprophagous species, M. spiracularis will readily establish itself, if accidentally introduced by man or transported by the wind, in novel regions.

Fig. 1. Megaselia spiracularis: male, showing the characteristic enlarged abdominal spiracles.

Australian Entomologist, 2012, 39 (3) 119

Of the 91 M. spiracularis specimens, 85.7% were males. The yellow pan traps were more productive than pitfall traps, which is typical for Phoridae in which both sexes are winged (e.g. Disney et al. 1982). The use of pitfall traps for sampling Phoridae is useful for the flightless females of mainly myrmecophilous and termitophilus species.

Acknowledgements

PG’s field work was funded by the Department of the Environment, Water, Heritage and the Arts. Thanks are due to Dan Bickel and the Australian Museum for access to the specimens. RHLD’s studies of Phoridae are currently supported by grants from the Balfour-Browne Trust Fund (University of Cambridge) and the Systematics Research Fund of the Linnean Society and the Systematics Association (UK).

References

BORGMEIER, T. 1967a. Studies on Indo-Australian phorid flies, based mainly on material of the Museum of Comparative Zoology and the United States National Museum (Diptera, Phoridae). Studia Entomologica, Petropolis 9: 129-328 (1966).

BORGMEIER, T. 1967b. Studies on Indo-Australian phorid flies, based mainly on material of the Museum of Comparative Zoology and the United States National Museum. Part Il. Studia Entomologica, Petropolis 10: 81-276.

DISNEY, R.H.L. 1990. Key to Dohrniphora males (Diptera: Phoridae) of the Australasian and Oriental Regions with descriptions of new species. Zoological Journal of the Linnean Society 99: 339-387.

DISNEY, R.H.L. 2001. The preservation of small Diptera. Entomologist’s Monthly Magazine 137: 155-159.

DISNEY, R.H.L. 2003. Tasmanian Phoridae (Diptera) and some additional Australasian species. Journal of Natural History 37: 505-639.

DISNEY, R.H.L. 2008. Six new species of Megaselia Rondani (Diptera: Phoridae) from mainland Australia. Zootaxa 1899: 57-68.

DISNEY, R.H.L. 201 1a. Three new species and a new key to the Diplonevra Lioy (Diptera: Phoridae) from Australia. Zootaxa 2792: 41-50.

DISNEY, R.H.L. 201 1b. Forensic science is not a game. Pest Technology 5: 16-22.

DISNEY, R.H.L. and BANZIGER, H. 2009. Further records of scuttle flies (Diptera: Phoridae) imprisoned by Aristolochia baenzigeri (Aristolochiaceae) in Thailand. Mitteilungen der Schweizerischen Entomologischen Gesellschaft 82: 233-251.

DISNEY, R.H.L. and BARTAREAU, T. 1995. A new species of Dohrniphora (Diptera: Phoridae) associated with a stingless bee (Hymenoptera: Apidae) in Australia. Sociobiology 26: 229-239,

DISNEY, R.H.L., ERZINCLIOGLU, Y.Z., HENSHAW, D.J. de C., HOWSE, D., UNWIN, D.M., WITHERS, P. and WOODS, A. 1982. Collecting methods and the adequacy of attempted fauna surveys, with reference to the Diptera. Field Studies 5(4): 607- 621.

DISNEY, R.H.L. and KISTNER, D.H. 1997. New species and new host records of Phoridae (Diptera) associated with termites (Isoptera: Termitidae). Sociobiology 30: 1-33.

120 Australian Entomologist, 2012, 39 (3)

DISNEY, R.H.L. and KISTNER, D.H. 1999. New species of Phoridae (Diptera) associated with Termites (Isoptera: Rhinotermitidae and Termitidae) in Australia. Sociobiology 34: 35-43. GREENSLADE, P. and FARROW, R. 2008. Coringa-Herald National Nature Reserve — Identification of invertebrates collected on the 2007 invertebrate survey for The Department of the Environment, Water, Heritage and the Arts, June 2008. Attp:/Avww.environment.gov.au/ coasts/mpa/publications/pubs/coringa-herald-terrestrial-invertebrate-survey-2007.pdf KANEKO, K. and FURUKAWA, E. 1977. Studies on phorid flies (Phoridae, Diptera) in Japan. Part II. Morphological notes on larvae and pupae. Journal of the Aichi Medical University Association 5: 65-72.

LIU, G-C., CHEN L., HE, X-H. and DENG, L. 2001. Life history of Megaselia spiracularis Schmitz (Diptera: Phoridae). Journal of Shenyang University 13: 1-2.

SCHMITZ, H. 1938. Drei neue aus toten Schnecken gezuechtete japanische Phoriden. Natuurhistorisch Maandblad 27: 80-83.

Australian Entomologist, 2012, 39 (3): 121-160 121

TAXONOMY AND BIOLOGY OF SYNEMON DISCALIS STRAND AND S. PARTHENOIDES R. FELDER (LEPIDOPTERA: . CASTNIIDAE) IN SOUTH AUSTRALIA

R. GRUND! and A. STOLARSKI?

19 Parkers Rd, Torrens Park, Adelaide, SA 5062 ?PO Box 423, Tailem Bend, SA 5260

Abstract

Adults and early stages of Synemon discalis Strand and S. parthenoides R. Felder sensu lato from South Australia are illustrated and compared. Synemon discalis is shown to be monotypic, while S. parthenoides s.l. is polytypic, comprising at least three allopatric and morphologically distinct taxa consistent with regionally isolated populations, viz. S. p. parthenoides sensu stricto from Adelaide and SE South Australia to western Victoria, S. parthenoides valma subsp. n. from Yorke Peninsula and S. larissa sp. n. from Eyre Peninsula.

Introduction

Synemon discalis Strand, 1911 and S. parthenoides R. Felder, 1874 belong to a complex of morphologically similar (but not necessarily closely related) Synemon Doubleday species that commonly occur in the temperate areas of Western Australia (WA), South Australia (SA) and Victoria (Vic). The first of this complex to be described was S. sophia (White, 1841) from Albany, WA. For the next 30 years (and also recently — Edwards 1996, Douglas 2008), similar species in SA were ascribed to S. sophia until Felder (1874) proposed the new name S. parthenoides for a large, Adelaide-region population. Klug (1850) had previously illustrated this latter species but treated it as S. sophia. Tepper (1882) probably confused S. parthenoides with his S. laeta Walker, although his specimens no longer exist and thus cannot be compared. It was not until much later that Strand (1911) recognised S. discalis as a smaller, cryptic species similar to a small S. sophia in appearance, although he did not indicate a locality for it. Strand (1911) also portrayed S. parthenoides as a larger species than S. sophia, but confusingly erected S. partita Strand (a synonym of S. parthenoides: see Edwards 1996) as a new species. There are also several other similar cryptic species

occurring in WA that are believed not to extend into SA (Edwards 1996, 2006).

However, the species found in SA are still very difficult to separate because of their similar pattern; consequently Tindale (1928) placed S. parthenoides as a subspecies of S. sophia. Even later, McQuillan and Forrest (1985) used the name S. sophia for the local S. parthenoides population in SA. However, subsequent Synemon workers (Edwards 1996, 2006, Douglas 2008) asserted that S. sophia occurs only in southwestern WA and that S. parthenoides only occurs in SA and Vic. They also stated that S. discalis is a valid species, possibly occurring across all three states, although Edwards (in Douglas 2004, 2008) qualified that by stating the latter may in fact be a separate new species in WA. Partial confirmation came from Kallies et al. (2008), using

122 Australian Entomologist, 2012, 39 (3)

DNA techniques based only on the COI mitochondrial gene and limited sampling of S. discalis [n = 1: Vic], S. parthenoides [n = 3: SA and Vic]) and only six other Synemon species; their work indicated that the first two species form a clade with a sister-taxon relationship.

When two specimens of similar size of S. discalis and S. parthenoides are compared, they can be very difficult to separate. The present authors therefore undertook a study of these two species, as presently recognised in South Australia, in order to determine if there is an easy way to reliably differentiate them.

We have an underlying interest in these species, having recognised the likely presence of both throughout temperate SA during previous surveys, but our studies were impeded by a lack of authoritative literature, compounded by the local collection of Synemon in the South Australian Museum, Adelaide (SAMA) being sent to the Australian National Insect Collection in Canberra in 1993, where it is still located. We have realised that our local observations are at variance with some of those previously documented and therefore present our findings here. We have examined relevant type images, descriptions and literature and have accepted Edwards’ (1996, 2006) conclusions regarding the arrangement of the species discussed here, primarily because his initial revision included an examination of original type specimens plus material from WA, which we were unable to do.

Methodology and preliminary adult differentiation

Edwards (2006 and unpublished data) and Douglas (2004, 2008) believed that S. discalis and S. parthenoides are separable by wing morphology and size. We agree but these characters are not necessarily diagnostic and we therefore sought to reinforce this view by an examination of all characters, including early stages, host plants and, particularly, the male genitalia. It was found that it is often possible to utilise the upperside (UPS) patterns on the inner-margin half of the forewing (FW) and the tornal area of the hind wing (HW) UPS (and sometimes the underside (UNS) pattern) for a quick provisional separation of the two species.

In the FW UPS of S. parthenoides there is a broad postmedian transverse white bar with (usually) in each space a dark, horizontal flattened ovoid area devoid of white scaling. The basad side of the bar is bordered by a broad black area. In the HW UPS there is a narrow, continuous, orange-coloured link between the diffuse marginal spot in cell 1A+2A of the tornal area and the postmedian spot in cell CuA2 (Figs 1-6). On the HW UNS the macular markings are usually orange coloured but sometimes marked with white centres in the costal region of the wing.

In the FW UPS of S. discalis the broad postmedian white bar is usually

completely filled with white scaling, with no dark intracellular ovoid area, and there is only a narrow black transverse zig-zag area basad of the white

Australian Entomologist, 2012, 39 (3)

bar, while in the HW UPS tornal area there is usually no continuous orange link between the diffuse marginal spot on the inner margin and the postmedian spot in space CuA2. On the HW UNS the macular markings are usually yellowish (Figs 52-65).

Once this separation was accomplished the male genitalia were examined. The genitalia of both species were found to be of a similar simple construction to those of the Synemon collecta group found in SA (Grund 2011), but differed primarily in having a long but bent, posteriorly directed ventral valva arm (harpe or valvula) about as long as the rest of the valva (e.g. Fig. 7). The ventral bulbous extension (coecum) of the aedeagal phallobase was found to be different in S. discalis and S. parthenoides (e.g. Figs 7 and 66). A broad distributional range of male genitalia were examined, initially from areas where it was generally agreed that the species occurred (not necessarily together), such as the Southeast, Adelaide and southern Eyre Peninsula Regions. The scope was then expanded to southern Yorke Peninsula and northern Eyre Peninsula, where the species were either rare or not previously recorded.

Based on the combination of male genitalia and other morphological attributes, we were able to differentiate three distinct groups in sS. parthenoides but found no differentiation in S. discalis. In the former, there is a nominotypical group (1) occurring in the Adelaide and southeast regions of SA and also western Vic; a group (2) on Yorke Peninsula; and a group (3) on Eyre Peninsula. A fourth group likely exists on Kangaroo Island (A. Young unpublished data 2010) but unfortunately we were unable to obtain any study material of this population. The work of Kallies et al. (2008) indicated that, genetically, S. parthenoides identified from Kangaroo Island formed part of a monophyletic group from Goolwa, SA and the Big Desert, Vic.

As expected for seemingly non-dispersive species, there were minor gradational clinal changes in morphology (wing pattern and male genitalia) across the S. parthenoides groups, but sharper breaks in morphology occurred

at biogeographic boundaries such as the Spencer and St Vincent Gulfs and the Mt Lofty Ranges.

Except where a holotype is illustrated, the adult images in this paper have been digitally repaired where possible, especially the termens. Adults are often damaged and scratched by their fast flight within vegetation and from copulation rituals (particularly noticeable in S. discalis). The FW UPS white surface scaling is also quickly lost, with a resultant loss of pattern, which was not repaired. Mounted material can also quickly fade in storage, with the FW UPS black background colour turning brown (also particularly noticeable in S. discalis). Most of the material examined came from the collection of R. Grund (RG); the rest came from the collections of A. Stolarski (AS), A. Lines (AL) and the residual collection at the SAMA.

124 Australian Entomologist, 2012, 39 (3)

Systematics and biology Synemon parthenoides parthenoides R. Felder (Figs 1-31)

Nominotypical Group (1) referred to above.

Synemon parthenoides R. Felder, 1874. (Type data: p. 9, pl. LXXIX, figs. 7-8, Syntype[s] [Q], in Natural History Museum, London (BMNH); type locality Adelaide [Region] [ex G.F. Angas collection?]).

Synemon partita E. Strand, 1911. (Type data: p. 1 and also J.-A. Boisduval 1875 [1874]; image in J.-A. Boisduval [1875], pl. 11, fig. 5. Type Q ex Becker collection; type locality Australia) (Synonymized by Edwards 1996).

Material examined (Figs 1-6, 14-19). SOUTH AUSTRALIA (ADELAIDE REGION): 14, 29, Kaiser Stuhl Scrub, 2.xii.2011; 19, Mt Bold, 23.xii.2003; 23, 12, Mt Bold, 22.xii.2011; 24, Mt Bold, 23.xii.2011; 153, 19, Mt Crawford, 24.xi.2011; 44, 49, Mt Crawford, 2.xii.2011; 1g, Scott Ck, 22.xii.2011 (in RG); 2ĝ, Aldinga, 21.xi.2010; 14, 19, Cherry Gardens, 24.xi.2011; 14, Onkaparinga Gorge, 23.xi.2008; 24, Onkaparinga Gorge, 30.xi.2008 (in AL). SOUTH AUSTRALIA (SOUTHEAST); 13, Binnie, 11.xi.2010; 34, Ferries-McDonald Conservation Park (CP), 15.xi.1995; 14, 19, Gosse Hill, 12.xii.2007; 19, Messent CP, 12.xi.2006; 13, Monarto, 3.xii.2010; 14, Monarto, 10.xi.2011; 74, 62, Monarto, 19.xi.2011; 19, Mt Rescue CP, 11.xi.2008; 18, Mt Rescue CP, 13.xi.2008; 19, Malinong, 17.xi.2010 (in RG); 19, Binnie, 17.xi.2009; 13, Binnie, 4.xi2010 (in AS). VICTORIA (NORTHWEST): 14, Dimboola, 4.xii.1997; 1d, 19, Mirranatwa, Grampians, 3.xii.1997 (in RG).

Description (Figs 1-12, 14-19). Male. Body: frons, head and thorax dark brownish grey-black above, a white line along each side of anterior half of thorax above, abdomen above dark brown anteriorly, golden brown to orange laterally and posteriorly, thorax pale grey below with a narrow orange neck collar, abdomen fawn below, labial palpi ascending, pale grey scales appressed, extending beyond the eye to the edge of the frons, apical segment long, cylindrically tapering to a point, slightly shorter than mid segment, proboscis unscaled well developed, eyes smooth, reflective eye pattern pale grey Type III when alive, antennae reach to or slightly beyond half the length of forewing (FW) costa or the end of the discal cell, shaft scaled, black, narrowly ringed white at the end of each segment, club broad, mucronate, black above, white below, mostly scaled but underside with a brown nudum area. Wing morphology: background colour of wings is black, very slightly brownish when freshly emerged, but turn more brownish with age; FW UPS patterned with white scaling, easily dislodged; a broad white margin, partly scalloped in appearance with some white scaling continuing basad along veins; two white curved subapical bands, widely spaced at the costa, converging and terminating at cell space M2 to form a curved V shape, the inner band is broken by black veining and is usually much stronger than the outer band but can sometimes weaken close to the convergence point, the outer band is strongly scalloped; a large white irregular blotch straddles the

Australian Entomologist, 2012, 39 (3) 125

Figs 1-6. Synemon parthenoides parthenoides, upper and undersides: (1) male (m) wing expanse 44 mm Mt Crawford, SA 24.xi.2011; (2) (m) 42 mm Mt Crawford 24.xi.2011; (3) female (f) 48 mm Mt Crawford 2.xii.2011; (4) (m) 41 mm

Onkaparinga Gorge, SA 30.xi.2008; (5) (m) 42 mm Mt Bold, SA 22.xii.2011; (6) (f) 46 mm Mt Bold 23.xii.2003.

distal cell-end of the discal cell, a large black roughly circular area basad of the white blotch within the discal cell; a wide white scaled post median band extending from cell M3 to near the inner margin at cell CuP, the inner area of the band in cells CuAl, CuA2 and CuP partially devoid of white scaling producing a dark area, usually one or two large irregular white spots incorporated into the band adjacent to the white discal cell end spot, the apical space continuing above the white band to the costa is black and devoid of white scaling; a wide and straight black tornal bar occurs distad of the white band usually coalescing with the apical black area distad of the discal cell end spot, the tornal bar constricts towards the tornus; a wide black submedian band occurs basad of the postmedian white band, weakly

126 Australian Entomologist, 2012, 39 (3)

coalescing with the large black distal spot in the discal cell, slightly curved basad, with the cross veins sometimes white scaled; the basal portion of the wing is covered with white scaling. HW UPS with three transverse rows of orange macular spots; an outer marginal (subterminal) row of smaller irregular spots, always three spots present in cells M3, CuAl and CuA2, usually widely spaced with one spot per cell space, sometimes additional inconspicuous single spots extend towards the apex in the cell spaces M1 and/or M2, the major marginal spots are sometimes elongated basally and sometimes join with the row of larger postmedian spots, there is a larger diffuse tornal spot in cell 1A+2A essentially forming part of a broad orange inner margin area extending from the wing base, each of the two large postmedian spots straddle two cells, each spot is offset slightly such that the spot nearest the apex is further away from the wing base; the spot closer to the inner margin in cells CuA2 and CuA1 is divided by a black coloured vein and is joined to the tornal marginal spot by a narrow curved orange band, a fourth inconspicuous postmedian spot sometimes occurs in space Sc+RI next to the costa; there is a large orange spot straddling the distal end of the discal cell; the basal inner margin area next to the spots is covered in orange scaling and brown hairs (setae). FW UNS black, 10 small weakly elongated marginal spots, tornal spot 10 weak or not developed, otherwise one spot per cell, the first two apical spots white coloured, the next 2-3 become increasingly more orange, the remainder are orange, spots 9-10 in tornal cells CuP and 1A+2A are usually joined together and to the postmedian band; there are broad irregular orange coloured subapical and postmedian transverse bands, the subapical band also usually overlain by a centred wash of white in each cell, sometimes there is a weak wash of white in the postmedian cells near the discal cell, the costal margin and the basal half of the discal cell is orange scaled. HW UNS black, usually seven small marginal spots that become increasingly larger and more elongated towards the tornus, the first two at the apex often inconspicuous and white, remainder mostly orange, the postmedian and discal orange spots found on the UPS are also present on UNS, the centres usually weakly washed with white, excepting the large apical postmedian spot which can have a strong wash of white, the small costal spot of the postmedian band is white coloured if present, usually a wash of white scaling at the wing apex, the inner margin and basal area is orange coloured. Termens above and below are dark brownish grey on the FW and also much of the HW but are dark yellow in the apex, tornal and inner margin areas of the HW, and pale grey along the costa of the HW.

Female. Similar to male although the white markings are generally better defined and more intense. Antennae reach to or slightly before half the length of forewing (FW) costa or the end of the discal cell.

Wing pattern morphology of both sexes is generally stable, except for minor variations mentioned above and rare aberrations, being mainly variation in the size, shape and number of macular spots and the degree of white scaling

Australian Entomologist, 2012, 39 (3) 127

Figs 7-13. Male genitalia, lateral views. (7-11) S. p. parthenoides: (7) Mt Crawford; (8) Mt Bold; (9) Monarto, SA; (10) Binnie, SA; (11) Dimboola, Vic. (12-13) S. larissa: (12) Hincks CP, SA; (13) Pinkawillinie CP (east), SA.

128 Australian Entomologist, 2012, 39 (3)

on the FW UPS. The latter is also partially controlled by the age (wear and tear after ecdysis) of the adult that can have a significant bearing on the configuration of the white scaling. There are no obvious pale and dark morphological forms as seen in the S. collecta species group (Grund 2011).

Figs 14-19. S. p. parthenoides, upper and undersides: (14) (m) 42 mm Monarto 19.xi.2010; (15) (f) 50 mm Monarto 19.xi.2010; (16) (m) 38 mm Mt Rescue CP, SA 12.xii.2007; (17) (f) 50 mm Mt Rescue CP 11.xi.2008; (18) (m) 45 mm Mirranatwa Grampians, Vic 3.xii.1997; (19) (f) 47 mm Mirranatwa 3.xii.1997.

Wing venation. Both sexes show the basic venation typical for Synemon (Edwards et al. 1999) and similar to all species examined in this project. FW discal cell about half length of costa, vein Sc reaches costa beyond the end of discal cell, bases of veins R1, R2, R3+R4+R5 originate from the discal cell, R4 and RS stalked, bases of MI and R3+R4+R5 not connate at discal cell, origin of M3 on discal cell usually equidistant between bases of M2 and CuAl; hind wing (HW) frenulum with one spine in males or 2-3 spines (usually 2) in females, bases of M3 and CuA1 usually not connate, origin of M3 on discal cell is much nearer to CuA1 than to M2.

Australian Entomologist, 2012, 39 (3) 129

Adult forewing expanse (length of forewing along the costa from centre of thorax to apex tip x 2). This is a large species. Wing expanse of females is usually considerably larger than that of males. Based on material in the authors’ collections, males from the Mt Lofty Range have a wing expanse of 39-46 mm (avg. 42 mm, n = 31) and females 44-56 mm (avg. 49 mm, n = 10), while Southeast males are 35-45 mm (avg. 41 mm, n = 20) and females 41-53 mm (avg. 48 mm, n= 14).

Male genitalia (Figs 7-13). Male (n = 8). Tegumen broad (viewed from above), short and shallow viewed from side, sclerotised sides sit directly on top of valves, dorsal part of tegumen weakly fused with the uncus where the latter also downturns; uncus about same length as tegumen, shallow from side, edges rolled over, a slight posterior ventral bulge on each side, broad (arrow-head shaped) and tapering posteriorly viewed from above, half width of tegumen and constricted about midway along uncus, then tapering quickly to a blunt posterior point, uncus with long peripheral hairs (setae); the fultura superior is exposed in the area below the uncus and tegumen junction on each side of the genitalia and is membranous, containing a long broad horizontal chitinous scaphial plate adjacent to the valve and anal tube, anteroventral edge of plate weakly fused to posteroventral edge of tegumen; anterior part of valve broad, bulging from side view, flattened from top view, anterior sclerotised edge slightly concave posteriorly, valve tapered posteriorly to join in line with a long flattened tapering arm-like extension (harpe) of the valve that curves or bends ventrally at an angle and ends with a short upward and inward turned spine, some very long hairs posterodorsally and anteroventrally on the harpe, the bases of the former may be so dense as to cause a rough granulated bulge along the valva edge; vinculum in lateral view narrow ventrally, usually sloping away from the valva at an angle, the dorsal part next to the valve broadening considerably until the posterior margin attaches to a short narrow valva hinge at the anterodorsal corner of the valve, the anterodorsal margin of the vinculum fused with the tegumen at the apex angularis and then continuing around the tegumen edge but forming a prominent rounded anterodorsal appendage on the tegumen, the ventral part of the vinculum side arms are bent anteriorly to form a bifurcate saccus, but are joined together at the curve by a wide, flattened, sclerotized cross-brace (Fig. 10), in the centre of which a broad weakly scleritised plate emanates dorsally (as part of the diaphragm) to attach wishbone-like to the underside of the aedeagus anterior of the vinculum arms, the anteroventral arms of the valva extend to attach to the dorsolateral part of the wishbone pedicle, the whole complex forming the juxta; the aedeagus is very long, tubular, slightly curving downwards, the posterior sclerotised edge slanting at a straight angle to a point ventrally, the posterior vesica without obvious cornuti, the aedeagus enlarges considerably in the vertical plain at its anterior end to form the Synemon sclerotised phallobase with dorsal and ventral (coecum) bulbous enlargements, the proximal orifice opening is posterior.

130 Australian Entomologist, 2012, 39 (3)

When compared with the male genitalia of other S. parthenoides group. species (that have an orange join of HW UPS tornal spots 1A+2A and CuA2) from Eyre Peninsula (Figs 12-13, 51) and S. discalis (Figs 67-70), it is immediately seen that these three groups have genitalia that are very different from each other (see below for details).

Hostplants. Tindale (1928) found early stages of nominotypical S. parthenoides on Lepidosperma carphoides (Cyperaceae) at Highbury (a northeast foothill suburb of Adelaide) and provided the first biological details for a Synemon species from SA. He was the first to record Synemon larvae living underground within the root zone of its hostplant. He was unable to find living pupae but did notice pupal exuviae projecting from silken burrows at ground level adjacent to the hostplant.

The present authors (and Douglas 2008) found that the primary hostplant for nominotypical S. parthenoides is L. carphoides, meaning that this sun moth is usually found in the presence of that particular host if it is available. Douglas (2008) also recorded nominotypical S. parthenoides utilising both L. carphoides and Schoenus racemosus (Cyperaceae) as a host in the dryer, northern areas of its range in western Victoria (central Big Desert). He also saw females probing the bases of Lepidosperma viscidum nearby in southeast Big Desert, but apparently they did not oviposit. One of us (AS) has also seen a female probing small plants of Austrostipa mundula (Poaceae) in the Upper Southeast of SA. However, both of us noticed that Synemon females were not averse to ovipositor probing other plants in the vicinity of the primary host, based on visual sightings, but when these females were caught and examined it was noticed they were usually old and had no or few (possibly infertile) eggs left in their abdomens.

We also observed that the size of the L. carphoides plant for egg laying was irrelevant; females would utilise all sizes of plant, unlike many other sun moth species in SA that prefer small, stunted hostplants. They tend to prefer healthy plants in the open but will still lay on plants that are dead in the middle (similar to a Triodia spinifex ring and possibly killed by the activity of the Synemon larvae) and pupal exuviae are often found in the dead central area or along the outside of the healthy outer part of the plant.

Habitat. Adults are found flying in the vicinity of their primary hostplant Lepidosperma carphoides, a dryland sedge requiring moderate rainfall (35-80 cm pa). It grows in deep, usually white-sand soils occurring in open woodlands and sedgelands, but will also grow in higher-rainfall forest provided it is open and sunny.

Distribution and flight period. Nominotypical S. parthenoides occurs in the Adelaide-Mt Lofty Ranges and Upper Southeast Regions of South Australia (extending into western Victoria). There are no confirmed records from Eyre Peninsula or Yorke Peninsula. However, the distribution of L. carphoides

Australian Entomologist, 2012, 39 (3) 131

includes the Lower Southeast, suggesting that S. parthenoides will probably be found in that locality.

| It is sympatric with S. discalis (see later) but adults tend to start flying during

the later parts of the S. discalis flight period. (The flight period for all sun moth species documented in this paper can be instigated or delayed by the

| Climatic nature of the season and the micro-climate of the locality). There is a | tendency for adults to start flying earlier in warmer areas (and also finish

flying earlier). Males also tend to fly and be more common earlier than females in any one locality, with females first appearing about a week after the males. Along the Mt Lofty Ranges the normal recorded flight times are from 3 November to | January. In the Southeast the flight times are from 27 October to 14 December. In western Victoria, Douglas (2008) recorded flight times from late October to early January.

Figs 20-21. S. p. parthenoides, eggs and eclosed larvae. (20) egg, egg shell, eclosed larvae (4 mm), Monarto, 9.xii.2011; (21) eclosing larva with exposed spinneret, egg 2.4 mm, Binnie, 6.xii.2011.

Egg (Figs 20-21). Eggs of S. parthenoides (both laid and extracted) are similar to those of all other group members found in SA. Egg size is not necessarily related to female size, nor are they of the same size in an individual female; longer eggs tend to be narrower and vice versa. They are

132 Australian Entomologist, 2012, 39 (3)

of an elongate, ellipsoidal spindle shape, 2.05-3.95 x 0.85-1.1 mm (n = 11), with 10-13 (n = 8) prominent equi-spaced longitudinal ridges converging at

each end of the egg and with numerous (~60) less prominent, very fine cross

ridges or striae that form an interlocking disjunction at the longitudinal ridges

(e.g. Fig. 2 in Common and Edwards 1981). The higher number of longitudinal ridges seems to be proportional to an increase in size of the eggs

and females. The longitudinal ridges in this group have the peculiarity of sometimes dividing into two, a phenomenon not yet seen in the eggs of other

Synemon in SA. Each end of the egg constricts to a blunt point, one of which

(usually the sharpest) contains the micropyle and which is also the end from

which the larvae usually eclose. Pale sub-translucent yellowish-white when

freshly laid, later turning white particularly near eclosion, which occurred

after 22-32 days (n = 19). Eclosion may be dependent on moisture in the soil

enabling the egg chorion to become flexible, as one egg did not eclose until

moistened after 45 days (not recorded in incubation period). The ovipositor

of the female is typically very long and the distal end very bristly, features

that are found in all the SA group species.

Larvae (Figs 20-26). First instar larvae at eclosion (Figs 20-21) are 3.5-4.0 mm long (extended) and are similar to larvae of the other group members mentioned in this paper. All larval stages have a similar shape, of witchetty- grub type, and known larvae of other species in the group in SA are also very similar in shape. Larvae are cylindrical, slightly flattened and taper posteriorly, with the posterior end rounded. Moderately long, fine, simple sensory setae are common at either end, but few laterally and elsewhere, (no attempt was made to produce a setal map). The mid-portion of each segment is enlarged; thoracic segments (TS) 1-3 are larger than abdominal segments (AbS), but AbS 3-6 are also larger than other AbS. The prothoracic plate on TS | is much enlarged, tending to overlap onto TS2 and smooth, presumably to help with burrowing. Roughened, elliptical-shaped ridges are present dorsally on the other segments, again presumably to help with compacting the burrow. Thoracic legs and abdominal prolegs are present but not fully functional and of little use for directional travel, although first instar larvae were able to gain traction and sometimes walk up the vertical sides of a glass jar (probably helped by moisture). Skin and head are smooth and shiny and the body subtranslucent. The colour of the first instar at eclosion is pale yellow, white posteriorly, prothoracic plate brownish yellow, head pale brown, paler dorsally, mandibles black. Larvae remain underground all their lives and, if exposed to light, will quickly burrow back into the ground.

Second instar larvae (12 mm, Fig. 22) are white (subtranslucent), the prothoracic plate off-white, brownish next to the brown head, end of posterior segment brown, sometimes a brown area dorsally midway along abdomen, stomach contents black; from about the third instar they start to show areas of pinkish colouration on the skin and on the yellowish prothoracic plate there is a pair of mid-dorsal, orange-brown frontal triangular marks next to the head,

Australian Entomologist, 2012, 39 (3) 133

26

Figs 22-26. S. p. parthenoides larvae: (22-25) larvae in captivity ex L. carphoides Monarto. (22) pre-moult late 2™ instar (12 mm), 28.x.2011; (23) late 5" instar (30 mm) 24.x.2011; (24-25) late 5" instar dormant (30 mm) dorsal and lateral 24.x.2011. (26) larva on L. carphoides Mt Crawford, close-up of anterodorsal portion of 5" instar (30 mm) 23.x.2011, showing head, prothoracic plate, dorsal elliptical ridges, setae.

134 Australian Entomologist, 2012, 39 (3)

divided by a yellowish longitudinal line; the anal plate is brown, the head dark brown anteriorly, pale brown posteriorly and divided centrally by a dark brown, triangular area basally emanating from the anterior dark area. The mature fourth and final instars become increasingly darker pink (Fig. 23, 26) then red, then finally dark orange-red near pre-pupation (Figs 24-25). Earlier instars occur underground in the culm, while the mature larvae are mostly seen in the root zone.

Under adverse conditions, especially when placed together in captivity, larvae will cannibalise each other if the hostplant does not remain alive in adequate quantities. Under such conditions, larvae will live off their own fat and shrink (at least by half) until such time as a food source is generated. When small and large larvae meet, the normal first response of the smaller larvae is to try and escape but they sometimes disgorge their stomach contents and become a lot smaller, presumably as a deterrent to the larger larvae. In captivity, a disgorge response by the smaller larvae can be fatal. Larvae will not eat dead hostplant tissue. Based on the size of larvae observed, at various times over the year on their hostplant and in captivity, we believe that larvae have the growth potential to reach prepupal maturity within two years. One mature larva in captivity has already been living in a semi-torpid condition for a further two years, having ignored two potential pupating events, suggesting they require exacting conditions before pupation.

Larval predators. The only possible insect predatory activity we saw was occasional large beetle larvae found in the culm and root zone of the hostplant; these might be predatory on Synemon larvae since, when such beetle larvae are themselves put together, they will cannibalise each other, There were sometimes small bandicoot or echidna-like diggings at the sides of the L. carphoides hostplants, which might have led to predation on Synemon larvae. In strong colonies, none of these possible predators appeared to be in sufficient numbers to have had any threatening impact on Synemon larvae.

Pupae (Figs 27-31). We were unable to find living pupae, but RG was eventually able to find some exuviae protruding out of silked prepupal tunnels (essentially cocoons) in their ecdysis position. The latter were found in several colonies within the Adelaide Hills and were seen either within the dead central area of a living tussock of L. carphoides (see above), or adjacent to a living tussock up to 42 cm away. Up to four exuviae were found together in the former situation and up to three together were found in the latter; presumably all exuviae seen in any situation were from that flight season (considering the prolific animal, bird and insect life in the areas at the time which would have soon obliterated the exposed exuviae). Only male exuviae were observed (Figs 27-31), which have typical Synemon morphology (similar to the male pupa of S. magnifica illustrated in Common and Edwards 1981), with two rows of dorsolateral flattened spines (similar to a pointed

Australian Entomologist, 2012, 39 (3)

Figs 27-31. S. p. parthenoides, pupa exuvia ex L. carphoides. (27-30) (m) pupa exuvia (23 mm) 24.xi.2011 Mt Crawford; (31) pupa exuvia protruding from dorsal end of pre-pupa silked tunnel ‘cocoon’, (7 cm) plus exuvia, Monarto 10.xi.2011.

spade) on AbS 2-7 and with the anterior row comprising much larger spines. The spines on AbS 2 are not well developed and only a single row of (large) spines occurs on AbS 8-9. Contrary to previously published observations, only short, silk-lined pre-pupal tunnels were observed (Fig. 31); these were about 6-7 cm x 8-11 mm in size and near-vertical in the (sand) soil below the surface (but reaching the surface). The lower end of the tunnel was sealed off with silk and presumably the top part was also, but this was not seen at the time in a situation either before or after adult ecdysis (but a ‘lid’ was detected by Tindale 1928 on a similar 6 cm silk tunnel at Highbury); the entire silked structure would by definition be called a cocoon. The prepupal skin was

136 Australian Entomologist, 2012, 39 (3)

present at the bottom of the sealed tunnel, while the exuvia occurred halfway out of the top end (Fig. 31). The rest of the original tunnel presumably made underground by the prepupal larva back to the hostplant (as reported by Tindale 1928) was not silked and could not be discerned.

The extracted exuviae were about 23-26 mm long, equating to about 19-22 mm actual pupal length (allowing for the abdominal expansion during ecdysis). The antennae are not fused to the thorax or wings. We could find no difference in pupal morphology between S. parthenoides and S. discalis (Figs 79-82), except for some minute detail posteroventrally, which requires further confirmation. Again contrary to previous studies, the nature of the coarse, posteriorly directed, flattened spines on the abdomen of the pupa suggests that movement in only one direction would be possible for a living pupa inside a tight silk tunnel, that being upwards and out of the tunnel, presumably at the time of ecdysis. There is no cremaster to impede movement.

Adult biology. Typically, adult males tend to stay close to the hostplants, preferring open spaces and either flying about the plants or by basking or patrolling over clear ground, car tracks or plant debris nearby. The flight is less rapid than in S. discalis, perhaps attributable to their larger size. They usually fly just above the hostplants but at times will fly higher, particularly in wooded areas with a higher understorey. They are not known to seek out hill or dune tops to patrol but will utilise them if their host is nearby. While in flight, males can detect females on the ground from a few metres away and immediately divert to where the pheromones are coming from. When disturbed both sexes fly rapidly, resembling a skipper in flight, generally flying up to 50 metres (usually much shorter) in one direction before settling. They fly in full sun, preferring temperatures above 18C, although in hot conditions they will fly with some high cloud present. Adults become active around 0930 h (DST), typically nectaring or basking on the ground to begin with, but increasing in activity with time. By midday there is maximum activity, which continues to about 1400 h.

In the afternoon females tend to fly just above hostplant height in search of suitable food plants, seemingly sensing the presence of hostplants while in’ flight by a combination of sight and olfaction. Once selected, females typically land on the ground close to the hostplant then walk to the base of the plant to test it, usually by flitting up onto the leaf stalk near ground level, then backing down to the ground to start probing the edges of the stalks below ground level to lay a single egg. Sometimes she will first land on the higher outer part of the plant then work her way down to the base, either through the plant (usually impossible) or flitting lower to an outer part of the plant. During this pre-oviposition stage the wings are regularly opened and closed. When laying is completed, the female usually moves on and repeats the process on another nearby plant, but sometimes return to the same plant

Australian Entomologist, 2012, 39 (3) 137

or will leave the area. The time taken to lay an egg can be short (~30 secs) or can take one or two minutes depending on how experienced she is ‘or how accessible the oviposition site is. Activity tends to decline after about 1400 h, but depends on adult numbers and ambient temperature. On warm days, some activity may continue to about 1700 h, including egg laying, but most active

males are by then sitting on the ground. We did not determine where they roost at night.

We have seen adults nectaring only rarely. RG observed nectaring in SA on Calytrix tetragona. In west Victoria, Douglas (2008) observed nectaring on Kunzea pomifera, Calytrix tetragona and Eucalyptus costata. AS observed nectaring on Leptospermum sp. in central Victoria, where the adult flapped its wings slowly as it moved from flower to flower.

Comments. Synemon parthenoides adults, when in good condition, clearly differ from those of other group members in their collective wing and male genitalia morphology and other biological attributes, as documented above and elsewhere in this paper. The distribution of the nominotypical group of S. parthenoides was found to continue eastward from the Adelaide Region into Southeast SA and further into central Victoria (CSIRO 2012) (Fig. 37). The wing pattern of eastern material (Southeast SA and Victorian Regions) is very similar to that of nominotypical material from the Adelaide Region, differing mainly in the white markings being more suppressed in males (Figs 1-2, 4-5, 14, 16, 18). The male genitalia (Figs 7-13) are also very similar, differing mainly in the amount of bending in the ‘harpe’, which tends to be more exaggerated in eastern specimens.

Synemon parthenoides valma subsp. n. (Valma’s Sun moth) (Figs 32-36) Yorke Peninsula Group (2) referred to above.

Types. Holotype 3, 43 mm, SOUTH AUSTRALIA (YORKE PENINSULA): Hardwicke Bay, 5.xi.2011, R. Grund (in SAMA). Paratypes (Figs 15-16): 98, 59, Hardwicke Bay, 5.xi.2011, R. Grund; 1d, orange form, Coonarie, 17.xi.1999, R. Grund (in RG). :

Description (Figs 32-35). As for S. p. parthenoides from the Adelaide Region except as follows. Male (Figs 32-33): FW UPS white markings more strongly developed and the submedian dark area has a distinct ‘three-leaf clover’ configuration. The ‘orange’ markings of the HW UPS and the FW and HW UNS are distinctly yellow in S. p. valma and are further accentuated by a white suffusion of variable intensity. The HW marginal spot overlying vein CuA2 is sometimes distinctly divided by the black scaling of the vein. The male paratypes include one worn specimen from Coonarie (Fig. 35) that has orange markings and the white suffusion was more suppressed compared to specimens from Hardwicke Bay, although the FW ‘clover-leaf submedian pattern was present.

138 Australian Entomologist, 2012, 39 (3)

Figs 32-35. Synemon parthenoides valma subsp. n., upper and undersides: (32-33) holotype (m) 43 mm Hardwicke Bay, SA 5.xi.2011; (34) paratype (f) 50 mm Hardwicke Bay 5.xi.2011; (35) paratype (orange form) (m) 40 mm Coonarie, SA 17.xi.1999,

Female (Fig. 34). Similar to male but the white markings above and white suffusion below are significantly more obvious and distinct.

Adult forewing expanse. Males from Hardwicke Bay have a wing expanse of 39-44 mm (avg. 41 mm, n = 11) and females 46-52 mm (avg. 49 mm, n = 5). The single male from Coonarie has a wing expanse of 42 mm.

Male genitalia (Fig. 36, n = 2). Genitalia of the yellow morphs from Hardwicke Bay are very similar to those of S. p. parthenoides. Differences noted include: the base of the harpe (where it attaches to the rest of the valva) is noticeably constricted, although a similar constriction is seen in the male genitalia from Binnie (Fig. 10); the harpe is bent rather than gradually curved.; the uncus is only very weakly fused to the tegumen, but the scaphial plate is more strongly fused basad to the tegumen; the ventral coecum elongation of the phallobase is better developed and easily reaching down to the base vinculum sclerotised cross-brace bridge (bifurcate saccus). The vinculum side arms are attached to the tegumen at two points on the apex angularis (clearly seen in Fig. 36), by two narrow pedicles emanating from the posterior and anterior edges of the vinculum; the juxta development between the aedeagus and base vinculum brace is better developed and stronger, where the juxta is more scleritised and forms a posteriorly bent wishbone-like structure (similar to a flattened spring-like vertical prop or strut once used under the seats of farmers’ tractors). The juxta appears to be in a more advanced state than in S. p. parthenoides.

Australian Entomologist, 2012, 39 (3) 139

36

Fig. 36. Male genitalia, S. p. valma lateral view, Hardwicke Bay.

Etymology. Named in honour of the late Yorke Peninsula volunteer Valma Stone, for humanity, ecology and wildlife work.

Hostplants. Lepidosperma carphoides does not exist on Yorke Peninsula. Females were observed ovipositing on L. congestum, which was common at Hardwicke Bay and is reported to be common throughout Yorke Peninsula by the State Herbarium of SA. The host for the orange form at Coonarie was not determined.

Habitat. The type locality at Hardwicke Bay is partially cleared coastal white dunes, with very open low mallee and coastal salt-tolerant type vegetation. At Coonarie the vegetation was low mallee, growing on red loam over limestone in a hill-top situation where the orange form was flying with S. discalis.

Distribution and flight period. S. p. valma is known from Hardwicke Bay and Coonarie (Fig. 37) and likely exists further west of Hardwicke Bay to Marion Bay in relict native vegetation, where author RG has previously seen single flying specimens (not examined) of either it and/or S. discalis. Tepper (1882) similarly reported S. parthenoides (as S. laeta Walker) occurring at Ardrossan although his specimens no longer exist for authentication. A specimen exists at SAMA (currently at ANIC) captured by N. B. Tindale at Moonta (CSIRO 2012). At Hardwicke Bay, this subspecies was common in early November. At Coonarie a few were flying in mid November.

Egg. Eggs (n = 2, infertile) were extracted from the ovipositor of two separate females and are very similar to others of the complex, having 14-15 longitudinal ridges (including bifurcation as for eggs of S. p. parthenoides), 2.35-3.1 x 0.9-1.05 mm. Pale subtranslucent yellowish white when fresh.

Larvae and pupae. Not observed. Larval predators. While examining the hostplants for early stages, a very

large dune scorpion was found in a tunnel into the root zone; presumably it would eat any Synemon it found.

140 Australian Entomologist, 2012, 39 (3)

2 8

22

ays &

Sy ey a

nn

oi G

= Ss s 8 2

2ge2 225

BOR EOE a

BSG G6 GX D

E E EE

37

r= 5 © = 2 8 ag of 68 2's = 2 @ 2 Sa EE mie a oLf ZO

38

Figs 37-38. Distribution maps for SA [and west Vic.]: (37) S. larissa and S. parthenoides subspecies and primary hostplant L. carphoides; (38) S. discalis and primary hostplant G. lanigera.

| Australian Entomologist, 2012, 39 (3) 14]

Adult biology. The Hardwicke Bay population was examined by RG on 5 November 2011 with temperatures reaching 34°C. Adults were already flying by 1000 h, mostly old male specimens either patrolling and sunning themselves on dune tops or flying around hostplants lower down in the inter- swale areas. By about 1130 h newly eclosed adults of both sexes were more frequent and began copulation. One newly eclosed female flew only a short distance before being chased by a newly emerged male, landing on a low plant then turning upright, the male landing below her and quickly walking to her left side before touching her abdomen, then quickly moving to her right side, facing in the same upright direction as the female and immediately commencing copulation. Soon afterwards another male arrived and terminated the copulation by flying onto the female, causing the original

couple to fly off for a short distance before they again landed and copulation resumed.

Flight activity ceased between 1300-1400 h, after which a few older females began ovipositing. One landed high up on an upright leaf at the edge of the hostplant with her head downwards (resembling a flower head), then walked downwards to near ground level, turned upright, then backed down to ground level before probing deep into the sand with her ovipositor along the edge of the leaf, all while continually opening and closing her wings. This probing activity was repeated a few times on this and other leaves in the clump before she flew away. Some females landed next to a black, congested flower head near the top of the hostplant, where they cryptically blended in with the flower head, often remaining there for 20 minutes or more. A few adults were still flying at 1440 h when the author left the area.

Comments. We believe the differences in both morphology and biology between S. p. valma and S. p. parthenoides are sufficient to warrant its erection as a subspecies. There is a break in the distribution of the primary hostplants, L. carphoides and L. congestum, between Gawler and north Yorke Peninsula and, in combination with the presence of the St Vincent and Spencer Gulfs, these features likely act as barriers to dispersal, creating a distinct morphological group on Yorke Peninsula consistent with a regionally isolated population, possibly the result of Pleistocene climate cycling as suggested for other Australian Lepidoptera such as the genus Theclinesthes Röber (Lycaenidae) (Rod Eastwood unpublished data 2006).

Some sun moths are renowned for their poor dispersal abilities (Douglas 2008) and S. parthenoides, being a large, heavy species is likely to be one such moth. Subspecies S. p. valma is allopatric with other S. parthenoides- like sun moths (orange linkage of HW UPS tornal spots 1A+2A and CuA2) and has a distinctive morphology, yet has a similar pattern and male genitalia to the latter; although it does not use the same hostplant it does utilise sedge plants comparable to those used by S. p. parthenoides, which is also the neighbouring taxon. Its wing colours may be influenced by a variation in

142 Australian Entomologist, 2012, 39 (3) |

flavonoid pigments sequestered from its local host plant (such as occurs in the skipper Hesperilla flavescens Waterhouse: Hesperiidae). The isolation of — S. p. valma, use of a different hostplant, unique wing pattern and minor changes in male genitalia support its recognition as a subspecies.

Synemon larissa sp. n. (Larissa’s Sun moth) (Figs 39-50)

Eyre Peninsula Group (3) referred to above.

Figs 39-44. S. larissa sp. n., upper and undersides: paratypes Hincks CP, SA 3.xi.2011 (39) (m) 38 mm, (40) (f) 47 mm; Heggaton east, SA 4.xi.2011, (41-42) holotype (m) 39 mm; paratypes (43) (m) 42 mm, (44) (f) 47 mm.

Types. Holotype 3, 39 mm, SOUTH AUSTRALIA (EYRE PENINSULA) (Figs 39- 40): Heggaton east, 4.xi.2011, R. Grund (in SAMA). Paratypes (Figs 41-49): 13, | Heggaton east, 4.xi.2005, R. Grund; 84, 59, Heggaton east, 4.xi.2011, R. Grund; 29, © Heggaton west, 2.xi.1998, R. Grund; 13, 19, Hincks CP, 3.xi.1998, R. Grund; 31, 139, Hincks CP, 3.xi.2011, R. Grund; 44, Hincks CP, 4.xi.2011, R. Grund; 33, Pinkawillinie CP (east), 13.x.1998, R. Grund; 33, Pinkawillinie CP (east), 10.x.2011, — R. Grund; 29, Pinkawillinie CP (east), 10.x.2011, R. Grund; 23, 19, Corrobinnie, 22.x.1998, R. Grund; 14, Kalanbi, 6.x.2011, R. Grund (in RG).

Australian Entomologist, 2012, 39 (3) 143

Figs 45-49. S. larissa sp. n., paratypes, upper and undersides: Pinkawillinie CP (east), SA (45) (m) 38 mm 13.x.1998, (46) (f) 48 mm, 10.x.2011; Corrobinnie, SA 22.x.1998 (47-48), (m) 38 mm, (f) 43 mm; (49) (m) 40 mm Kalanbi, SA 6.x.2011.

Description (Figs 39-49). As for S. p. parthenoides from the Adelaide Region except as follows. Male. Thorax lacking a pair of white lines and below with only a very weak orange neck collar; antennal club with nudum black; FW UPS white subapical markings weakly developed and discal cell-end white mark tending to have an apically directed point; FW UPS white spots on the postmedian white band next to the discal cell edge seen in S. parthenoides usually not developed; submedian dark area in males usually with a scattering of white scales causing a dusky appearance; orange UNS markings tend to be slightly smaller, creating an overall darker aspect than in S. parthenoides; FW UNS tornal marginal spots tend to be weakly developed or absent; HW UNS large orange markings next to the costa always with an extensive white area, this feature is almost diagnostic within the SA species but a weaker version present in S. discalis, HW UNS postmedian spot next to the inner margin tending to be smaller and divided by a black vein or space creating two separate spots. The three large HW marginal spots in the tornal region of the wing tend to be more like those in S. discalis, with the first two (next to the

144 Australian Entomologist, 2012, 39 (3)

tornus) being block-like and square-sided, while the third spot in cell M3 is elongated.

Female. Similar to male, except the white subapical and discal spots are larger and better developed. The FW UPS white spots on the postmedian white band next to the discal cell edge seen in S. parthenoides are sometimes weakly developed.

Wing venation. Similar to other SA species in the group except the origin base of M3 in the HW is unstable, ranging from being closer to M2 or closer to CuAl or connate with CuA1 (all on the discal cell), to being stalked on CuAl.

Adult forewing expanse. The size of S. larissa is quite variable, with some smaller specimens approaching S. discalis in size, yet some females are almost as large as those of female S. parthenoides. Females tend to be significantly larger than males, compared with the other group species where the size difference is less noticeable. Males have a wing expanse of 34-42 mm (avg. 38 mm, n = 54) and females 46-52 mm (avg. 47 mm, n = 24).

50

Fig. 50. Male genitalia, S. /arissa lateral view, Heggaton east.

1

Male genitalia (Figs 12-13, 50; n = 11). Similar in appearance to those of S. p. parthenoides but with some significant differences. The overall size of the genitalia tends to be relatively smaller due to their more compact construction. The tegumen-uncus-scaphial plate complex is similar but tends to be more robust; the posterior lateral edges of the tegumen are bulging; the anterior ventral edges of the scaphial plate are broadly fused to the tegumen. The ‘harpe’ is smaller and more compact, the anterior half broader, while the posterior bent half is shorter than in S. parthenoides and the dorsal edge is

) Australian Entomologist, 2012, 39 (3) 145

‘weakly upturned rather than down-turned. The anterior dorsal edge of the harpe is bulging and roughened due to granulation of the setal bases. The . ventral edge of the valva is strongly convex or bulging, the anteroventral arm ‚of the valva extends anteriorly to very weakly join dorsally with the juxta wishbone prop. The vinculum in lateral view is wide and shortened and has a vertical or squared aspect relative to the valva (compared with S. parthenoides, where it is narrow and slopes away anteriorly) before sharply l bending anteriorly at the base to form the combined bifurcate saccus and l vinculum cross-brace (as found in S. parthenoides). The juxta is similar to ! that of S. p. valma but is more robust and the attachment point on the aedeagus is in line with the vinculum arms; the vinculum arms next to the , lower part of the valvae gradually widen dorsally but expand significantly at . its join with the tegumen just dorsal of the apex angularis (the vinculum is ` narrow in its basal half in S. parthenoides before suddenly widening . dorsally), where it then becomes very narrow as it fringes the anterior side of the tegumen and also producing a rounded anterodorsal appendage on the tegumen (similar to S. parthenoides but half the size). The aedeagus is relatively shorter, more curved and slightly thicker; the ventral enlargement of the phallobase coecum is long as in S. p. valma; the dorsal enlargement of the phallobase is present but sometimes weak.

Etymology. Named in honour of a benefactor of this project.

Hostplants. One of us (RG) saw females ovipositing in the manner typical for the group on Lepidosperma carphoides in the Hincks and Heggaton areas of Eyre Peninsula. However, L. carphoides only occurs in southern Eyre Peninsula to as far north as Heggaton in northeast Eyre Peninsula; it does not occur in northwest Eyre Peninsula. In the above areas and in other areas where S. larissa flies in the absence of L. carphoides, females were attracted to the sedges L. congestum and Schoenus racemosus, which are likely hostplants although this could not be confirmed. No eggs were seen to be laid and no larvae or pupal exuviae were seen on or near the latter plants.

Habitat. Synemon larissa occurs primarily in mallee habitat, both open and closed.

Distribution and flight period. This species has only been seen on Eyre Peninsula (Fig. 37), occurring in mallee country as far north as the dog fence to the north of Ceduna. It has yet to be recorded from the extreme southern parts of Eyre Peninsula and was not looked for in the Port Lincoln area by the authors, but it is highly likely to occur in that area due to the presence of a primary hostplant L. carphoides. Its northern range is likely to be limited by the presence of its probable hostplants L. congestum and S. racemosus, being about its present known limits.

In northern Eyre Peninsula, males (n = 8) were recorded flying during 6-22 October, females (n = 3) 10-22 October. In central Eyre Peninsula, males (n =

146 ” Australian Entomologist, 2012, 39 (3)

46) were recorded flying during 3-4 November, females (n = 14) 3 November. These sparse observations imply that S. larissa starts flying earlier in the northern parts of its range and that males also start flying earlier in the season than females. It is sympatric with S. discalis and typically tends to fly later than the peak flight period of S. discalis in any one season.

Egg (Fig. 51). Eggs are very similar to others of the group, having 8-11 longitudinal ridges (n = 27), ~50 cross striae, 2.1-2.9 x 0.85-1.0 mm, (n = 29, both laid and extracted). The longitudinal ridges are sometimes divided. Pale subtranslucent yellowish white when freshly laid but white at eclosion, which occurred after 19-32 days (n = 19).

Fig. 51. S. larissa, egg shells and eclosed larvae 3.0-4.5 mm, Hincks CP 9.xii.2011.

Larvae (Fig. 51). First instar larvae at eclosion are 3.0-4.5 mm long (extended) and typically are similar to larvae of other species of the group. Older larvae were not found.

Pupae. Despite the large number of adults seen flying, no pupae or exuviae were observed.

Adult biology. This sun moth can be very common locally. In Hincks CP, RG saw potentially hundreds of males and females flying together in a small area along a track, presumably the result of a joint mass ecdysis. The numbers persisted further along the track; they were so huge that females were unable to oviposit because as soon as they stopped flying they were pounced on by the males. The track acted as a flight path for males, which continuously patrolled the area for females that ventured in to either oviposit or visit a nectar source. On a day that reached 35°C, both sexes (mostly worn) were active by 0800 h, initially sunning themselves in the open, but by 0820 h they were common and actively nectaring, doing so for most of the morning as

Australian Entomologist, 2012, 39 (3) 147

numbers continued to increase. When adults in the open were disturbed they did not fly very far, usually less than 30 m. Adults were still nectaring on flowers at 1000 h but some females were probing L. carphoides and others were seen investigating S. racemosus. Examination of adjacent native vegetation produced only the occasional female looking for hostplants. By midday adults were very active. Some females started nectaring again by 1400 h while the males patrolled. By 1500 h both sexes were nectaring and

by 1600 h they began to disappear or bask on the ground. By 1700 h they had mostly disappeared.

In the Hincks and Heggaton areas both sexes spent a lot of time in the early morning and late afternoon nectaring from flowers; they showed a preference for Calytrix sp and white-flowered Homoranthus wilhelmii, but a yellow- flowered Glischrocaryon sp (Golden Pennant) was sometimes used. Elsewhere, nectaring was not obvious. At Heggaton, a few males were seen

patrolling a large dune top during the midday heat and a few females in oviposition mode appeared interested in both L. carphoides and S. racemosus. A large population of these sun moths were later seen in a gully at 1500 h, the males flying near the hostplants and the females still attempting oviposition on L. carphoides. Activity diminished by 1630 h, with many flying off to roadside flowers for nectar.

Comments. This cryptic species has morphological features of both S. discalis and S. parthenoides, especially with the S. p .parthenoides population east of the Adelaide Region (and includes the orange linkage of HW UPS tornal spots 1A+2A and CuA2). Even though it is allopatric with respect to S. parthenoides, we believe this taxon should be treated as a new species, for reasons similar to those discussed above for S. p. valma. It has both distinctive wing pattern features and male genitalia. Inhibition of dispersal by the Spencer Gulf in the east and the aridity of the far northern Eyre Peninsula and the Nullarbor Plain presumably maintain its geographical isolation. We are unsure whether it could be a now stable species of hybrid origin or was historically derived from Western Australia.

Synemon discalis Strand (Figs 52-82) Synemon discalis E. Strand, 1911. (Type data: p. 2, Castniidae, pl. 9. Holotype 3 in

Zoological Museum, Berlin (ZMB), 26 mm, type locality Australia). Precise type locality not stated, but inferred to be South Australia (Douglas 2004).

Material examined (Figs 52-65). SOUTH AUSTRALIA (SOUTHEAST): 43, 19, Binnie, 11.xi.2010; 19, Binnie, 16.xi.2010; 19, Ferries McDonald CP, 19.xi.2010; 53, 19, Malinong, 8.xi.2010; 1g, Malinong, 11.xi.2010 (in RG); 192, Binnie, 17.xi.2009; 14, Binnie, 4.xi.2010; 14, Binnie, 1.xi.2010; 19, Malinong, 6.xi.2009; 12, Malinong, 9.xi.2009; 14, Malinong, 6.xi.2010; 4¢, Malinong, 12.xi.2010 (in AS). SOUTH AUSTRALIA (YORKE PENINSULA): 13, Coonarie, 17.xi.1999 (in RG). SOUTH AUSTRALIA (EYRE PENINSULA): 44, 39, Hincks CP, 6.x.1998;

148 Australian Entomologist, 2012, 39 (3)

33, 12, Hincks CP, 5.xi.2005; 83, 39, Inila, 8.x.2011; 14, Pinkawillinie CP (east), 13.x.1998; 19, Pinkawillinie CP (east), 1.x.2011; 19, Pinkawillinie CP (east), 10.x.2011 (in RG).

Figs 52-57. S. discalis, upper and undersides: Southeast Region SA. Binnie (52) (m) 32 mm 11.xi.2010, (53) (f£) 40 mm 16.xi.2010; Malinong (54) (m) 32 mm, 12.xi.2010, (55) (m) 32 mm 11.xi.2010; (56) (f) 38 mm Ferries-McDonald CP 19.xi.2010; Yorke Peninsula SA. (57) (m) 32 mm Coonarie 17.xi.1999.

Comparative description with S. parthenoides and S. larissa (Figs 52-65). This cryptic species has a similar wing pattern to both S. parthenoides (SP) and S. larissa (SL) but differs as follows. Male. Body: frons, head and thorax dark grey above, indistinctly speckled pale grey, white lateral thoracic line absent (present in SP but not SL); abdomen above dark golden to orange brown, thorax pale grey below, orange neck collar absent (present in SP and SL), abdomen fawn below; labial palpi pale grey ascending with appressed scaling (similar to SP and SL), apical segment ~3/4 length of mid segment (similar to SP and SL); proboscis unscaled and well developed, eyes smooth, reflective eye pattern pale grey Type III when alive; antennae reach to or just

Australian Entomologist, 2012, 39 (3) 149

beyond half the length of FW costa or the end of the discal cell, similar to SP and SL except nudum area very dark brown (brown in SP, black in SL). Wing morphology: background colour of wings dark brown-black when freshly emerged, becoming paler with age; FW UPS patterned with white scaling, easily dislodged; a broad white margin (subterminal), partly scalloped in appearance with some white scaling continuing basad along veins; two white curved subapical bands, widely spaced at the costa, converging and terminating at cell M2 to form a curved V shape, the inner band much stronger and clearer near the costa than the outer band but weakening close to the convergence point, the outer band strongly scalloped; usually a poorly developed white spot at the distal end of the discal cell, a large black area basad of the white spot within the discal cell; a wide white scaled postmedian band extending from cell M3 to near the inner margin at cell CuP, the inner area of the band in cells CuAl, CuA2 and CuP with weaker scaling (but not fully black as can occur in SP and sometimes SL), the basad edge of the band at vein CuP sharply extended basad, the apical space between the white postmedian band and the costa black and devoid of white scaling; a narrow and usually straight black tornal bar distad of the white band, usually coalescing with the apical black area distad of the discal cell end spot, tornal bar constricted towards tornus; a narrow black submedian band basad of the postmedian white band, strongly angulate basad at vein CuP to produce a narrow zig-zag appearance to the submedian band that is diagnostic for S. discalis when not damaged (absent in SP and SL), coalescing with the large black distal spot in the discal cell; basal portion of wing covered with white scaling. HW UPS similar to SP and SL except macular spots yellowish orange in S. discalis (usually orange in SP and SL, ignoring S. p. valma), the tornal marginal spot in cell 1A+2A usually not joined to the postmedian spot in cell CuA2 by a narrow ‘orange’ band (almost diagnostic for S. discalis, whereas these spots are usually joined by an orange band in SP and SL except when aberrant), the three marginal spots in cells CuA2, CuAl, M3 tending quadrangular and closer together than in SP and SL, spot M3 (especially on UNS) and tending quadrangular while spot M3 is usually elongated basad (especially on UNS) whereas in SP the three spots are of similar size and of irregular shape and spaced further apart than in S. discalis, in SL the spots are similar to the latter but are smaller and spaced apart as in SP, the fourth inconspicuous postmedian spot sometimes occurring in space Sc+R1 next to the costa in SP is not present in S. discalis or SL on the UPS (but is on rare occasions seen on the UNS of females of those two species and is white coloured if present); the UNS ‘orange’ markings are yellowish as on UPS and tend to be similarly placed as in SP and SL, but are slightly larger than in SL and the FW UNS tornal marginal spots in SL differ by being weakly developed or absent; the FW UNS postmedian black bar often tending parallel-sided apically (a feature notably remarked upon by Strand 1911), particularly in Southeast and near Adelaide specimens, whereas in SP and SL it is usually constricted posterior of vein M2; HW UNS ‘orange’ markings

150 Australian Entomologist, 2012, 39 (3)

well developed as in SP whereas in SL the markings are slightly smaller, the ‘orange’ markings in S. discalis UNS have a white wash, particularly on the HW, this wash is also present in SP and SL to varying minor extents, except in SL the large spots next to the costa always have an extensive white area; the termens are similar to SP and SL.

Figs 58-65. S. discalis, upper and undersides: Eyre Peninsula SA. Murray Point, Port Lincoln SA 4.xi.1997, (58) (m), (59) (f); Hincks CP, (60) (m) 35 mm 5.xi.2005, (61) (f) 37 mm 6.x.1998; (62) (m) 33 mm 6.x.1998; (63) Pinkawillinie CP (east) (f) 38 mm 1.x.2011; Inila, SA 8.x.2011, (64) (m) 34 mm, (65) (£) 44 mm.

Australian Entomologist, 2012, 39 (3) 151

Female. Similar to males although the white markings are generally better defined and more intense. The UNS yellowish orange markings are more

yellowish; the antennae reach to or just before half the length of FW costa or the end of the discal cell.

There are no obvious pale and dark morphological forms as seen in the S. collecta species group (Grund 2011). There seems to be some tendency for the FW of S. discalis, especially in the Southeast population, to be slightly narrower than for SP (Douglas 2008) and SL, but the data were not consistent.

Wing venation. Sexes similar. FW discal cell about half length of costa, vein Sc reaches costa beyond the end of discal cell, bases of veins R1, R2, R3+R4+R5 originate from the discal cell, R4 and R5 stalked, bases of M1 and R3+R4+R5 connate or nearly connate at discal cell, origin of M3 on discal cell variable but usually nearer base of M2 than CuA1; hindwing (HW) frenulum with one spine in male or 2-3 spines (usually two) in female, origin

of M3 on discal cell variable but usually either connate or nearer base of CuA1 than M2.

Adult forewing expanse. This is the smallest species of the group. Males are easily separated from the others by their size, although females can be of similar size to males of the other species and can be misidentified unless the differences noted above are used. Based on material in the authors’ collections, specimens from Eyre Peninsula tend to be slightly larger than those from the Southeast. The former males have a wing expanse of 31-36 mm (avg. 34 mm, n = 15) and females 37-44 mm (avg. 41 mm, n = 9), while Southeast males are 28-35 mm (avg. 32 mm, n = 17) and females 28-40 mm (avg. 38 mm, n = 7).

Male genitalia (Figs 66-70; n = 12). Closer in appearance to those of S. larissa (SL) than S. parthenoides (SP). The tegumen, uncus and scaphial plate complex are typical of the group. The ‘harpe’ is similar to that of SL, but the posterior dorsal edge is straight rather than upturned as in SL, the anterior dorsal edge of the harpe is not bulging and roughened due to granulation of the setae bases as in SL, the ventral edge of the valva is strongly convex or bulging, the harpe base is not constricted, the anterior-ventral arm of the valva extends anteriorly to very weakly dorsally join with the juxta wishbone prop (the last three attributes all similar to SL). The vinculum in lateral view is moderately wide, either straight-sloping or weakly curved anteriorly (different from SL, which has a vertical or squared aspect relative to the valva) before sharply bending anteriorly at the base to form the combined bifurcate saccus and vinculum cross-brace (as found in the group); the juxta is robust as in SL but the juxta wishbone pedicle is even more robust and the attachment point on the aedeagus is slightly anterior of the vinculum arms as in SP. The vinculum arms next to the valvae slightly widen dorsally from the base, where they attach to the valvae and also noticeably dorsal of the apex

152 Australian Entomologist, 2012, 39 (3)

angularis on the tegumen; a small rounded anterodorsal appendage on the tegumen is present (similar to SL). The aedeagus is slightly down-curved (similar to SL); the phallobase is not enlarged in S. discalis, which is diagnostic within the group in SA, the proximal orifice opening is posterior (similar to SP and SL). The female genitalia were not studied.

Figs 66-70. Male genitalia, S. discalis lateral views: (66) Hincks CP; (67) Binnie; (68) Hincks CP; (69) Pinkawillinie CP (east); (70) Inila.

Australian Entomologist, 2012, 39 (3) 153

Hostplants. The present authors, as well as others (Edwards 2006, Douglas 2008) have found that the primary hostplant for S. discalis is Gahnia lanigera (Cyperaceae), meaning that this species is usually found in the presence of that particular host. However, confirmed S. discalis is not averse to using other sedge plants in the vicinity of the primary host, based on visual sightings of female oviposition and the presence of early stages. In the Southeast Region, AS has seen females utilise the small sedges Schoenus breviculmis and Schoenus deformis (Cyperaceae) as hostplants. Douglas (2008) noted that in confusion after fire, female S. discalis oviposited on L. carphoides in northwestern Victoria.

Habitat. We have found S. discalis to occur only in the presence of its primary host G. lanigera. This plant is a dryland sedge favouring open mallee type habitat having a limestone base.

Distribution and flight period. S. discalis closely follows the distribution of its primary hostplant G. lanigera and has been found in the Regions of the Southeast-east Mt Lofty Ranges (extending into northwestern Vic), southern Yorke Peninsula and Eyre Peninsula (Fig. 38). There are no S. discalis records from Kangaroo Island, northern Yorke Peninsula or areas north of Adelaide. On the basis of Gahnia lanigera being the primary hostplant of S. discalis, then the latter should have a broader range than is currently documented.

Although sympatric with the other species in the group, S. discalis has always been found to be in peak flight earlier than the others. Males generally start to fly first, followed by the females about a week later, and there is usually a short peak period of emergence when both sexes tend to be more obvious (even though flight numbers tend to be fewer than for the other group species). The flight period for S. discalis has not been fully documented, but the flight occurs earlier in the warmer northern parts of its range than in the cooler south. It is likely contingent on weather conditions in early spring. In the Southeast the flight period lasts about three weeks, with the greatest number being present approximately 10 days after season commencement, with males seemingly outnumbering females.

On northern Eyre Peninsula, flight has been noted as early as 26 September, peaking in early October and then finishing by late October. On southern Eyre Peninsula the flight is during October to mid November, peaking in early November at Port Lincoln (CSIRO 2012). They have been recorded in early to mid November on southern Yorke Peninsula. In the Southeast they occur during November. A similar north to south range of flight periods from early-October to mid-November occurs across northwestern Victoria (Douglas 2008).

Egg (Fig. 71). Very similar to those of the other species, having 10-13 longitudinal ridges (n = 10), 1.9-2.5 x 0.85-1.0 mm (n = 14) and ~38 cross

154 Australian Entomologist, 2012, 39 (3)

striae (n = 1). The ridges are sometimes divided. Pale sub-translucent yellowish white when freshly laid but white at eclosion, which occurred after 32 days (n = | from Pinkawillinie CP east).

Fig. 71. S. discalis eggs and eclosed larvae: egg with 12 longitudinal ridges, ~38 striae, laid 8.x.2011, Inila; egg shell, eclosed larvae, Pinkawillinie CP (east).

Larvae (Figs 71-77). A first instar larva at eclosion (n = 1 ex Pinkawillinie CP east) was 3.0 mm long (extended) (Fig. 71). A near-mature larva (20 mm) was found by RG in a small G. lanigera plant from north of Ceduna. It was sub-translucent greenish grey when fresh (Fig. 72) (presumably it had been eating fresh culm or leaf material), but soon lost the greenish colour (Fig. 73) after being removed from the hostplant. It was observed in the culm just below ground level. The Gahnia was dead in its central part, possibly due to consumption by the larva.

Suspected immature larvae (Fig. 74) were observed by AS on G. lanigera and on Schoenus breviculmis and S. deformis in the Malinong-Boothby area of the Southeast Region; these were also found in the culm below ground level. (Identification of these larvae as S. discalis is based on adults being seen to oviposit on these plants.) These larvae were sub-translucent, pale grey-white in colour. A probable near-mature larva (21 mm) was also found in the culm of G. lanigera and had a sub-translucent pinkish white colour, with some dark brown dorsal areas and a dorsal line (Figs 75-77). The pink

Australian Entomologist, 2012, 39 (3) 155

markings were seen on fat-like platelets under the skin (Fig. 78). It possessed a large, smooth and shiny, orange-yellow dorsal pro-thoracic plate on thoracic segment (TS) 1, the edges of the plate were darker and there were a pair of separated dark orange-red triangular markings on the front edge of the plate. The head was brown, smooth and shiny with black mandibles, the anal segment was pale brown peripherally, with a large dorsal dark brown half- circle anal plate at the anterior-dorsal margin. Scattered, moderately long, fine dark setae were present on the body and head, slightly longer on the anal segment. No attempt was made to map the setae distribution. Generally of slightly flattened, posteriorly tapered, cylindrical shape, typical for Synemon (c.f. S. magnifica in Common and Edwards 1981), the thoracic segments enlarged and the abdominal segments with rudimentary legs generally unsuitable for traction. The dorsal anterior and mid segments often have a roughened elliptical patch (Fig.73), presumably used for either gripping or compacting their tunnels. There was no reliable morphological character that could be used to separate this larva from that of S. parthenoides, except perhaps for the seemingly different arrangement of the ‘fat platelets’, which requires more study.

Figs 72-73. S. discalis mature larva 20 mm on G. lanigera from Kalanbi.

Larvae from the Southeast were observed at the base of the hostplant (culm) below ground level, but not in the root zone. They created a smooth shelter cavity to suit their size, but no silk was used. They are very sensitive to light and will hide if exposed. Based on the size of larvae observed at varying times of the year, and the experience with S. parthenoides larvae, we believe the larvae take two years to complete their growth.

Larval predators. Similar possible predators observed in the vicinity of S. parthenoides larvae have also been observed with larvae of S. discalis.

156 Australian Entomologist, 2012, 39 (3)

Figs 74-78. S. discalis, larvae ex Southeast, SA: (74) immature 9 mm 25.11.2011 ex S. breviculmis; (75-77) mature 21 mm, 25.ii.2011 ex G. lanigera; (78) mature larva (21 mm) from G. lanigera, close-up of anterodorsal portion 25.ii.2011, showing head,

prothoracic plate, dorsal elliptical ridges, setae.

Australian Entomologist, 2012, 39 (3) 157

Pupae (Figs. 79-82). No pupae or pupal exuviae of Synemon were found by RG on Eyre Peninsula. A suspected pupa of S. discalis was found by AS in a small S. deformis plant in the Mt Rescue CP on 19.x.2011. The pupa occurred head-upwards in the culm, 1.5 cm below ground level in a ‘made to size’ cavity. A silked tunnel (or ‘cocoon’, as used by S. parthenoides) was not noted. The pupa was male, brown 14.5 x 3.6 mm (Figs 79-82), cylindrical and although smaller, was essentially identical to the ‘pupae’ (pupa exuviae) of S. parthenoides (Figs 27-31). The S. discalis pupa was critically injured during extraction and so could not be used to confirm the identification of the adult by way of ecdysis. It is apparent from the work of Douglas (2008) and Edwards (in Douglas 2008), and from our observations, that S. discalis larvae do not leave the hostplant like S. parthenoides to pupate, and the construction of a silken tunnel or cocoon is also not obligatory. The flattened spines on the pupal abdomen (Fig. 82) are strong (similar to S. parthenoides) and constructed such that any movement that they might allow the pupa would primarily be in a forward (head) direction. A cremaster was not present on the pupa.

Figs 79-81. S. discalis pupa from Schoenus deformis, (m) 14.5 mm 9.x.2011.

158 Australian Entomologist, 2012, 39 (3)

Fig. 82. S. discalis pupa, closeup of posterior-dorsal spines.

Adult biology. Typically, males tend to stay close to the hostplants, preferring open spaces, either by flying above the plants or by basking or resting on clear ground, car tracks or plant debris nearby. They tend to fly closer to the ground than other species in the group, possibly because their hostplants are normally smaller than Lepidosperma spp. They are not known to actively patrol on hill and dune tops, but will utilise them if their host is nearby. While in flight males can detect females on the ground from a few metres and immediately divert to where the pheromones are coming from. Adults fly rapidly when disturbed, resembling the flight of skippers. When disturbed, females tend to fly a distance between 10-30 m in one direction before settling. Males have a tendency for a part return flight. Their normal flight tends to be in a fast, irregular zig-zag fashion. Both sexes react to intrusions by other sun moths or insects with females simply flying away, while males engage in ‘dogfights’ before resettling. Adults fly in full sun; however in hot conditions they will fly under high cloud. Their flight is seemingly fast and active and their exceptional vision (similar to other Synemon) is such that they easily evade most intrusions, responding to approaches from roughly 3-5 metres.

Adults become active around 1000 h. Males are active before females, which usually become active around midday, with the greatest number of individuals flying from midday until 1400 h. Males tend to check hostplants for females early and, if they cannot find any, then tend to fly off to other areas or rest on cleared ground. Females, as with other species in the group, tend to fly close to the ground in search of suitable hostplants, sensing the presence of the plants while in flight, we believe by both sight (initially) and later by chemical cues. Once selected, females typically land on the higher outer part of the plant, then walk down head first to the base. When performing this, the wings are held upright with regular slow, flapping

Australian Entomologist, 2012, 39 (3) 159

movements. At the base she turns upright and the wing movements stop, then she descends backwards to ground level and deeply probes the ground or sides of the plant with her ovipositor before oviposition. Presumably only one egg is laid, based on the time expended, but minimal effort was made by us to try and find the egg(s), due to their small size and well camouflaged location. When oviposition concludes, the female will fly on and repeat the process some 2-3 m away. The time taken to lay eggs is about one minute. We have not seen adult S. discalis nectaring on flowers even though the proboscis is fully developed. Douglas (2008) observed males nectaring on Dampiera rosmarinifolia in northwestern Victoria.

Comments. The morphological and biological information on the cryptic Synemon species discussed in this paper show there are differences between them that can be used to taxonomically differentiate them. Synemon discalis adults, when in good condition, clearly differ from those of other SA group members by their collective wing and male genitalia morphologies. The wing pattern has a diagnostic difference and there is a diagnostic difference in the male genitalia, i.e. the lack of an expanded phallobase. The overall similarity of the wing patterns and male genitalia indicate that the three species are congeneric, while the collective character differences indicate that they (plus one subspecies) are taxonomically distinct.

The presence of S. discalis on Eyre Peninsula suggests that the species has considerable dispersal ability, especially given its presence throughout temperate SA and northwestern Vic and possibly also in WA.

Interestingly, even though the male genitalia of S. discalis and S. parthenoides are very similar, and yet dissimilar to those of the S. collecta Swinhoe species group (Grund 2011), Kallies et al. (2008) nested the discalis-parthenoides clade within the latter species group in their phylogenetic analysis.

Acknowledgements

Specimens collected by the authors in South Australia were obtained under permit numbers U23970 and A25806 issued by the Department for Environment and Heritage. We are grateful to Peter Hudson for access to the remnant SAMA Castniidae collection; to Len Willan, photographer of the images of S. discalis displayed on the CSIRO Entomology website ‘Australian Moths Online’, for permissions to use his images in this paper; and to Andrew Lines for access to his Synemon collection. Plants mentioned

in this paper were identified by Rosemary Taplin at the State Herbarium of SA.

References

BOISDUVAL, J.A. 1875. Sphingides, Séstides, Castnides. In: Boisduval, J.A and Guenée, A. (eds), Histoire Naturelle des Insectes. Species Général des Lépidoptéres Hétérocéres. Librarie Encyclopedique de Roret Vol. 1, Paris; 568 pp. [dated 1874, in French]

160 Australian Entomologist, 2012, 39 (3)

COMMON LF.B. and EDWARDS. E.D. 1981. The life history and early stages of Synemon magnifica Strand (Lepidoptera: Castniidae). Journal of the Australian Entomological Society 20: 295-302.

CSIRO, 2012. Australian National Insect Collection Taxon Database, Castniidae. Viewed 6 March 2012. <http://anic.ento.csiro.au/database/biota_details.aspx?BiotalD=26617>

CSIRO Entomology, 2012. Australian Moths Online, Castniidae. Viewed 6 March 2012. <http:/www1.ala.org.au/gallery2/main.php ?g2_itemld=10720>

DOUGLAS, F. [2004]. Five threatened Victorian sun-moths (Synemon_ species). Action Statement, Flora and Fauna Guarantee Act 1988, Victoria. No. 146; 11 pp.

DOUGLAS, F. 2008. The sun-moths (Lepidoptera: Castniidae) of Victoria, with a detailed study of the pale sun-moth (Synemon selene Klug, 1850). Master of Applied Science Thesis, University of Ballarat, Accepted June 2008; 323 pp.

EDWARDS, E.D. 1996. Castniidae. P. 138, in: Nielsen, E.S., Edwards, E.D. and Rangsi, T.V. (eds), Checklist of the Lepidoptera of Australia. Collingwood : CSIRO Publishing; 529 pp. EDWARDS, E.D. 2006. Australian Faunal Directory. Australian Biological Resources Study , Canberra. Superfamily Castnioidea, Complete Review. Viewed 1 July 2009. <http:/Avww. environment.gov.au/biodiversity/abrs/onlineresources/fauna/afd/taxa/castnioidea/complete> EDWARDS, E.D., GENTILI, P., HORAK, M., KRISTENSEN, N.P. and NIELSEN, E.S. 1999. The Cossoid/Sesioid assemblage. Pp 181-197, in: Kristensen, N.P. (ed.), Lepidoptera, moths and butterflies. Vol. 1: Evolution, systematics and biogeography. de Gruyter, Berlin.

GRUND, R. 2011. A new species of Synemon Doubleday (Lepidoptera: Castniidae) from the Colona Plains, South Australia. Australian Entomologist 38(4): 167-178.

KALLIES, A., BRABY, M.F., HILTON, D. and DOUGLAS, F. 2008. The extent of genetic variability between and within the parthenogenetic morphs of the pale sun-moth, Pp 217-229 [appendix], in: DOUGLAS, F. 2008, ibid.

KLUG, [J.C.F.] 1850. Uber die Lepidopteren-Gattung Synemon. Nebst einem Nachtrage über Castniae. Abhandlungen der Kéniglichen Akademie der Wissenschaften zu Berlin 1848 (Physikalische part): 245-257. [In German]

McQUILLAN, P.B. and FORREST, J.A. 1985. A guide to common moths of the Adelaide Region. Special Educational Bulletin Series (No. 5), South Australian Museum, Adelaide. STRAND, E. [1911]. Castniidae. In: Seitz, A. (ed.), The Macrolepidoptera of the World. Vol 10. Bombyces and Sphinges of the Indo-Australian Region. 2 parts. Alfred Kernen, Stuttgart; 909 pp, 100 pls. [English Version]

TEPPER, J.G.O. 1882. The Papilionidae of South Australia. Royal Society of South Australia Transactions and Proceedings 4: 25-36, pls 2-3.

TINDALE, N.B. 1928. Preliminary note on the life history of Synemon (Lepidoptera, Fam. Castniidae). Records of the South Australian Museum 4: 143-144.

Australian Entomologist, 2012, 39 (3) : 161-177 161

A REVIEW OF THE NEW GUINEAN GENUS PARAMECOCNEMIS LIEFTINCK (ODONATA: PLATYCNEMIDIDAE), WITH THE DESCRIPTION OF THREE NEW SPECIES

A.G. ORR!, V.J. KALKMAN? and S.J. RICHARDS?

'Griffith School of the Environment, Griffith University, Nathan, Qld 4111 ?National Museum of Natural History, Leiden, Netherlands 3 South Australian Museum, North Terrace, Adelaide, SA 5000 and Museum and Art Gallery of the Northern Territory, PO Box 4646, Darwin, NT 0801

Abstract

The genus Paramecocnemis Lieftinck, previously known from two species from northern New Guinea, is redefined on the basis of new material recently collected in the Sepik Basin and Western Province of Papua New Guinea. Three new species are described: P. spinosus sp. n. and P. similis sp. n. are quite close to the generic type species, P. erythrostigma Lieftinck, while P. eos sp. n. is more distantly related to known species and probably of basal stock.

Introduction

The zygopteran family Platycnemididae, which is absent from Australia (Kalkman et al 2008, Kalkman and Orr 2012), is richly represented in New Guinea by over 40 species in 11 genera in the subfamily Calicnemiinae (excluding Hylaeargia Lieftinck and Palaiargia Forster). Almost all New Guinean members of the subfamily may be recognised by the distinctive marginal crenulations at the wing tips, a feature found elsewhere only in the Philippine subgenus Risiocnemis (Risiocnemis) Cowley.

The genus Paramecocnemis Lieftinck, 1932, was erected to accommodate the long-bodied P. erythrostigma Lieftinck, 1932, which exhibited several venational features not then known in the related Jdiocnemis Selys, including the fusion of M3 and Rs for a short section slightly distad of their independent origins, a feature unknown in other Platycnemididae. The main diagnosis of the genus, as given by Lieftinck (1932), is based on venational characters found in both sexes. Various other unique male characters relating to the venter of the thorax and abdomen were also listed. Subsequently, another even longer bodied species, P. stillacruroris Lieftinck, 1956, was described. It too exhibited the critical fusion of M; and Rs, possessed roughly similar male terminal appendages and shared with P. erythrostigma several of the unique male characters already identified in Lieftinck (1932).

Recent collections sponsored by Conservation International’s RAP (Rapid Biodiversity Assessment Programme) in the Muller Range and by Xstrata Copper in the Sepik Basin have yielded representatives of three new species from Papua New Guinea, two of which are very similar to P. erythrostigma with the exception that their abdomens are much shorter, and another more distant species which, while lacking the diagnostic venational characteristics of the genus, possesses other male structures which clearly ally it more closely with Paramecocnemis than Idiocnemis, the other genus in which it might be placed. These three species are described here.

162 Australian Entomologist, 2012, 39 (3)

It is, however, necessary first to review Lieftinck’s (1932) original diagnosis of the genus, with greater emphasis placed on clear synapomorphies in male structures, with wing venation generally and abdomen length being shown as more labile and, in consequence, less reliable in generic definition. Terminology used follows Westfall and May (2006), with exception of the anal appendages, where we follow Watson et al. (1991). Type specimens are deposited in The National Museum of Natural History, Leiden (RMNH), the Museum and Art Gallery of the Northern Territory (NTM) and the South Australian Museum (SAM).

Generic definition of Paramecocnemis

In his original definition of the genus Paramecocnemis, Lieftinck (1932) stressed the shape of the wing and venational characters, particularly with respect to the differences between the typical species, P. erythrostigma, and known members of the genus /diocnemis, with which Paramecocnemis is undoubtedly most closely allied (Gassmann 2005). This definition was repeated with few alterations in a key to genera of Platycnemididae (Lieftinck 1949). Since the original definition, numerous new species of /diocnemis have been discovered, some of which exhibit characters already listed in the original description as unique to Paramecocnemis. In addition, new material collected in the last decade, evidently belonging to Paramecocnemis, does not conform to Lieftinck’s generic characters. Therefore, the following generic traits with respect to Idiocnemis must be discarded: (1), wings less strongly petiolated with distal portion narrower — this is a tendency only, with several exceptions in Idiocnemis; (2), quadrilateral longer with the lower distal angle more acute — this is also a general trend associated with long thin wings and the acute distal angle is not especially noticeable in the type species, P. erythrostigma; (3), Mz and M;, more widely separated at the origin in Paramecocnemis — this is a trend only, with the two veins separated by 2 crossveins in the forewing and 3 crossveins in the hindwing commonly occurring in both Paramecocnemis and Idiocnemis; (4), Rs and M; arising separately near subnodus but fused for one cell breadth or more — in the Platycnemididae this character is unique to Paramecocnemis but the present study includes one species in which the fusion does not occur, although the two veins approximate very closely just beyond their origins — in another species the character is variable, with complete fusion for nearly a cell breadth in some specimens, none in others. Other male characters noted were: (5), abdomen very long, at least twice as long as hindwing — abdomen length is highly unreliable as a character and clear examples of Paramecocnemis are now known in which the abdomen is of moderate length; (6), abdominal segments 5-7 with patches of long fine ventral setae — these are present only in some species and may be confined to S7; (7), male anal appendages highly modified — this character refers mainly to the strongly down-turned superior appendage, not present in all members of the genus as recognised here.

Australian Entomologist, 2012, 39 (3) 163

Lieftinck (1932, 1949) also noted several characters, unique to the male, not found in Idiocnemis or other platycnemidid genera, which together completely define the genus (see Fig. 6). These include: (1), ‘posterior third of poststernum rather swollen in its middle, the convex surface being closely beset with a bunch of soft golden hairs which are directed caudad’ — although the degree of swelling is somewhat variable, a dense narrow tuft of long caudally directed setae is present in all species of Paramecocnemis (in Idiocnemis only sparse unbunched setae occur in a few species and there is no swelling; in other unrelated genera a swelling and fairly dense setae may be present but never in the same arrangement or of the same length); (2), the sides of the first tergite project straight down, rather than turning to enclose the segment, and each bears a fringe of long coarse setae — this character appears unique although it is variably developed, being especially prominent in P. erythrostigma and reduced in P. stillacruroris; (3), the lower margin of the second tergite with strong tooth, well before caudal margin — this character does not occur in Jdiocnemis and is a clear synapomorphy. Lieftinck (1956) noted that this character is less developed in P. stillacruroris, but it is nevertheless clearly present.

Characters present in male P. erythrostigma, not mentioned by Lieftinck in his generic diagnosis and found in some but not all species sharing the above three characters include: (1), median lobe of prothorax with strong projecting cone on either side (much reduced in P. stillacruroris); (2), gonopore of abdominal S9 situated slightly beyond midpoint of segment; (3), genital valves flanking gonopore large and bearing long setae, giving the segment a ventrally notched appearance in profile; (4), abdominal S9 produced ventrally and bearing a dense tuft of long, backward directed setae. None of these characters of S9 are present in Idiocnemis, where the gonopore is situated nearer the apex of S9 and the genital valves lack long setae, but this condition also occurs in one species which appears best placed in Paramecocnemis.

Owing to lack of material, female Paramecocnemis cannot yet be unambiguously defined, but the following venational character reliably separates them from /diocnemis in all cases known so far: Rs and M; arising separately near subnodus but united near base for half one cell breadth or more.

Paramecocnemis eos sp. n. (Figs la-e)

Type material. Holotype 6, PAPUA NEW GUINEA: Western Province, CI Muller Range expedition, Camp | (Gugusu), 05°43.751’°S, 142°15.797°E, 515 m asl, 04- 11.ix.2009, leg VJ Kalkman; deposited in RMNH.

Diagnosis. A finely built damselfly of small-medium size; ground colour dark with pale green and cerulean blue markings. Males with ventral tufts of setae on post sternum and on tergites of the first abdominal segment. Wings

164 Australian Entomologist, 2012, 39 (3)

with moderately dense reticulation; pterostigma small, dark and lozenge shaped; distal margins crenulated. It may be distinguished from its congeners by its shorter abdomen and/or the form of the male anal appendages.

Description. Head (Fig. la): short and lightly built. Labium pale bluish white, dark at extremities; medium lobe with small shallow ‘V’ shaped incision. Remainder of underside of head dark. Labrum and clypeus black. Front of head, including mandibles, genae and frons pale bluish green, forming a broad transverse band not reaching antennal sockets. Remainder of head black except for two medium sized, bright blue postocular spots, roughly triangular in outline; several long black setae arise from these spots. Antennae black; 2nd segment long; remaining segments missing from specimen. Eyes black above, probably light green below in life (pale ochre in specimen).

Thorax (Fig. la): prothorax: anterior lobe small, dark, posteriorly curving into a groove marking boundary with median lobe; median lobe mainly pale green laterally; dorsally with two strong conical horns, pale bluish green on their outer face, otherwise dark; dorsal area of median lobe dark, these extending laterally anterior to and along part of the base of the horns; posterior lobe dark; produced into a flat small semi-erect process, roughly rectangular in profile with a slight curve to its posterior margin. Synthorax: dorsally with pale bluish green antehumeral band, broad anteriorly, terminating acutely level with a point at two thirds of length humeral suture, inner margin of band obscured posteriorly; mesepimeron with upper two thirds pale bluish green except for fine black line bordering humeral suture, remainder dark; metepisternum pale green; merging with pale area of mesepimeron, except for a narrow dark margin along metapleural suture, becoming broader toward metinfraepisternum; metepimeron with pale yellowish green triangular patch posteriorly, separated from green of metepisternum by broad black band; mesinfraepisternum and metifraepisternum both black except for small pale green mark in posterior comer; venter of synthorax pale yellowish; posterior third of post sternum with elevated tubercle bearing tuft of long, dark, coarse setae. Legs missing beyond trochanters on synthorax. Legs of prothorax short with dense, long, fine spines; overall coxae pale green; trochanters pale on meso and metathorax with posterior dark marking; on prothorax legs trochanters and remainder of legs dark, except for inner surface of trochanters and femora which are pale yellowish. Wings (Fig. Ib): hyaline with relatively dense neuration; weakly petiolated but fairly broad (max. breadth: length ratio — 0.20 forewing; 0.21 hindwing); distal margins crenulated; M3 and Rs arising near subnodus; closely approximated near origin but not fused at any point; quadrilateral moderately long, distinctly longer and narrower in hindwing, with lower distal angle strongly acute in both wings; origins of M3 and M,, separated by two cross veins in forewing, three in hindwing; forewing 17.5 Px; hindwing 15.5 Px; pterostigmata small and black, diamond-shaped,

Australian Entomologist, 2012, 39 (3) 165

covering less than one cell in forewing and one cell exactly in hindwing. Articulated sclerites at each wing base (costal plates - as viewed with wings closed) with large external bright blue spot.

Fig. 1. Paramecocnemis eos sp. n., male: (a) right lateral view of thorax and first two abdominal segments and dorsal view of head; (b) right wings; (c) S10 and anal appendages in left lateral view; (d) S10 and anal appendages in dorsal view (inferiors

shaded); (e) distal part of S9, with genital valves, S10 and anal appendages in ventral view (inferiors shaded).

Abdomen: long and very thin; slightly expanded in basal segments (S1 and $2); strongly laterally expanded in terminal segments (S8-S10); S1 with lateral margins of tergites slightly produced downward and not wrapped under the base of the segment and bearing definite tuft of fairly short but

166 Australian Entomologist, 2012, 39 (3)

thick setae; S2 with well defined subapical tooth on ventral margins of tergites. Ground colour of abdomen dark; basal segments marked with pale blue-green as shown in Fig. 1a; S4-S7 unmarked; S8-S9 with broad bright cerulean blue patches on their dorsal surface, somewhat tapered inward towards base of S8. S10 and appendages black; S10 (Figs 1b-d) about as long as deep; superior appendages about as long as S10; curved downward strongly from a point about two-thirds of the way along the dorsal margin; forcipate in dorsal view; inferiors basally broad, attenuating rapidly to incurved, slightly bifid tip, which reaches just beyond inner margin of superiors; basal part with thick ventral tuft of long setae; gonopore situated near posterior margin of S9; genital valves email) lacking long setae, not visible in profile. Fine ventral setae occur sparsely along the abdomen but are best developed on S1 and the tooth of S2 (Fig. 1a). No obvious ventral setae

on S5-S7. Measurements (mm): forewing, 23; hindwing, 22; abdomen + appendages, 36.5.

Variation. Unknown; the holotype is the only known specimen.

Etymology. The name eos is a noun in apposition from the Greek ’nwe, meaning ‘dawn’, a reference to the probable basal position of the species

within the genus.

Habitat. Only a single specimen was seen, which was collected along a small and steep stream in virgin forest.

Paramecocnemis spinosus sp. n. (Figs 2a-d, 3a-b) Type material. Holotype 3 (1008588), PAPUA NEW GUINEA: West Sepik Province, upper Sepik Basin, 4°39’S, 141°43’E, 800 m asl, 07.vi.2010, leg S.J. Richards; deposited in NTM. Paratypes: 1 3, 1 Q (supposition), same locality, 06.vi.2010; 3 33 3 2, collected within 200 m radius of type locality between 30.xi.2009- 4.xii.2009, leg S.J. Richards. All deposited in RMNH.

Diagnosis. A finely built damselfly of small-medium size; ground colour dark with pale blue and cerulean blue markings (the former discoloured in preserved specimens). Males with ventral tufts of setae on post sternum and on tergites of the first and sternum of last abdominal segment. Wings with moderately dense reticulation; distal margins crenulated; pterostigma small, dark and lozenge shaped. It may be distinguished from its congeners by its shorter abdomen and/or the form of the male anal appendages.

Description of Holotype male. Head: lightly built; labium entirely black bearing sparse long setae; median lobe with shallow ‘V’ shaped incision; labrum and clypeus black, margin of labrum with long coarse setae; mandibles, genae and lower half of frons with narrow transverse pale blue band, broadly stepped slightly caudad on genae; upper part of frons black

Australian Entomologist, 2012, 39 (3) 167

with paired low prominences, each bearing a tuft of long setae, anterior and interior to the antennal sockets; remainder of head black except for postocular lobes which bear large, bright blue spots. Antennae black; segments 2-7 relatively long . Eyes black above, pale blue below in life (Fig. 4).

Fig. 2. Paramecocnemis spinosus sp. n. male holotype: (a) right lateral view of thorax and first two abdominal segments and dorsal view of head; (b) right wings; (c) posterior section of S9, S10 and anal appendages in left lateral view; (d) S9, S10 and anal appendages in dorsal view.

Thorax: posterior lobe small, black, with slight transverse furrow; median lobe with lower half of sides pale blue; upper half black; dorsum black with two prominent blunt conical horns; posterior lobe small, slightly elevated triangular flap; black except for slight blue edging laterally. Synthorax finely

168 Australian Entomologist, 2012, 39 (3)

built with black ground colour; antehumeral bands about half breadth of mesepisternum, parallel sided for most of their length and ending diffusely at about a point at 7/8ths of length of mesepisternum; laterally synthorax with thin pale blue band extending along the length of the metepisternum and separated from metapleural suture along its length by a black band; small contiguous patch of pale blue curving up to form diffuse narrow block of colour around the upper one quarter of the mesepimeron. Metepisternum with irregularly defined, long, pale blue patch in its posterior half; mesinfraepisternum with blue mark at _ posterio-ventral corner; metinfraepisternum black; venter of synthorax except for posterior 2/Sths of post sternum, which is pale, including a raised protuberance bearing a tuft of heavy, long, golden brown setae. Legs fine and moderately long, bearing long thin spines; mainly dark with pale markings posteriorly on coxae, anteriorly and internally on trochanters and femora, the pale coloration becoming paler from the pro- to the metathoracic legs. Wings (Fig. 2b): hyaline with relatively dense neuration; weakly petiolated and moderately broad (max. breadth: length ratio — 0.19 forewing; 0.20 hindwing); distal margins crenulated; M3 and Rs arising near subnodus; closely approximated near origin and fused for about 1 cell length in forewing and half a cell length in hindwing to about the level of Px1; quadrilateral long, distinctly longer and narrower in hindwing, with lower distal angle strongly acute in both wings; origins of M) and Mj, separated by two cross veins in forewing, four in hindwing; forewing 17.5 Px; hindwing 15 Px; pterostigmata small and black with fine pale margin, diamond-shaped, covering less than one cell in forewing and one cell exactly in hindwing.

Abdomen: long and thin; slightly expanded in basal segments (S1 and S2); distinctly laterally expanded and flattened in terminal segments (S8-S10); S1 with lateral margins of tergites produced downward bearing dense tuft of long black setae; S2 with well defined subapical tooth on ventral margins of tergites. Ground colour of abdomen dark; basal segments marked with small pale blue patches as shown in Fig. la; S4-S7 unmarked; S6 with patch of fine long ventral setae towards apex, S7 with patch of fine long ventral setae in basal half; S8-S9 with broad bright cerulean blue patches on their dorsal surface, that of S8 triangular, tapered to a rounded point 2/3rds of way to the base of the segment. S10 and appendages black; S10 (Fig. 2b, c,) almost as long as deep with strong ventral swelling bearing dense tuft of long black setae; superior appendages about as long as S10; heavy and curved downward more than 90° from a point about halfway along the dorsal margin; outer margin forcipate in dorsal view but with interior surface filling almost all intervening space ventrally; inferiors basally broad, attenuating rapidly to incurved, slightly bifid tip, which reaches just beyond inner margin of superiors; arising from around the midpoint of the outer part of the appendage is a long strong spine, directed inwards, upwards and slightly cephalad, the pair nearly meeting in dorsal view; gonopore situated slightly

Australian Entomologist, 2012, 39 (3) 169

distad of midpoint of S9; genital valves large, bearing long setae, visible in profile as a distinct notch in the underside of the segment.

Measurements (mm): forewing, 21.5; hindwing, 20.5; abdomen + appendages, 36.

Female (supposition) (Figs 3a-b). Head: marked as in the male but with pale coloration more extensive; anterior transverse band across head covers almost all of frons being level with markings on genae; frontal prominences not defined as entire frons is slightly protruding, but tufts of long dark setae arise from similar locations to those on male; postocular lobes are pale blue and more extensive, being connected by a fine band along the occipital bar.

Thorax: prothorax anterior lobe black; remainder pale blue; median lobe without horns found in male; posterior lobe a short, triangular flap. Synthorax marked as in male but pale areas more extensive; antehumeral bands broader and nearly reaching alar triangle; upper one quarter of mesepimeron and most of metepisternum pale; posterior half of metepimeron pale; posterior one third of poststernum pale. Legs moderately long, mainly black; coxae and trochanters paler than in male; femora marked as in male with pale inner streaks. Wings moderately broad (breath: length 0.20 in both wings) but not strongly petiolated; outer margins crenulated. Differs from male slightly in venation; M and Mj, separated by 1 and 2 cross veins in the forewing and hindwing respectively. M3 and Rs fused for one cell length in forewing and half a cell length in hindwing. Pterostigma medium brown.

Fig. 3. P. spinosus sp. n., female: (a) left lateral view of thorax and first two abdominal segments and dorsal view of head; (b) left lateral view of terrninal abdominal segments showing ovipositor and anal appendages.

170 Australian Entomologist, 2012, 39 (3)

Abdomen: S1 and S2 both with a blue saddle mark; small subdorsal blue flecks present at the posterior margin of S2; base of S3 with thin dorsal blue marking; remainder of segments black except for S8 and S9 which are broadly cerulean blue dorsally (Fig. 3b). Terminal segments distinctly clubbed. Anal appendages thin and conical, slightly shorter than S10; valves with slightly pale tip, serrated ventrally, extending just beyond level of anal tubercle.

Measurements (mm): forewing, 21-22.5; hindwing, 20.5-22; abdomen + appendages 29.5-32.

Variation. The following variation occurs in males: The pale marking on the mesepimeron may be either slightly more extensive than in the holotype, occupying most of the upper quarter, or absent, resulting in a single thin, regular stripe laterally. The poststernum may be deeply infuscated, with only the raised protuberance clearly pale. M3 and M,, may be separated by 3 cross veins in the forewing and/or hindwing and two wings are not always symmetrical in this character. The degree of fusion of M3 and Rs varies, especially in the hindwing, with no fusion in the hindwing of one specimen. Variation in size is negligible.

Females show slight variation in the extent of pale banding on the side of the synthorax and variation in size as noted. In two female specimens, the pale blue marking on the dorsum of andominal segments S8-S9 is not clearly evident but this appears to be an artefact of poor preservation.

Etymology. The name spinosus, a Latin adjective, refers to the distinctive spine on the inferior appendage of the male.

Habitat. All specimens were found in sun patches in rainforest along trails near clear, rocky streams.

Paramecocnemis similis sp. n. (Figs 4, Sa-d) Type material. Holotype 6, PAPUA NEW GUINEA: upper Sepik Basin, West Sepik Province, 4°44’S, 141°47’E, 425 m asl, 18.11.2010, leg S.J. Richards; deposited in RMNH. Paratypes: 1 G, same data; 1 ĝ, same locality, 19.ii.2010; 1 4, same locality, 20.11.2010. All deposited in RMNH.

Diagnosis. A finely built damselfly of small-medium size; ground colour dark with pale blue and cerulean blue markings. Males with ventral tufts of setae on post sternum and on tergites of the first and sternum of last abdominal segment. Wings with moderately dense reticulation; distal margins crenulated; pterostigma small, dark and lozenge shaped. It may be distinguished from its congeners, especially P. spinosus, by its slightly darker markings and the form of the male anal appendages.

Australian Entomologist, 2012, 39 (3) 171

Fig. 4. Paramecocnemis similis sp. n. in nature.

Description of Holotype male. This species is so similar to P. spinosus that it is best defined by comparative notes: In general slightly darker than P. spinosus (Fig. 5a). Head with pale band across frons slightly narrower, just visible in dorsal view; postocular spots smaller, darker blue and more definitely triangular. Prothorax in lateral view darker than in P. spinosus, with reduced lateral pale markings barely reaching anterior lobe and just touching coxa at a point directly below the median lobe processes. Synthorax with pale antehumeral band shorter, terminating sharply at a point about 2/3 of the length of the mesepisternum. Laterally with pale, irregularly edged band covering anterior half of mesepisternum and extending slightly in places onto mesepimeron; no specimens with broad pale coloration on mesepimeron as found in some P. spinosus specimens; metepimeron with posterior 1/3 pale; poststernum dark, rather than pale, as in P. spinosus. Legs with posterior pale marking on meso- and metacoxae broader than in P. spinosus; otherwise similar to that species. Wings of the holotype with M, and Mj, separated by two cross veins in forewing, three in hindwing; M3 and Rs fused for about | cell length in both wings to about the level of Px;; pterostigmata small and black without fine pale margin found in P. spinosus. Abdomen in shape and ventral setae like P. spinosus, S1 with small obscure ventrolateral pale mark as well as reduced dorsal blue saddle mark; S2 with ventral margin anterior to tooth with obscure pale streak, longer than in P. spinosus; no pale marking in S3; S8-S9 as in P. spinosus with broad bright cerulean blue

172 Australian Entomologist, 2012, 39 (3)

patches on the dorsal surface. S10 less projected ventrally than in P. spinosus, but also bearing dense tuft of dark setae. Superior appendages (Fig. 5c) bent downward, slightly less strongly than in P. spinosus. Inferior appendages distinct; not quite reaching tips of superiors; apically strongly bifurcated but apparently lacking strong upwardly directed inner spine visible in profile in P. spinosus (Fig. 2b); strong inner spine nearly meeting its partner interiorly; not visible in lateral view but evident in dorsal view (Fig. 5d). In dorsal view superiors more smoothly rounded than in P. spinosus.

en oS Dumon mamman pezsi

SE = SSS SY

REE Sa

Fig. 5. Paramecocnemis similis sp. n. male holotype: (a) right lateral view of thorax and first two abdominal segments and dorsal view of head; (b) right wings; (c) posterior section of S9, $10 and anal appendages in left lateral view; (d) S9, S10 and anal appendages in dorsal view.

Australian Entomologist, 2012, 39 (3) 173

Measurements (mm): forewing, 20.5; hindwing, 20.0; abdomen + appendages, 35.

Variation. In two specimens the antehumeral band reaches nearly to the alar triangle, thus this character does not separate all specimens from P. spinosus. There are slight differences in the outline of the lateral band on the thorax, particularly along its irregular anterior margin. M, and M,, may be separated by one cross vein in the forewing and/or two in the hindwing and two wings are not always symmetrical in this character. The degree of fusion of M3 and Rs varies, especially in the hindwing, with no fusion in the hindwing of one specimen. Variation in size is slight, the hindwing ranging from 20-22 mm; abdomen+appendages from 35-36.5 mm.

Etymology. The name similis, a Latin adjective (= similar), is derived from its similarity to the previous species, P. spinosus sp. n.

Habitat. Found along small, clear streams in dappled sun in primary rainforest.

Discussion

Table 1 lists the distribution of the main characters which serve to define the genus. The first three, relating to the poststernum, abdominal S1 and S2 are essentially similar in all species, although developed to varying extents (Fig. 6). Similarly, the prothorax in all species bears two conical horns although these are slightly reduced in P. stillacruroris. Although similar structures are known in distant genera they do not occur in Jdiocnemis, the presumed sister

group of Paramecocnemis. The remainder of the characters show distinct variation within the genus.

Allowing that abdomen length is labile, the greatest similarity is shown by P. erythrostigma, P. spinosus and P. similis. The lack of fusion of M; and Rs in some wings of some specimens of P. spinosus and P. similis may not be of great significance, especially given that the wings in those two species are

slightly shorter than in P. erythrostigma, while the body stature is very similar.

It is fairly clear that P. stillacrororis is less closely allied, as remarked by Lieftinck (1949), but P. eos appears to be still further removed. The male anal appendages in this species do not differ significantly from certain species of Idiocnemis and in the single known specimen M; and Rs are clearly separate. Based on these comparisons it appears to be the most basal member of the genus Paramecocnemis.

The five species of Paramecocnemis are confined to parts of the central mountain range and to the hills in the northern lowlands of New Guinea (Fig. 7). P. eos, P. spinosus and P. similis are each known from a single location in, or on the edge of, the central mountain range at heights of respectively 515, 800 and 425 m but, given the extent of suitable mountainous habitat in

174 Australian Entomologist, 2012, 39 (3)

Table 1. Summary of significant male characters in known Paramecocnemis spp.

Species erythrostig- stillacruroris | spinosus similis ma

Post- with with with with with sternum protuberance | protuberance protuberance | protuberance protuberance bearing bearing bearing bearing bearing dense tuft of | dense tuft of

dense tuft of long setae

dense tuft of long setae

dense tuft of long setae

long setae long setae

fringe of fringe of

Sl venter | fringe of fringe of fringe of

setae at setae at setae at setae at setae at margins of margins of margins of margins of margins of tergites tergites tergites tergites tergites creating creating creating creating creating distinct tuft distinct tuft distinct tuft distinct tuft distinct tuft

(less well

(less well developed)

developed)

ventral

ventral

Tergum ventral ventral ventral S2 margin margin margin margin margin

dentate slightly dentate dentate dentate

dentate

Prothorax | 2 conical 2 reduced 2 conical 2 conical 2 conical

horns conical orns, | horns horns horns Fusion of M; Rs fused M; Rs fused M; Rs fused M; Rs fused M; Rs not Mzand Rs | for at least for at least for nearly for nearly fused

one cell- one cell- one cell- one cell-

length in length in length in one | length in one

both wings both wings or both or both

wings wings S7 ventral | long fine setae absent long fine long fine setae absent setae vetral seate venral setae ventral setae Male large, small, large large small, genital inserted at inserted near | inserted at inserted at inserted near valves (of | point just midpoint of point just point just distal end of S9 venter) | beyond S9 creating beyond beyond S9, lacking long setae

midpoint of S9 creating notch seen in

midpoint of S9 creating notch seen in

notch seen in profile — lacking long

midpoint of S9 creating notch seen in

profile — setae profile - profile -

bearing long bearing long | bearing long

setae setae setae S10 tuft of long lacking long | tuft of long tuft of long lacking long venter setae setae setae setae setae Superior bent bent bent bent bent append- downwards downwards downwards downwards downwards ages but slightly

elongated

moderate moderate

very long, moderate more than twice as long

as hindwing

long, about twice as long as hindwing

Abdomen length M

Australian Entomologist, 2012, 39 (3) 175

erythrostigma

stillacruoris

eos spinosus

similis \ Cc

Fig. 6. Comparison of the structure of the poststernum and abdominal S1 and S2, and the distriution of setae on the post sternaum and ventral margins of abdominal S1: (a) ventral tooth on S2; (b) ventral tuft of setae on S1; (c) tuft of setae on poststernum.

both northern and southern New Guinea, the new species almost certainly all have broad distributions in the region. P. stillacruroris is known from three locations and seems widespread in the central mountain range seemingly only occurring at higher altitudes from 900-1300 m (Lieftinck 1949, Oppel 2005, 2006). P. erythrostigma is the only species known from the northern lowlands of New Guinea and has a wide altitucinal range (250-1000 m). All

species are found on steep rocky streams in forest. Further details on habitat or behaviour are lacking.

176 Australian Entomologist, 2012, 39 (3)

Fig. 7. The distribution of the five known species of Paramecocnemis: green circles — P. erythrostigma; green squares — P. stillacruroris; red diamond — P. eos; blue circle —

P. spinosus; blue square — P. similis.

Acknowledgements

Material from Papua New Guinea was obtained during a Rapid Assessment Programme (RAP) biodiversity survey organised by Conservation International with support from the Porgera Joint Venture (PJV), and during a field survey organised and supported by Xstrata Copper. The authors are extremely grateful to Conservation International, PJV and Xstrata Copper for their support. SJR is grateful to the PNG National Research Institute for assistance with his Research Visa and to the PNG Department of Environment and Conservation for issuing export permits for voucher specimens. Fieldwork by VJK in the Indonesian province of West Papua was made possible by the Uyttenboogaart-Eliasen Foundation.

References CARLE, F.L., KJER, K.M. and MAY, MLL. 2008. Evolution of Odonata, with special reference

to Coenagrionoidea (Zygoptera). Arthropod Systematics & Phylogeny 66: 37-44.

DAVIES, D.A.L. and TOBIN, P. 1984. The dragonflies of the world: a systematic list of the extant species of Odonata. Volume | Zygoptera, Anisozygoptera. Societas Internationalis Odonatologica Rapid Communications (Supplements) 3: 1-127.

FRASER, F.C. 1957. A reclassification of the order Odonata. Royal Zoological Society of New South Wales, Sydney; 134 pp.

GASSMANN, D. 2005. The phylogeny of Southeast Asian and Indo-Pacific Calicnemiinae (Odonata:Platyenemididae). Bonner Zoologische Beiträge (2004) 53: 37-80.

Australian Entomologist, 2012, 39 (3) 177

KALKMAN, V.J., CLAUSNITZER, V., DIJKSTRA, K.-D.B., ORR, A.G., PAULSON, D.R. and TOL, J. van. 2008. Global diversity of dragonflies (Odonata) in freshwater. Hydrobiologia 595: 351-363.

KALKMAN, V.J. and ORR A.G. 2012. The Australian monsoon tropics as a barrier for exchange of dragonflies (Insecta: Odonata) between New Guinea and Australia. Hydrobiologia 693: 55-70.

LIEFTINCK, M.A. 1933. The dragonflies of New Guinea and neighbouring islands. Part II. Descriptions of a new genus and species of Platycneminae (Agrionidae) and of new Libellulidae. Nova Guinea 17: 1-66.

LIEFTINCK, M.A. 1949. The dragonflies (Odonata) of New Guinea and neighbouring islands. Part VII. Results of the Third Archbold expedition 1938-1939 and of the Le Roux expedition 1939 to Netherlands New Guinea (II. Zygoptera). Nova Guinea (N.S.) 5: 1-271.

OPPEL, S. 2005. Odonata in the Crater Mountain Wildlife Management Area, Papua New Guinea. IDF-Report 7: 1-28.

OPPEL, S. 2006. Comparison of two Odonata communities from a natural and a modified rainforest in Papua New Guinea. International Journal of Odonatology 9: 89-102.

TSUDA, S. 2000. A distributional list of World Odonata. Private Publication, Osaka.

TOL, J. van and GASSMANN, D. 2007. Zoogeography of freshwater invertebrates of Southeast Asia, with special reference to Odonata. Pp 45-91, in: Renema, W. (ed.), Biogeography, time and place - distributions, barriers and islands. Topics in Geobiology, Vol. 29. Dordrecht (Springer). WATSON, J.A.L., THEISCHINGER, G. and ABBEY, H.M. 1991. The Australian Dragonflies: a guide to the identification, distributions and habitats of Australian Odonata. CSIRO.

WESTFALL, M.J. and MAY, M.L. 2006. Damselflies of North American. Revised Edition. Scientific Publishers, Gainesville, Florida; 503 pp.

178 Australian Entomologist, 2012, 39 (3)

TWO NEW RECORDS OF OEDASPIS LOEW SPECIES (DIPTERA: TEPHRITIDAE: TEPHRITINAE) FROM QUEENSLAND

DAVID L. HANCOCK 8/3 McPherson Close, Edge Hill, Cairns, Qld 4870

Abstract

Oedaspis escheri (Bezzi) and O. perkinsi Hardy & Drew are newly recorded from southeastern Queensland.

Introduction

Hardy and Drew (1996) recorded only two species of the tephritine genus Oedaspis Loew from Queensland, viz. O. goodenia Hardy & Drew and O. mouldsi Hardy & Drew. This genus belongs to a group of flies (subtribe Platensinina) that is known to form stem galls on various species of Asteraceae, Goodeniaceae and Onagraceae (Hancock 2001), with Oedaspis utilising the first two families. Two additional species are recorded here, based on material in the Queensland Museum, Brisbane (QMB).

Oedaspis escheri (Bezzi) QUEENSLAND: | 9, Brisbane, 18.x.1961, light trap (QMB).

Previously recorded from Western Australia, Northern Territory and New South Wales (Hardy and Drew 1996), the above specimen is the first record from Queensland.

Oedaspis perkinsi Hardy & Drew QUEENSLAND: | Q, Six Mile Ck, 27.015°S 152.977°E, 10 m, 24.ix-9.x.2010, G. Monteith, Malaise trap, euc/wallum (QMB).

The above female, the first specimen recorded since the holotype male collected in Victoria in 1859 (Hardy and Drew 1996), differs from the male in the slightly reduced hyaline areas on the wing. The hyaline band across cells r2,3 and r4,5 is reduced to an isolated spot in each cell and the spot in the distal half of cell dm is isolated and not joined with the posterior indentation in cell cu,, while a single marginal indentation in cell m is short, broad and deeply concave anteriorly. The oviscape is black and short, about as long as tergite V. The characteristic two hyaline spots in the stigma (cell sc) are present and some sexual dimorphism frequently occurs in Australian species of Oedaspis, suggesting that this SE Queensland specimen is conspecific with the male, despite the considerably extended distribution.

References

HANCOCK, D.L. 2001. Systematic notes on the genera of Australian and some non-Australian Tephritinae (Diptera: Tephritidae). Australian Entomologist 28(4): 111-116.

HARDY, D.E. and DREW, R.A.I. 1996. Revision of the Australian Tephritini (Diptera: Tephritidae). Invertebrate Taxonomy 10(2): 213-405.

Australian Entomologist, 2012, 39 (3): 179-187 179

OVIPOSITION BEHAVIOUR IN THE DART-TAILED WASP, CAMERONELLA DALLA TORRE (HYMENOPTERA: PTEROMALIDAE: COLOTRECHINAE)

A.X. WANG and L.G. COOK

The University of Queensland, School of Biological Sciences, Brisbane, Qld 4072 (E-mail: ugxwan18 @ug.edu.au)

Abstract

The first description of oviposition behaviour by a dart-tailed wasp, Cameronella Dalla Torre, 1897, is provided based on observations and a video recording of an adult female attempting to oviposit into a gall of Apiomorpha ovicola (Schrader, 1863). The oviposition behaviour of the female of Cameronella is similar to that of other pteromalids that have an expended ovipositor. Three major behaviours associated with oviposition were observed: antennation (including at the orifice of the host's gall), drilling and preening.

Introduction

Dart-tailed wasps, Cameronella Dalla Torre, 1897, are endemic to Australia and are specific parasitoids of the gall-inducing scale insect Apiomorpha Riibsaamen, 1894 (Hemiptera: Eriococcidae) (Boucek 1988). The common name of the wasp is derived from the modified epipygium of the adult female, which resembles the tail of a dart although it is more similar to the straight fletching of an arrow, in that there are only three vanes (Fig. 1). These wasps are rarely caught by hand-netting or Malaise traps and are poorly represented in collections. For example, the Australian National Insect Collection has 48 specimens, the South Australia Museum has three specimens and the Queensland Department of Agriculture, Fisheries and Forestry has only two specimens. Seven described species were listed by Boucek (1988) but he suggested that only three might be valid. Because only a few of the specimens in museum collections have been identified to species, it is difficult to identify any recently collected specimens to species.

In accord .with the rarity of specimens, the biology and ecology of Cameronella are little known. We determined that Cameronella is an ectoparasitoid, because early stage larvae were found attached externally to the cuticles of females of Apiomorpha extracted from galls (Wang et al. unpublished). The biology of the other 11 Australian genera of Colotrechninae (Boucek 1988) is even less known. Six associate with unidentified galls on Eucalyptus and Casuarina or with twig-boring beetles, while nothing is known of the other five genera (Boucek 1988).

Oviposition behaviour of pteromalids has rarely been described. Both Cheiropachus quadrum (Fabricius, 1787) and Anisopteromalus calandrae (Howard, 1881) are pteromaline parasitoids attacking larvae of small beetles (olive bark beetle and bruchid beetle). Their oviposition behaviours have been described as host searching, antennation, drilling, piercing and inserting, oviposition, preening and, sometimes, feeding on host body fluids (Fig. 2) (Carlos et al. 1999, Begum 1995). The oviposition of the well-studied genus

180 Australian Entomologist, 2012, 39 (3)

Nasonia Ashmead, 1904 (Pteromalinae) on fly puparia has also been described and, based on observations by Edwards (1954) and others (e.g. Girault and Sanders 1910, Altson 1920, Jacobi 1939), includes the same behaviours as Ch. quadrum and An. calandrae. However, all of these differ from Cameronella in that they lack extended ovipositors.

1 mm ‘

1 mm

Fig. 1. An adult female of Cameronella sp. (above) and the “dart tail” of another adult female of Cameronella (below), both from Western Australia.

Oviposition behaviour of pteromalids with extended ovipositors has been less well described. Oviposition in one fig-wasp parasitoid group, Apocrypta Coquerel (Sycoryctinae), was described by Ulenberg and Niibel (1982) and Zhen et al. (2005), who focused mainly on abdominal movements during

Australian Entomologist, 2012, 39 (3) 181

drilling. Oviposition was simply described as consisting of three phases: searching for a receptive host, penetrating the host and oviposition before withdrawing the ovipositor (Zhen et al. 2005). This is similar to the process described for other pteromalids but lacks feeding on the host. Female wasps with expended ovipositors often need to drill through thick plant tissue to get to the host. Consequently, the female is unable to reach the host with her mouthparts and cannot feed on host body fluids. Species of Apiomorpha live within tough woody galls on their eucalypt hosts and thus we expected Cameronella would have similar oviposition behaviour to that of Apocrypta.

HOST SEARCHING

HOST CONTACT ANTENNATION

4

DRILLING

y

Brest mm Acts NE >| PIERCING AND i i INSERTING PREENING "1 FEEDING ese ; OVIPOSITION

LEAVING HOST

Fig. 2. Flow chart of the oviposition behaviours of females of Cheiropachus quadrum (Pteromalidae: Pteromalinae), from locating the host and ovipositing, through to leaving the host. Solid lines indicate invariable paths and dotted lines indicate alternative pathways. (After Carlos et al. 1999).

182 Australian Entomologist, 2012, 39 (3)

Methods

All observations were carried out at The University of Queensland, St Lucia, Brisbane. The adult female of Cameronella sp. observed ovipositing was reared from Apiomorpha ovicola (Schrader, 1863) collected by P. J. Mills from Eucalyptus microcarpa (Maiden) Maiden at Dimboola, Victoria on 9 July 2011. The female wasp emerged 53 days after the gall was collected and was kept in a plastic box and fed with honey solution.

The 7-day old virgin adult female was presented with a gall of an adult female of A. ovicola collected from Eucalyptus polyanthemos Schauer at Chiltern-Mount Pilot National Park, Victoria. The gall contained a live adult female, as indicated by the fresh wax at the apical orifice (Fig. 4). The gall was kept at room temperature (18~24°C) before being exposed to the parasitoid. Oviposition behaviour was recorded using a Canon EOS 7D, with an attached 100 mm macro lens, as 1080p high definition video under fluorescent lighting.

In 2011, a soft, green gall of A. ovicoloides was collected (LGC, 8 September 2011, Higginsville, Western Australia) that contained a developing larva of Cameronella sp. A three-dimensional model of the wall of one quarter of the gall was constructed using sliced images with 3DMed software (http://www.mitk.net). The gall tissue was cut into 18 slices, each about 0.5 mm thick, by hand with a scalpel and both sides of each slice were photographed using an Olympus Stereo Microscope. Dark brown tracks (presumably produced by oviposition) were visible against the light yellow tissue of the gall wall. Thirty-six images of sliced gall were used to construct a 3D image in 3DMed using a “Z distance” of 14 mm. Images were enhanced for contrast and colour was inverted to show bright traces against a dark background. A stereoscopic 3D image was captured from two angles of the 3D model and aligned for parallel 3D viewing using Photoshop.

Results

After being released directly onto the gall, the wasp started antennation by walking over the gall and rapidly tapping its surface with the tips of her antennae (Fig. 3). Each time she approached the apical opening of the gall (Fig. 4) she stayed and tapped around the opening for about 5 sec (Fig. 5). Occasional preening behaviours were observed during the antennation phase. This apparent “investigating” behaviour lasted between 100-180 sec, until the female stopped midway along the gall and began drilling (Figs 6-9).

Prior to drilling, the female stopped tapping the gall with her antennal tips and moved forward such that, when the abdomen was raised and the ovipositor was placed against the gall tissue (by folding down at the junction of the first and second tergite), the tip of the ovipositor was at the place where the female had been tapping with her antennae (Fig. 6). When the tip of the ovipositor contacted the gall (Fig. 7), she separated the ovipositor

Australian Entomologist, 2012, 39 (3) 183

sheaths and the “dart tail” by about 30° laterally to detach the ovipositor (Fig. 8) and started to drill.

Drilling into the gall tissue appeared to involve two different movement patterns: a horizontal swinging of the abdomen combined with rapid vibration of the “dart tail” and a vertical movement. The first horizontal swinging movement consisted of a slight swing of the abdomen from side to side across an arc of about 20° at a frequency of about once every 3 sec, combined with rapid shaking of the “dart tail”. After approximately 80 sec, vertical movement was added by vibrating the abdomen in the vertical plane as the legs bend and straighten to move up and down at a frequency of 2-3 times per sec. The combined movement, both horizontal and vertical, lasted about 90 sec. After that, the wasp stopped drilling for 5 seconds and then removed the ovipositor from the gall tissue by pulling it upwards (Fig. 9) and lifting it to replace it in the ovipositor sheaths (Fig. 10).

Figs 3-12. Oviposition behaviour of Cameronella sp.: (3) antennation; (4) apical orifice of the gall of A. ovicola showing wax produced by the female inside; (5) focused antennation at the apical orifice of the host; (6-8) start of drilling; (9) removing the ovipositor; (10-12) preening behaviour using hind-leg and ovipositor sheaths.

184 Australian Entomologist, 2012, 39 (3)

Preening behaviour was observed after the wasp removed its ovipositor from the gall tissue. The hind legs were used to brush the ovipositor (Fig. 10) and plant material adhering to the tip of the ovipositor was detached by moving the ovipositor in and out of the sheaths (Figs 11-12). After 3 mins resting and cleaning, the wasp walked away from the gall. The observed sequence of behaviours is summarised in Fig. 13.

HOST SEARCHING -------- ANTENNATION

------>

Fig. 13. Ovipositon behaviour pattern of Cameronella sp.: solid lines indicate the sequence of behaviours described for the last observed attempt (see text), whereas dashed lines indicate alternatives. The question mark indicates a path not observed in

this study.

Australian Entomologist, 2012, 39 (3) 185

Four separate attempts at drilling into the gall tissue were observed. The first three each lasted no longer than 30 sec and only used the swing movement followed by further antennation, whereas the fourth attempt lasted about 170 sec. The holes drilled by the wasp during oviposition were about the same diameter as the ovipositor. It is unlikely that the wasp laid an egg during any of the four attempts because the depth to which the ovipositor penetrated was less than the thickness of the gall wall. When the gall was later opened, the scale insect was still active and showed no sign of immobilising venom

having been injected. No apparent damage or other parasitoids were found on the scale insect.

The quarter of the wall of the gall of Apiomorpha that housed a developing larva of Cameronella showed signs of four attempts at drilling (Fig. 14). In two of these, the ovipositor apparently did not penetrate all the way though the gall wall, whereas in the other two attempts the ovipositor appeared to reach the inner chamber of the gall, or close to it.

Discussion

This is the first description of oviposition behaviour in Cameronella and in Colotrechinae. Compared with other pteromalids, the oviposition behaviour of Cameronella is more similar to the fig-wasp parasitoid Apocrypta than to Nasonia, in that the former two genera have not been observed to feed on host body fluids whereas females of Nasonia feed on haemolymph that exudes from the oviposition wound site. Here, the female of Cameronella did not appear to pierce the host but it is unlikely that she could feed on haemolymph given that the host is inside a gall. Most species of Apiomorpha are associated with ants (Gullan 1998) that, according to our observations, can stimulate the female scale insect to secrete honeydew by tapping it with their antennae. Honeydew can also be elicited from Apiomorpha by tickling the female with a human hair. Cameronella might also elicit honeydew production. by tapping using their antennae, explaining the prolonged antennation at the apical orifice, but this has not been observed by the authors. Alternatively, prolonged antennation might assist the wasp in detecting the status of the potential host, for example whether it is alive and/or already parasitised. Adults of Cameronella likely feed on the nectar of flowers, given that a female has been netted on eucalypt flowers (collection details of E. Exley on a pin-mounted specimen in QM). Further observations

are needed to test the idea that Cameronella might also feed on host honeydew.

The oviposition attempts reported here might have been unsuccessful because the gall of the Apiomorpha used for trials had become dry and hard after being picked from the tree in Victoria and transported to Brisbane, Queensland. However, the failed attempts observed in this study are apparently not rare in the field. The field-collected gall containing a developing larva of Cameronella showed several failed oviposition traces in

186 Australian Entomologist, 2012, 39 (3)

the soft walls of the gall (Fig. 14). The thickness and hardness of the gall wall could vary at different locations and it changes through the development of the gall-inducing scale insect. It is possible that females of Cameronella need several attempts to find a satisfactory drilling location.

Fig. 14. Stereoscopic 3D image (parallel view) of the wall of a quarter of an Apiomorpha gall that housed a developing larva of Cameronella showing oil glands in the gall wall (bright dots) and oviposition traces (straight aligned dots). Four Oviposition attempts can be observed in this part of the gall. Arrows indicate where the ovipositor appeared to reach the inner chamber of the gall, or close to it. Arrowheads indicate attempts that apparently did not penetrate all the way though the gall wall. (R: right eye view; L: left eye view; upper images: hind view; bottom images: lateral view).

Supplementary material Videos of Cameronella oviposition have been edited and uploaded to http://vimeo.com/28772701 under Creative Common license of Attribution-

ShareAlike 3.0 Unported (CC BY-SA 3.0).

Australian Entomologist, 2012, 39 (3) 187

Acknowledgements

We are grateful to Penelope J. Mills for collecting many specimens of Cameronella. This project was supported by a RHD scholarship from the School of Biological Sciences, The University of Queensland, to.Andy X. Wang, ARC Discovery Projects funding to Lyn G. Cook and ABRS funding to Lyn G. Cook and Penelope J. Gullan. We thank DEC WA and DSE VIC for permits to collect scale insect galls. The manuscript was improved with suggestions from Chris Burwell and John La Salle.

References

ALTSON, A.M. 1920. The life history and habits of two parasites of blow-flies. Proceedings of the Zoological Society of London 1920: 195-243.

BEGUM, S. 1995. Mating and oviposition behaviour of Anisopteromalus calandrae (Howard) (Hymenoptera: Pteromalidae). Bangladesh Journal of Zoology 23: 29-34.

BOUCEK, Z. 1988. Australasian Chalcidoidea (Hymenoptera). A biosystematic revision of genera of fourteen families, with a reclassification of species. CAB International, Wallingford; 832 pp.

COMSTOCK, P. 1881. Report of the entomologist. Report of the United States Department of Agriculture 1880: 273.

EDWARDS, R.L. 1954. The host-finding and oviposition behaviour of Mormoniella vitripennis (Walker) (Hym., Pteromalidae), a parasite of muscoid flies. Behaviour 7: 88-112.

GIRAULT, A.A. and SANDERS, G.E. 1910. The chalcidoid parasites of the common house or typhoid fly (Musca domestica Linn.) and its allies. Psyche, Cambridge, Massachusetts 17: 9-28.

GULLAN, P.J. 1984. A revision of the gall-forming coccoid genus Apiomorpha Riibsaamen (Homoptera: Eriococcidae: Apiomorphinae). Australian Journal of Zoology Supplementary Series 99; 1-203.

JACOBI, E.F. 1939. Uber Lebensweise, auffinden des Wirtes und Regulierung der Individuenzahl von Mormoniella vitripennis Walker. Archives Neerlandaises de Zoologie, Leiden 3: 197-282.

LOZANO, C., GUERRA, O.A. and CAMPOS, M. 1999. Host-finding and oviposition behaviour of Cheiropachus quadrum (F.) (Hym.: Pteromalidae), a parasite of olive bark beetles (Col.: Scolytidae). Mitteilungen der Schweizerischen Entomologischen Gesellschaft 72: 89-93. SCHRADER, H.L. 1863. Observations on certain gall-making Coccidae of Australia. Transactions of the Entomological Society of New South Wales 1: 1-6.

ZHEN, W.Q., HUANG, D.W., XIAO, J.H., YANG, D.R., ZHU, C.D. and XIAO, H. 2005. Ovipositor length of three Apocrypta species: Effect on oviposition behavior and correlation with syconial thickness. Phytoparasitica 33(2): 113-120.

188 Australian Entomologist, 2012, 39 (3)

A REPLACEMENT NAME AND NEW COMBINATION FOR LAPHRIA NIGROCAERULEA KIRBY, 1889 (DIPTERA: ASILIDAE: LAPHRIINAE)

GREG DANIELS

School of Biological Sciences, University of Queensland, St Lucia, Brisbane, Qld 4072

Abstract

Laphria nigrocerulea Kirby, 1889 is found to be a junior homonym of Laphria nigrocaerulea van der Wulp, 1872 and is replaced by Laphria christmasensis nom. n. This species is also transferred to the new combination Orthogonis christmasensis (Daniels).

Discussion Laphria nigrocerulea Kirby, 1889, p. 555, from the Australian Territory of

Christmas Island in the Indian Ocean, is a junior homonym of Laphria nigrocaerulea van der Wulp, 1872, p. 194, from New Guinea. A new name Laphria christmasensis nom. n. is therefore proposed as a replacement name

for Laphria nigrocerulea Kirby, 1889.

Apart from Kirby’s publication, Hull (1962: 323) appears to be the only other author to refer to Kirby’s species. He records it from Oceania, a region that excludes the Indian Ocean. In the ‘Catalog of Diptera of the Oriental Region’, Oldroyd (1975: 115) refers only to van der Wulp’s species from New Guinea (as Orthogonis nigrocaerulea), even though Christmas Island falls within the scope of the catalogue.

The structure of the terminalia of specimens of L. nigrocerulea Kirby from Christmas Island in the author’s collection is typical of genus Orthogonis Hermann and the species is hereby transferred to that genus, as Orthogonis

christmasensis (Daniels), comb. n.

References HULL, F.M. 1962. Robber flies of the World. The genera of the family Asilidae. Bulletin of the

United States National Museum 224: 1-907.

KIRBY, W.F. 1889. On the insects (exclusive of Coleoptera and Lepidoptera) of Christmas Island. Proceedings of the Zoological Society of London 1888: 546-555.

OLDROYD, H. 1975. Family Asilidae. Pp 99-156, in: Delfinado, M.D. and Hardy, D.E. (eds.), Catalog of Diptera of the Oriental Region. Vol. 2. University Press of Hawaii, Honolulu; [ix] + 459 pp.

van der WULP, F.M. 1872. Bijdrage tot de kennis der Asiliden van den Oost-Indischen Archipel. Tijdschrift voor Entomologie 15: 129-279, pls. 9-12.

Australian Entomologist, 2012, 39 (3): 189-194 189

NEXT GENERATION INSECT LIGHT TRAPS: THE USE OF LED LIGHT TECHNOLOGY IN SAMPLING EMERGING AQUATIC MACROINVERTEBRATES

DOUGLAS GREEN, DUNCAN MACKAY and MOLLY WHALEN

School of Biological Sciences, Flinders University, GPO Box 2100, Adelaide, SA 5001 (Email: douglas.green@ flinders.edu.au)

Abstract

LED lights were trialled as a replacement for traditional fluorescent bulbs for catching emerging aquatic macroinvertebrates. Initial trials with white LEDs were disappointing, with the catch amounting to chance contact with the trap, but when ultraviolet LEDs were used, there was no significant difference from the traditional fluorescent trap of the same design. While the fluorescent trap used most or all of the available battery power, the LED lights used less than 10% of the available power. It is suggested that LEDs can be used to replace the more power- demanding traditional lights for use in light traps.

Introduction

Light traps have long been a popular choice for baseline surveys of winged invertebrates from mosquitos to moths and there have been many variations on light trap designs over the years. While their use in urban environments is facilitated by the availability of close power sources, field use has always been limited by the requirement of power to run traditional lights. Traditional fluorescent tubes often do not run for more than 12 hours from a traditional 12-volt power source such as a car battery.

Light traps have been used for insect trapping for over 100 years. In that time there have been many variations in design with some being extremely complex, involving both lights and fans (Venter et al. 2009), while others have remained simple (Scanlon and Petit 2008). The source of light has also varied, beginning with flames and moving on to incandescent bulbs and, in more recent times, fluorescent tubes. Most current traps employ either an incandescent bulb or actinic fluorescent tube as the light source, as the spectrum of light emitted from these bulbs is effective for attracting insects (Sambaraju and Phillips 2008). However, the power used by these light sources has always been an issue. Typically, small bulbs of around 6-9 watts are used which require either a fixed power source or a large power supply to power the light for an entire night. A common power source used is a 12-volt battery which will power such lights for approximately 6-8 hours, depending on the amp-hours of the battery. Given that the flight period of different insects varies from dusk until dawn, this means that standard light sources

may fail to attract a portion of the available insect population (Williams 1935, Scalercio et al. 2009).

Over the last decade, light-emitting diodes (LEDs) have become increasingly popular as a replacement for standard incandescent bulbs or fluorescent bulbs as they are cheaper, run cooler, are more resistant to damage and use

190 Australian Entomologist, 2012, 39 (3)

considerably less power. LEDs are also a much more focused light source with a narrow spectrum of light (generally 5 nanometres) and either a narrow beam (generally 25 degrees) or wide beam (Moreno and Sun 2008). This allows for specific lighting characteristics to be selected and tailored for a specific purpose. Previous work has indicated that the use of LEDs increased capture rates of sandflies by 50% (Cohnstaedt et al. 2008); however, the effectiveness of LEDs in attracting other types of insects has been little investigated. The purpose of this study was to examine whether LEDs could be used as a substitute for an actinic fluorescent bulb in a conventional light trap, and to examine the effect of this substitution on capture rates of emerging aquatic macroinvertebrates.

Methods For this study three different lights were trialled. All light sources used were

attached to a “heath” style trap that employs three transparent upright vanes radiating out from a central point and light source. The vanes sit over a vertical funnel leading into a chamber where the insects are trapped until collection. In order to keep the trap stable under windy conditions the vanes were anchored to a stake. All lights were attached to an 18 amp-hour 12-volt battery (5-in-1 Power station/Jump starter (MB-3594), PowerTech). The first light source trialled was a commercially available 8 watt actinic fluorescent bulb (E700, Australian Entomological Supplies Pty. Ltd, Australia). The second was two banks of four white LEDs (6500 nm, 3000 millicandela), and the third was two banks of nine 2000 millicandela ‘UV/black light (395 nm)’

LEDs (Fig. 1).

These traps were trialled in the Sturt River Gorge, South Australia, from 5-8 December 2011. Given the documented variation in catch due to weather conditions (Williams 1940, Yela and Holyoak 1997) and moonlight (Bowden and Church 1973, Yela and Holyoak 1997), these details were recorded. Two of the actinic fluorescent light type and two of the UV LED light trap were trialled over four consecutive nights. The traps were placed alongside pools separated by a minimum of 50 meters and at least one riffle section (Fig. 2). No other trap was visible from the trap location. The LED light traps were always directed towards the water, facing the steep side of the river valley. Traps were set at 8pm and collected at 7am.

Collected individuals were identified to Order using the CSIRO online invertebrate key (CSIRO 2011). In order to rule out any effect of sampling date on the results a one-way ANOVA was used. Differences between the samples collected by the different styles of trap were analysed using a series of independent samples t-tests for total number of individuals sampled per trap, total orders sampled per trap and the number of each order sampled per trap, treating the nightly catches as replicates. All statistical analysis was performed in IBM SPSS Statistics (Version 19).

Australian Entomologist, 2012, 39 (3) 19]

Fig. 1. Constructed light trap showing banks of LEDs and general set-up of upright clear vanes positioned over a funnel.

sa wa m a d ATA ty ett

Fig. 2. Sampling sites used for trialling the light traps in the Sturt River Gorge, South Australia. Site a: 35°2'58.49"S, 138°36'25.96"E. Site b: 35°2'57.18"S, 138°36'27.73"E. Site c: 35°2'58.49"S, 138°36'30.52"E. Site d: 35°3'0.69"S, 138°36'32.77"E.

192 Australian Entomologist, 2012, 39 (3)

Total Individuals Caught

New LED

Actinic Fluro Trap type

Fig. 3. Box plot of total individuals caught in the different styles of trap per night generated using IBM SPSS Statistics Version 19 (8 replicates). Bars represent minimum and maximum number of individuals caught per night, the middle bar

represents the median.

Trap_type

Actinic Muy LED

=

100

Mean Individuals Caught per Night

Trichoptera Coleoptera Lepidoptera Diptera Ephemeroptera

Order

Fig. 4. Mean and error bar plot (+/- 1 standard error, 8 replicates) of the five most abundant orders caught in both UV LED and Actinic light traps (generated using IBM

SPSS Statistics Version 19).

Australian Entomologist, 2012, 39 (3) 193

Results and discussion

The weather conditions varied little over the sampling period. There was light cloud cover ranging from 10-20% on each of the sampling nights. The moon phase was day 11 through 15. The wind direction and speed varied from night to night; however, due to the location of the trapping site, a well vegetated river gorge, the effect of wind was likely minimal. There was no significant effect of sampling date on the invertebrates caught shown by the one-way ANOVA conducted for the total number of individuals caught, as well as on each individual order (all results p>0.05).

The White LED light traps were relatively ineffective, with the insect catch apparently amounting to no more than incidental collision with the clear vanes (total 7 individuals) and were discarded after the first two nights. Therefore, we focused on comparing the UV LED traps and the actinic fluorescent trap. The results indicated that there was little difference between the catch from either trap type. The most commonly caught insects were Trichoptera, followed by Coleoptera (Fig. 4). When looking at the total insect abundance, there were on average slightly fewer individuals caught in the UV LED traps; however, this difference was not significant (Fig. 3, t=0.490, df=13.982, p=0.631). Independent samples t-tests were also done on individual orders to see if there was an order specific difference in the sample. There was a trend towards more Lepidoptera and Diptera in the actinic light traps; however, this was found to be not significant using an independent samples t-test for the four replicates (p>0.05). It is possible that these results are related to the 360 degree spread of light from the actinic bulb rather than the 120 degree spread of light from the UV LED traps. In addition, the light from the UV LEDs was directed largely over the water body, rather than towards the vegetation. Given that all orders trapped in this study appear to be attracted to both light sources, we hypothesise that, given a full 360 degree spread of light (achieved by adding more LEDs or modifying the arrangement of the LEDs), the results may have been more similar.

Power consumption was measured using the inbuilt voltmeter on the jump starter battery packs and analysed using an independent samples t-test. The power consumption significantly differed between the two trap types as expected (t=32.16, df= 8.84, p<0.00, n=4). While running off 18 amp-hour batteries the LED light traps used, on average, less than 10% of the available power while the actinic fluoro used, on average, 92.5% of the available power, with some trials using 100%. This may have led to discrepancies

among catches as it was unclear when the battery power was exhausted for some of the fluorescent light traps.

Given the results of this study, we propose that UV LEDs may often be used in place of traditional light sources in insect light traps. LEDs can be easily retrofitted to any existing light trap and are inexpensive to buy. They are also more durable, longer lasting, more power efficient and easier to repair. The

194 Australian Entomologist, 2012, 39 (3)

LED light traps used in this study were constructed from commonly available materials for less than $60AUD each. LEDs also commonly run on 12 volts DC, which reduces the risk of electric shock to the operator as fluorescent tubes may require high voltages to start and inverters to run. This study found no significant differences in the abundance or composition of the insects caught by LED-based and fluorescent tube based light traps, even when the LEDs only illuminated 120 degrees while using less than an eighth of the power of the fluorescent lights. While we believe that UV LED light traps are a good replacement for actinic light traps, largely because of their lower power consumption and more robust design, we believe considerably more work is required to assess the relative attractiveness of LED and traditional light sources to specific insect orders.

References

BOWDEN, J. and CHURCH, B.M. 1973. The influence of moonlight on catches of insects in light-traps in Africa. Part II. The effect of moon phase on light-trap catches. Bulletin of Entomological Research 63: 129-142.

COHNSTAEDT, L., GILLEN, J.I. and MUNSTERMANN, L.E. 2008. Light-emitting diode technology improves insect trapping. Journal of the American Mosquito Control Association 24: 331-334.

CSIRO. 2011. Key to the Invertebrates [Online]. [Accessed 6 December 2011]. Available: hitp:/www.ento.csiro.au/education/key/couplet_O1.himl

MORENO, I. and SUN, C. 2008. Modeling the radiation pattern of LEDs. Optics Express 16: 1808-1819.

SAMBARAJU, K.R. and PHILLIPS, T.W. 2008. Responses of adult Plodia interpunctella (Hubner) (Lepidoptera: Pyralidae) to light and combinations of attractants and light. Journal of Insect Behaviour 21: 422-239.

SCALERCIO, S., INFUSINO, M. and WOIWOD, I.P. 2009. Optimising the sampling window for moth indicator communities. Journal of Insect Conservation 13: 583-591.

SCANLON, A.T. and PETIT, S. 2008. Biomass and biodiversity of nocternal aerial insects in an Adelaide City park and implications for bats. Urban Ecosystems 11: 91-106.

VENTER, G.J., LABUSCHAGNE, K., HERMANIDES, K.G., BOIKANYO, S.N.B., MAJATLADI, D.M. and MOREY, L. 2009. Comparison of the efficiency of five suction light traps under field conditions in South Africa for the collection of Culicoides species. Veterinary Parasitology 166: 299-307.

WILLIAMS, C.B. 1935. The times of activity of certain nocternal insects, chiefly Lepidoptera, as indicated by a light trap. Transactions of the Royal Society of London 83: 523-556. WILLIAMS, C.B. 1940. An analysis of four years capture of insects in a light trap. Part HI. The effect of weather conditions on insect activity; and the estimation and forecasting of changes in the insect population. Transactions of the Royal Society of London 90: 227-306.

YELA, J.L. and HOLYOAK, M. 1997. Effects of moonlight and meteorological factors on light and bait trap catches of noctuid moths (Lepidoptera: Noctuidae). Entomological Society of America 26: 1283-1290.

Australian Entomologist, 2012, 39 (3): 195-196 195

A NOTE ON THE IDENTITY OF ‘ACANTHONEVRA’ INERMIS HERING (DIPTERA: TEPHRITIDAE: ACANTHONEVRINI)

DAVID L. HANCOCK 8/3 McPherson Close, Edge Hill, Cairns, Qld 4870 Abstract

Rioxoptilona inermis (Hering), comb. n., described from southern India, is transferred from Acanthonevra Macquart and the female recorded for the first time. The type localities of Lumirioxa affluens (Hering), L. ornatipennis (Hering) and Rioxoptilona ochropleura (Hering) are confirmed as Kambaiti, northern Burma.

Introduction

Hancock (2011) retained the Indian fruit fly species Acanthonevra inermis Hering within that genus and placed it in a key to all known members of the Acanthonevra complex of genera as then defined. However, recent examination of the holotype male and two newly identified females (all located in the Natural History Museum, London (BMNH)), has revealed that Hering’s (1951) illustration of the wing was misinterpreted with respect to the curvature of vein R3, leading to its incorrect retention within Acanthonevra Macquatt. Its correct placement is discussed below. It should also be noted that some specimens of Ptilona conformis Zia have a narrow, longitudinal hyaline streak in cell 14,5 below the stigmal/r2,3 indentation that does not cross the cell; this should be considered when using the key.

Hancock (2011) also suggested that the type locality of Rioxoptilona ochropleura (Hering, 1951) was possibly incorrect and noted that those of Lumirioxa affluens (Hering, 1951) and L. ornatipennis (Hering, 1951) were merely recorded as ‘Burma’; more precise details are provided below.

Rioxoptilona inermis (Hering, 1951), comb. n. (Figs 1-2)

Acanthonevra inermis Hering, 1951: 5. Type locality Anamalai Hills, S India. HT 3 in BMNH; examined.

Material examined. INDIA: Holotype ĝ, Anamalai Hills, S. India, 4000-5000’, 27.1x.1946; 1 9, Naraikkadu, 2500-3000’, Tinnevelly Dist., S. India, 11-13.iii.1936; 1 9, Bababuddin Hills, Mysore, 4700’, 1.vi.1915, Ramakrishna coll. (all in BMNH).

Discussion. In the male (Fig: 1), wing vein R,3 is noticeably undulate but the tip reaches the costa at an acute angle, not almost perpendicularly as previously indicated. This vein is less undulate in the female (Fig. 2), which also has the hyaline indentations and discal spots more extensive than in the male, those near the apex of cell dm forming a broad band rather that two distinct spots. The female abdomen is medially fulvous on terga II and MI.

This species keys to couplet 46 in Hancock (2011), differing from the otherwise similar R. formosana (Enderlein) and R. setosifemora (Hardy) in having a red-brown scutum without any indication of dark longitudinal vittae. It is known only from southern India.

196 Australian Entomologist, 2012, 39 (3)

Figs 1-2. Rioxoptilona inermis (Hering), wings of (1) holotype male; (2) female (with abdomen inverted). Photos by K. Goodger © Natural History Museum, London.

Type localities

Hering (1951) recorded the type localities of Lumirioxa affluens (Hering), L. ornatipennis (Hering) and Rioxoptilona ochropleura (Hering) as ‘Burma’ without further details. Holotypes of all three species are in BMNH and carry the following locality data which, despite doubts raised by Hancock (2011), must be assumed to be correct: ‘N.E. Burma, Kambaiti, [R.] Malaise’, with additional data ‘2000 m, 4.iv.1934’ for L. affluens; ‘7000 ft, 28.iv.1934’ for L. ornatipennis; and ‘1800 m, 17.vi.1934’ for R. ochropleura.

Acknowledgement ] I thank Kim Goodger (BMNH) for the photographs and access to specimens.

References

HANCOCK, D.L. 2011. An annotated key to the species of Acanthonevra Macquart and allied genera. Australian Entomologist 38: 109-128.

HERING, E.M. 1951. Neue Fruchtfliegen der Alten Welt. Siruna Seva 7: 1-16.

Australian Entomologist, 2012, 39 (3): 197-207 r 197

FIRST RECORD OF THE BASE-BORER WEEVIL, SPARGANOBASIS SUBCRUCIATA MARSHALL (COLEOPTERA: CURCULIONIDAE: DRYOPTHORINAE), FROM OIL PALM (ELAEIS GUINEENSIS JACQ.) INPAPUA NEW GUINEA AND ITS ASSOCIATION WITH DECAYING STEM TISSUE

CHARLES F. DEWHURST and CARMEL A. PILOTTI

Papua New Guinea Oil Palm Research Association, West New Britain and Milne Bay Provinces, Papua New Guinea

(Email: charlesf:dewhurst @ pngopra.org.pg) Abstract The native base-borer weevil, Sparganobasis subcruciata Marshall, is recorded for the first time from tissues of cultivated oil palm (Elaeis guineensis Jacq.) in Papua New Guinea. Adults, larvae, pupae and damage are illustrated. Evidence suggests that attack is initiated by odours produced by fungal decay of palm tissues caused by Ganoderma boninense Pat. and secondary decomposers or by Thielaviopsis paradoxa (de Seynes) in oil palm frond axils. Introduction

In September 2010, the Technical Services Division (TSD) at Higaturu (New Britain Palm Oil, Northern (Oro) Province, Papua New Guinea (PNG), reported that oil palms suspected of being attacked by the fungus Ganoderma boninense Pat. were also infested by insect larvae. The oil palms were growing at Mamba Estate, a plantation of mature oil palms planted at high densities (143 palms/ha) and located in the Mamba Valley, on the northern side of the Kumusi River, at an elevation of about 384 m. The palms affected

were approximately 22 years old and were due for felling before replanting was undertaken.

Specimens were sent to specialists in the United Kingdom and Australia, where they were identified as Sparganobasis subcruciata Marshall by C. Lyal (Natural History Museum, London) and R. Oberprieler (CSIRO, Canberra), a weevil that appears to be endemic to the island of New Guinea and known as the base-borer weevil (Froggatt 1936). It was originally described from three specimens collected at the Utakwa River in the Sudirman (Snow) Mountains in ‘SW Papua’ (Irian Jaya, now West Papua). A further four specimens (3 34, 1 Q) were from Andai [South of Manokwari, Doberai Peninsula] and Sele [NW Birdshead Peninsula] in the former Dutch New Guinea (now West Papua) and Batchian and Misol Islands in the Moluccas (Maluku) (all now part of Indonesia). They are housed in the Pascoe collection in the Natural History Museum, London (Marshall 1915).

Sparganobasis subcruciata was first reported as a pest of coconut palms by Simmonds (1925) in Madang Province of mainland PNG, where the feeding activity of the weevils eventually caused the coconut palms to collapse. The first documented record of this weevil from cultivated plantation oil palms (Elaeis guineensis Jacq.) is provided here.

198 Australian Entomologist, 2012, 39 (3)

Voucher specimens collected at Mamba Estate are deposited in the PNGOPRA reference collection, the National Insect Collection (NIC) in Port Moresby, Papua New Guinea and CSIRO in Canberra, Australia. Specimens are illustrated on the Museum Victoria Pests and Diseases Image Library

(PaDIL) website.

Morphology and biology Adults of S. subcruciata are variable in size, with a mean length of 21 mm.

Without the rostrum, both sexes are about 16.5 mm long (n = 12), males averaging 15.5 mm and females 17.5 mm. When palm trunks were cut open, all except the egg stage of S. subcruciata were found, with many of the adults in a teneral condition (recently eclosed from the pupal cell), paler in colour but with a more clearly defined dorsal pattern. Adults are also variable in colour and intensity, varying from reddish with paler markings to black with little obvious markings on the elytra. Well marked adults may be recognised by the broad, pale markings on the pronotum and the broad, pale diagonal bands converging at the centre line of each elytron, contrasting against the

darker background (Fig. 1).

The elytra are oblong-ovate and broadest at the shoulders, with raised longitudinal carinae. The lateral edges of the abdomen, legs and rostrum are densely covered with small punctures (punctate), clearly visible in lateral view (Fig. 2). These markings are much less obvious on darker specimens, except when viewed through a 10x hand lens. The prothorax is longer than broad and the head and prothorax are densely covered with circular, pale punctures. The wings are sooty-coloured and well developed (Fig. 3).

The female has a more obviously curved rostrum than the male and its basal part is covered with larger punctures. The distal part lacks punctures, while the rostrum of the male has similar, smaller punctures throughout its entire length. The legs are black and pustulate and the tibiae possess sharp terminal spines that enable the beetle to retain a firm grip on the substrate (Marshall 1915). Males are smaller than females and the sexes may also be distinguished by differences at the distal part of the abdomen, which in ventral view is wavy in outline and slightly angular in males but dull and rounded in females (R. Oberprieler pers. comm.).

Although fully winged, S. subcruciata adults were not observed to fly during the day and are probably crepuscular or nocturnal, as was reported by Froggatt (1936) for the banana weevil, Cosmopolites sordidus Chevrolat, in Australia.

Two samples of S. subcruciata adults (70 in total) from Mamba Plantation yielded 28 males and 42 females (sex ratio 1:1.5). Adults were also recently collected by one of us (CP, in 2011) from G. boninense infected oil palm at Milne Bay Estates, Milne Bay Province, Papua New Guinea.

Australian Entomologist, 2012, 39 (3) 3 199

Figs 1-2. Sparganobasis subcruciata. (1) dorsal views of adult male and female; (2) lateral view of adult male. Photos: Bill Page, PNGOPRA.

200 Australian Entomologist, 2012, 39 (3)

Larvae, pupae and adults were collected from the Mamba Estate plantation in September and November 2010. Immature stages were abundant among the tissues of the lower part of the trunk up to about 2 m above the ground and were concentrated in the outer tissues of the trunk beneath bark (Fig. 4).

Figs 3-5. Sparganobasis subcruciata. (3) adult showing extended wing; (4) larvae in situ beneath bark; (5) larval head capsule. Photos 3 & 5: Bill Page (PNGOPRA).

Australian Entomologist, 2012, 39 (3) f 201

Larvae of S. subcruciata are apodous, clearly segmented and with a noticeably setose, chestnut-brown head capsule with an inverted Y-shaped epicranial suture on the frons, between the arms of which is a raised area with two large, lateral pits (Fig. 5). They are similar to, although smaller than, those of the cane weevil borer, Rhabdoscelus obscurus (Boisduval), adults of which are commonly found on freshly cut frond bases together with other species of Dryopthoridae such as the lesser coconut weevil, Diocalandra frumenti (Fabricius) (Fig. 6). Once removed from palm wood, the larvae of S.

subcruciata are immobile except for the rhythmic pulsations of the entire body.

Sparganobasis subcruciata larvae were found among the outer tissues of the trunk, below the fibrous outer layer, with evidence (from a larva found with rot tissue) that they entered from the frond basal area, where organic detritus collects. Lever (1969) similarly reported larvae of S. subcruciata tunnelling into the trunk of a coconut palm, from ‘the point of junction of a leaf petiole’. Larvae live in well defined tunnels among the pale living tissue; however, no larvae were collected from dead, dark brown palm trunk tissue. From one collapsed and rotten palm, larvae and cocoons of the much larger black palm weevil, Rhynchophorus bilineatus Montrouzier, were also found.

Pupal cells, made from chewed palm wood tissue and lined with a smooth, light brown coating, were found in the larval tunnel. The head of the pupa was orientated towards the outside of the palm and the tunnel was plugged with palm fibre, permitting the emerging beetle to exit to the exterior of the palm (Fig. 7). The pupa is ca 21 mm long, pale cream in colour and sparsely spinose, particularly at both anterior and posterior ends. Pupae were very active when disturbed, making vigorous circular movements of the abdominal segments that caused them to rotate rapidly in the cell (Fig. 8). Pupae become darker as they near eclosion.

External evidence for the presence of the weevil was not obvious; however, close inspection of the palms, especially those that lacked old frond bases (i.e. with the trunk quite smooth), revealed signs of weevil presence in the form of small patches of oozing sap and sawdust from circular depressions, often plugged with vascular tissue, indicating the exit holes (Fig. 9), a feature also observed by Lever (1969).

Attraction to rotting wood:

As all but one of the S. subcruciata samples were collected from palms that had been attacked by the fungus G. boninense, the presence of an attractant was assumed. The nature of the attractant odour(s) is unclear. Odours are produced by both T. paradoxa and additional invasive organisms that cause a

secondary, ‘soft rot’, which is commonly seen in G. boninense-infected palms in advanced stages of the disease.

202 Australian Entomologist, 2012, 39 (3)

Figs 6-7. (6) adults of Rhabdoscelus obscurus and Diocalandra frumenti on oil palm cut frond base. (7) larva of Sparganobasis subcruciata in tunnel in oil palm trunk, with head to left.

Australian Entomologist, 2012, 39 (3) 4 203

‘igs 8-12. (8) pupae of S. subcruciata, ventral and lateral views. (9) cut bark showing exit holes. (10) two round rot patches. (11) dead palm tissue with many larvae and a banded millipede. (12) section of outer bark removed to show damage penetration.

204 Australian Entomologist, 2012, 39 (3)

A simple attraction trial was carried out at Mamba Estate. Ten adult weevils were placed at the centre of a large (52 cm diameter), blue plastic bowl and two small pieces of either fresh oil palm wood fragments or fragments with ‘rotted’ wood were added at opposite sides. They were left undisturbed overnight. The following morning, nine weevils had moved to the ‘rotted’ tissues and a single beetle remained at the centre of the bowl.

Since the fungus T. paradoxa was isolated from wood tissue in which larval tunnels of S. subcruciata were found, a similar experiment using T. paradoxa-inoculated oil palm wood and four weevils collected in Milne Bay Province was undertaken, with similar results. Interestingly, the rotting frond bases from which larval tunnels originated showed clear evidence of a wet rot that might have been caused by the fungus 7. paradoxa, as indicated by the coloration and odour of the tissue. Species of the teleomorph Ceratocystis produce volatile compounds (Hanssen 1993), which might be attractants for S. subcruciata. We believe that odours that are produced during the breakdown of the oil palm tissues by secondary invasion of other micro- organisms (e.g. yeasts or bacteria) might also be involved in attracting the weevils. It is not clear which of these odours is the primary attractant.

Damage

Fig. 10 shows a rot, possibly initiated by T. paradoxa that is often found in association with decay by G. boninense. As the development of an infestation of weevil larvae progresses, damage to the vascular tissue may become almost total (Fig. 11). In three palms where larvae were found, their trunks had snapped and the palms had collapsed. Close inspection after crude dissection of one of these palms confirmed that the infestation was caused by larvae of S. subcruciata.

Numbers of a brightly coloured banded millipede (Family Platyrhacidae: H Enghoff in litt.) were found on the exposed and decaying tissues of collapsed and broken palm trunks assisting with the breakdown process.

Nine palms with signs of infection by G. boninense (‘suspect palms’) and two palms that showed no outward signs of infection were felled and crudely dissected. The ‘suspect palms’ contained varying levels of weevil infestation, with adults, larvae and pupae present in the tissues. One of the latter two palms also contained S. subcruciata larvae in tissues near its periphery; the other did not contain larvae. Damage was concentrated in the lower part of the palm, to a height of about 2 m above the root base. Also examined was palm tissue that had been cut out of one host palm in July 2010 (4 months previously); although dry and friable, the tissues were still largely intact and no larvae, pupae or adult weevils were found.

Larval damage to the tissues spread from the outer tissues towards the centre of the palm (Fig.12). Heavy infestations were detectable by the presence of

Australian Entomologist, 2012, 39 (3) . 205

emergence holes and holes blocked with vascular tissues that are readily visible, being particularly obvious when the outer bark tissue is removed.

Monitoring and control

Larvae were heard feeding inside palms by Mamba Estate plantation workers on two occasions during these investigations. Sound production by the larvae of weevil species feeding within the trunk tissues was reported by Froggatt (1936) and was also investigated by Al-Manie and Alkanhal (2005) in Saudi Arabia, using ultra-sound recorders for larvae of Rhynchophorus ferrugineus (Olivier) in date palms (Phoenix dactylifera L.). A medical stethoscope was used at Mamba Estate, to listen for the larvae/pupae of S$. subcruciata; however, the hirsute nature of the palm surface caused too much background interference and no definite sound of larval/pupal activity was detected.

As the weevil appears to be closely linked to the presence of what may now be called ‘frond-base rot’ (FBR) and G. boninense-induced rot, treating a palm to kill the weevils at this stage will be too late to save the infected palm. Once G. boninense is established in a palm, that palm will eventually die without fungicidal intervention.

One palm showing symptoms of a G. boninense infection was injected with 90ml of glyphosate [Roundup™] and killed. Four months later it was felled and although there was no sign of G. boninense, a thriving population of weevil larvae was found among the tissues, suggesting weevils as the cause. The injection of glyphosate did not appear to affect weevil development, at least while the tissues remained firm. There is currently no evidence to suggest that the weevil is a vector of G. boninense as the fungus was not isolated from any weevils subsequently screened.

There is no monitoring system presently available for this weevil; however, the following options should be investigated using traps to monitor adult populations: (1), using natural attractant material from G. boninense or T. paradoxa infected tissue, as indicated by the Milne Bay trial; (2), development of synthetic S. subcruciata attractants based on the above chemicals; (3), identification and synthesis of a pheromone produced by S. subcruciata for use in traps.

Options for the control of weevil larvae in G. boninense-infected palms include: (1), timely application of recommended control procedures for G. boninense-induced basal stem rot should prevent secondary rots from developing, thereby reducing the likelihood for attraction of S. subcruciata; (2), reducing palm planting densities will result in palms with shallower frond bases, (3), if infestations of either G. boninense or S. subcruciata are identified, then emerging adults, larvae and pupae may be killed by following PNGOPRA recommendations (Pilotti 2006) and spreading the chipped wood out to dry before burying the chips.

206 Australian Entomologist, 2012, 39 (3)

Laboratory observations of larval and pupal weights

One hundred and five larvae were collected from Mamba Estate in September 2010 and 27 live larvae were subsequently weighed in the laboratory (78 died in transit between Mamba Estate and PNGOPRA office at Higaturu). The mean weight of the live larvae was 0.81 g (sd = 0.22). Among them was a cohort of much lighter (younger) larvae, weighing between 0.2-0.3 g. (Fig. 13), indicating the development of a new generation.

A small sample of three pupae was also collected in September 2010, which had a mean weight of 0.64 g (sd = 0.12). This was almost double that of the 21 pupae collected in November 2010, which had a mean weight of 0.37 g (sd = 0.10).

S.subcruciata : September 2010

No. larvae D

À 7 Loe

2-30 31-4 41-5 .51-.6 61-7 .71-.8 81-9 .91-1.0 1.01- 1.11- 1.21- 1.1 1.2 ik)

Weight class (gm)

Fig. 13. Live weights of 27 S. subcruciata larvae collected at Mamba Estate in September 2010.

Discussion

These observations suggest that S. subcruciata poses a potential threat to oil palms, especially in areas where high density planting and high rainfall results in long frond bases remaining on palms. These are typically produced in light-restricted valleys (W. Griffiths-NBPOL pers. comm.). In such high rainfall areas, organic matter and rainwater collect in the frond bases, which encourage the development of ‘frond-base rot’ and subsequent invasion by saprophytic fungi, including T. paradoxa. Chemical emissions from tissue breakdown caused by T. paradoxa, as well as other micro-organisms in the frond bases of oil palms, are the most probable sources of attractants for adult weevils. It is currently unknown if the initial attack by S. subcruciata was

Australian Entomologist, 2012, 39 (3) l 207

direct or was triggered by odours associated with T. paradoxa (as a result of the decay of wet organic detritus accumulating in the frond bases), G. boninense or a combination of factors.

Although the development cycle of S. subcruciata is still unclear, results indicate a clear temporal change in the phenology of larval populations, as younger (not measured) larvae were found in November 2010. Traps using natural or synthetic derivatives of the fungus T. paradoxa, or simply rotting G. boninense-infected wood, should be tested, while the possibility of extracting a pheromone from the weevils should also be investigated.

Acknowledgements

We thank Seno Nyaure (PNGOPRA, Entomology), TSD Manager Brian Gurisa (NBPOL) and Leo Guro (Plantation Manager, Mamba Estate) for their assistance in the field. The Sister-in-charge, Kokoda Hospital, is thanked for the loan of a stethoscope. Some locality details gathered from The Papua Insects Foundation gazetteer . Bill Page, Drs Rolf Oberprieler, Andrew Mitchell and Dale Smith are sincerely thanked for their thorough review of the manuscript and constructive comments. Bill Page is