ENTOMOLOGIST'S MONTHLY MAGAZINE 225 TAXONOMY AND BIONOMICS OF THE VESPA TROPICA GROUP (HYM., VESPINAE) BY MICHAEL E. ARCHER of Vespa tropica (Linnaeus, 1758) is a common and widespread hornet south-eastern Asia which until recently (Kojima, 1982) was con- sidered a single species. This paper reports on taxonomic, distributional and bionomic studies of Vespa tropica resulting in the recognition of three species. These three species, designated the Vespa tropica species group, can be readily recognised from other Vespa species by the elongated first gastral tergum (fig. 1 A) and the lateral apical margin of the clypeus produced as a blunt triangular projection (fig. 1 C). Both these character states are derived so the V. tropica species group is monphyletic (Archer, unpublished). This paper attempts to reduce the number of colour subspecies by eliminating polytopic subspecies and synonymising 'subspecies', which are really microgeographic races, with their macrogeographic race or subspecies. In addition informal names are proposed to replace the subspecies names. TAXONOMY The taxonomy of the Vespa tropica species group has been largely determined by Bequaert (1936) and van der Vecht (1957, 1959) although these two authors considered that the species group consisted of one species, namely V. tropica, which was divided into 15 colour subspecies. Bequaert (1936) showed that du Buysson (1904,1905) confused two V. cincta Fabricius,, 1775, viz.: V. tropica and V. afftnis Linnaeus, 1764. Bequaert (L936) selected the oldest specific name available from Sphex tropica, Linnaeus, 1758. Van der Vecht (1957) found the holotype of Sphex tropica and showed that it came from Java. Under V. tropica, Bequaert (1936) included the eight colour varieties: deusta Lepeletier, 1836, ducalis Smith, 1852, eulemoides du Buysson, 1905, esakii Sonan, 1925,, philippinensis de Saussure, 1854, pulchra du Buysson, 1905, sorror du Buysson, 1905, trisignata P€rez, 1910 and described three new colour varieties: anthracina, haematospecies under des and loochooensis. Van der Vecht (1957) raised the status of Bequaert's (1936) varieties to subspecies and described two new subspecies by separating leefmansi from tropica and trimeres from trisignata. He recognised that the subspecies unicolor Smith, 1864 was distinct from deusta, and that the variety soror was not a subspecies of V. tropica as suggested by Bequaert (1936) but a subspecies of V. mandarinia Smith, 1852. Van der Vecht (1959) introduced the new subspecies pseudosoror to replace soror for the appropriate specimens of V. tropica. 25th November, 1991 Yol.l27 (1991) ENTOMOLOGIST'S MONTHLY MAGAZINE 226 Kojima & Reyes (1984) described the subspecies cebuana but Starr (1989) found that the types of cebuana represented a colour anomaly and synonymised it with anthracina. Kojima (1982) found that the subspecies philippinensis and anthracina were sympatric in distribution and that females of philippinensis were separable from V. tropica on a structural character, and therefore raised philippinensis to specific rank. I have found that within the geographical distribution of the subspecies trimeres, the subspecies trisignata is found in two widely separated areas of the island of Ceram and south-eastern New Guinea. The more widespread subspecies trimeres has given rise to the polytopic subspecies trisignata tn at least two geographical areas. To avoid the use of the polytopic subspecies concept it is proposed that the subspecies trisignatabe a new synonym of the subspecies trimeres. Also I have found that the subspecies ducalis and pseudosoror are sympatric with the subspecies haematodes and leefmansi in North and South Vietnam and southern China but remain distinct colour forms. V. tropica should lose several colour subspecies which can be grouped together as the species V. ducalis (Table 1). The species V. tropica of Bequaert and van der Vecht can now be seen as consisting of three species (Table 1). Wilson & Brown (1953) put forward arguments that the subspecies nomenclature should not be used and the subspecies categories be replaced by a brief statement of locality or range of distribution. Such a set of informal names is proposed for the eight colour subspecies of I/. tropica (Table 1). The distribution (fig.2 A) and recognition of the eight informal colour forms can be described as follows. TABLE 1. THE SPECIES OF THE V. TROPICA SPECIES GROUP, TOGETHER WITH THE- SUBSPECIES AND INFORMAL COLOUR FORM NAMES. Subspecies Species Informal colour form name ducalis 1. ducalis to include new synonyms lo ochoo e ns is and p s e udos oror 2. pulchra to include new synonym esakii Chinese philippinensis none none 1. anthracina south Philippine 2. 3. 4. 5. 6. Andaman Indian-Chinese Malayan Sulawesi-New Britain tropica deusta eulemoides haematodes Japanese north Philippine leefmansi trimeres to include new synonym trisignata 7. tropica Javan 8. unicolor Buruan ENTOMOLOGIST'S MONTHLY MAGAZINE 227 The Indian and south Chinese colour form (: Indian-Chinese colour form) is characterised by reddish markings on the head and thorax and has the first gastral tergum black and the second gastral tergum orange. The reddish markings on the head and thorax become black in the colour form found in Indo-China, Malaysia, Sumatra and Borneo (: Malayan colour form) although the Indian-Chinese and Malayan colour forms have a wide area of overlap with intermediate colour forms from Sikkim to Thailand. In the Malayan colour form the orange of the second gastral tergum sometimes extends onto the third gastral tergum while in the colour form from the Andaman and Nicobar Islands (: Andaman colour form) all the gastral terga are orange except for the first and reddish marks appear again on the head and thorax. The Java and Bali colour form (: Javan colour form) is similar but darker than the Malayan colour form. In particular the second gastral sternum is mainly orange on the Malayan colour form but mainly black on the Javan colour form. The colour form from Sulawesi, Flores and Sumba to New Britain (: Sulawesi-New Britain colour form) has the orange of the second gastral tergum extending onto the first gastral tergum to a varying extent. Melanic colour forms are found in the north Philippine Islands (: north Philippine colour form), south Philippine and Palawan Islands (: south Philippine colour form) and the island of Buru (: Buruan colour form). The north Philippine colour form has its wings a uniform brown with a yellow tinge while the south Philippine and Buruan colour forms, which are indis- tinguishable from each other, have their wings strongly darkened in the basal half but apically are like the north Philippine colour form. Concerning V. ducalis I have found examples of the nominate subspecies from mainland China which are similar to the subspecies loochooensrs and which can, therefore, be considered a synonym of the nominate subspecies. Similarly the subspecies pseudosoror, which I have found to be present in Taiwan, Hainan island, southern China and Vietnam, can be considered as part of the subspecies ducalis and hence a synonym of the nominate subspecies. The subspecies esakii, which occurs within the geographical distribution of the subspecies pulchra may be considered a race of the subspecies pulchra and hence a new synonym of the subspecies pulchra. Thus V. ducali^t would seem to consist of two rather similar subspecies (Table 1). The two subspecies may be distinguished as follows. The central transverse markings on the first and second gastral terga of the subspecies pulchra are black, but are brown or reddish in the subspecies ducalis. The mesoscutellum is entirely yellow or reddish-brown in the subspecies ducalis but mainly black with some reddish-brown marks in the subspecies pulchra. The geographical distribution of V. ducalis is shown in tig.2 B. The two subspecies names may be replaced by the following informal colour form names (Table 2): the Chinese, Taiwan and Vietnam colour forms (: Chinese colour form) and the Japanese and Korean colour forms (: Japanese colour form). Specimens of an intermediate coloration are found in Korea and U.S.S.R. (southern Ussuriland). Bequaert (1936) indicated that the Japanese colour form extended into northern China. The distribution of V. philippinensis (fig. 2 C) is endemic to the Philippine Islands and has been recorded on Luzon, Leyte, Samar and Negros. ENTOMOLOGIST'S MONTHLY MAGAZINE 228 PTC A c tropica group. A, dorsal view of first gastral tergum; B, lateral view Fig. 1. -Vespa (PTC : pretegula carina); C, front view of clypeus (LAA : lateral apical of pronotum angle); D, ventral view of the sixth (6) and seventh (7) gastral sterna of the male. V. tropica, V. ducalis and V. philippinensis can be separated from all other species of Vespa by the following combination of characters of which L, 3 and 4 are considered derived and2 and 5 primitive (Archer, unpublished). 1. First gastral tergum long, as seen from above, about half as long or longer than wide (fig. 1A). 2. Pretegula carina complete (fig. 1 B). 3. Lateral apical margin of the clypeus produced as a blunt triangular projection, greater than a semicircle (fig. 1 C). In addition the following male characters should also be used. 4. Apical margin of the seventh gastral sternum slightly concave (fig. 1 D). 5. Apical margin of the sixth gastral sternum with one shallow emargination, less than a semicircle (fig. 1 D). ENTOMOLOGIST'S MONTHLY MAGAZINE 229 The three species may be separated with the following key. 1 Vertex black. First three 2 gastral terga mainly or extensively orange-yellow. On the .... philippinensis male the left eye touches the clypeal margin Vertex often not black; if black then the first three gastral terga either all black or only one or two terga with orange-yellow bands or markings. On the male the left eye does ....... 2 not touch the clypeal margin Yertex usually black; if not black the second gastral tergum entirely or orange-yellow s;;;i ;;;ffi ;t;h ;;; ;; ;;;; u""0, r".i';i;;:T!-; - n;;;;; ;;, bi;;k S;;";; .. band ducalis apical orange-yellow Kojima (1982) found that the pronotum of the female in dorsal view is moderately convex laterally on V. philippinensis but less convex or often nearly straight on V. tropica and V. ducalis. NESTING BIOLOGY The nesting biology of V. ducalis is well known from a study of the Japanese colour form in southern Japan (Matsuura, 1984). Nests are found in covered aerial positions or in large cavities underground. Matsuura (1980, pers. comm.), from a sample of.72 nest sites, found 25 in attics, 7 in hollow trees, 9 in other covered aerial sites and 31 underground. Martin (pers. comm., 1990) found three nests on the Ryukyu Islands, Japan of which one was inside a hollow log, one in a small rock cave and one underground. Yamane, So (1977) found nests in Japan in hollow trees and underground and in Taiwan in attics and underground. Sonan (1929) found nests of V. ducalis in Taiwan in the gaps of the walls of dwelling houses. Over-wintering queens emerge from the end of May until early June and the nest is founded by a single queen from early to mid-June. The first workers emerge by the end of July and the sexuals by late August. The queen usually dies in August and the demise of the colony is from late September until early October. There are about four months of colonial activity and this may be considered a short cycle. The queen nest consists of a comb of up to 35-40 cells, covered by an envelope shaped like an inverted bowl with a wide ventral opening. The mature nest has two or three combs of about L70 cells although some have over 300 cells. The envelope consists of a few plate-like sheets with imperfect air pockets and its wide basal opening exposes the lowermost comb. The mean developmental period of a worker is 35-36 days (eggs 6 days, larvae l2-I3 days, sealed brood 17 days). The mean length of life of an adult worker is 35 days with most workers having died by their 50th day. The average colony probably produces about 45 workers, 40 males and 45 queens although as many as 71 workers, 6L males and 75 queens has been recorded from one colony. The maximum mean number of workers found in a colony is about 15 although up to 40 workers have been found. Thus the colonies of V. ducalis are rather small. Matsuura (1983) also reported that the Chinese colour form of V. 230 ducalis ENTOMOLOGIST'S MONTHLY MAGAZINE in Taiwan and south China produced small nests like the Japanese colour form. V. ducali.r seems to be dependent on the brood of polistine wasps, particularly Polistes jadwigae Dalle Torre, as a source of food. Matsuura & Yamane (198a) related the four month duration and small size of colonies of V. ducalis to the four month nesting cycle of Polistes colonies and the low density of Polistes colonies in any one area. They estimated that a colony of V. ducalis would need 150-200 Polistes colonies for full development. Sakagami & Fukushima (1957) also give information about the nesting behaviour of the Japanese colour form of V. ducalis (which agrees with that given by Matsuura, 1984), and note the small nest size, with each comb rarely consisting of more than 100 cells. Yamane, So (1977) reported nest data which agreed with that given by Matsuura (1984). The only available information about the nesting biology of V. philippinensis is from the examination of an underground colony by Starr (1987), Starr & Jacobson (1990) taken at the end of September. The nest consisted of four combs with 1560 cells and was estimated to have produced 1480 adults although a cell was never used to produce more than one adult. Large and small cells were clearly discernible. Each sex showed a bimodal size distribution although the size difference between males was not aS clear aS that between gynes and workers. This colony was clearly much larger than any recorded f.or V. ducalis. Little information is available on the nesting biology of. V. tropica. Pagden (1952) recorded the Malayan colour form as building its nest inside aerial structures, e.g. hollow trees, under roofs and underground. Chan (1972) also recorded the Malayan colour form as building its nests inside aerial structures such as roofs or inside artificial structures such as sheds and boxes. Kojima (1982) recorded a colony of the north Philippine colour form in the hollow of a dead log which was partially covered by soil. Matsuura (1983) recorded a nest of the Malayan colour form in a river bank with some of the envelope visible. Starr & Jacobson (1990) found six colonies of the north Philippine colour form nesting underground; the three colonies that were dug up being 10-20 cm below the surface of the ground. Matsuura & Yamane (1984) reported that colonies of the Malayan colour form in central Sumatra could be initiated by more than one queen and that colonies may contain two to six queens even after the emergence of the workers. The nest of V. tropica found by Kojima (1982) in late August consisted of three combs of about 2600 cells. Chan (1972) showed a photograph of a nest of two or possibly three combs and over 200 cells are visible on the lowermost comb. Matsuura &" Yamane (1984) reported that the nests of the Malayan colour form contained 5-6000 cells and produced 500-1000 queens. One nest contained 5667 cells and ENTOMOLOGIST'S MONTHLY MAGAZINE 23r V) q \ V) O l) A) ca (.) \\ \ V) c:.: o ! IJ I c.i ob t! 232 ENTOMOLOGIST'S MONTHLY MAGAZINE 711 workers, 62 males and 245 queens (Matsuura 1983). Yamane, Sk (1974) found that workers and queens of V. tropica of the Indian-Chinese colour form were difficult to distinguish by size. Starr (1987) also referred to the difficulty of separating workers and queens of the north and south Philippine colour forms because of the presence of intermediate sized individuals. In this respect V. tropica differs from V. philippinensis where workers and queens are readily distinguished. Van der Vecht (1957) summarised several observations recording thatV. tropica preys on a variety of social Vespidae including Polistes, Ropalidia, Parischnogaster and Stenogaster. Matsuura (1983) also recorded the dependence of V. tropica on the larvae and sealed brood of the Polistinae and Stenogastrinae. The use of such prey is similar to that of V. ducalis. Clearly the mature nest size of V. ducalis is small in comparison with those of. V. philippinensis and V. tropica. These characteristics could be species specific or a consequence of V. ducalis'only having been studied in a temperate region and V. philippinensis and V. tropica in tropical regions. Study of V. tropica and V. ducalis in their overlap areas of North and South Vietnam and southern China should resolve this problem. REFERENCES Bequaert, J,, 1936, The common Oriental hornets, Vespa tropica and Vespa ffinis, and their color forms, Treubia, 15:329-351. Buysson, R. du, 1904, Monographie des gu6pes ouVespa. Annls Soc. ent. Fr.,72:260-288;1905, Monographie des gudpes ou Vespa. Annls Soc. ent. Fr.,73'.485-556, 565-634. Chan, K.L., 1972, The hornets of Singapore: their identification, biology and control, Singapore med. 1., 13: 178-187. Kojima, J.,7982, The genus Vespa in the Philippines (Hymenoptera, Vespidae), KontyA, 50: 434-444. Kojima, J. & Reyes, S.G., 1984, A new subspecies of Vespa tropica from Cebu Island, the Philippines (Hymenoptera, Vespidae). Kontytt, 52:260-261. Matsuura, M." 1983, Tropical hornets and honey bees, Insectarium, 20: 4-9; 1984, Comparative biology of the five Japanese species of the genus Vespa (Hymenoptera, Vespidae), Bull. Fac. Agric. Mie Univ.,69: 1-131. Matsuura, M. & Yamane, Sk., 1984, Comparative ethology of the Vespine Wasps, Hokkaido Univ. Press. Pagden, H.T., 1.952, Hornets in Malaya, Malay. Nat. J., 7: 78-83. Sakagami, S.F. & Fukushima, K., 1957, Some biological observations on a hornet, Vespa tropica var. pulchra (Du Buysson), with special reference to its dependence on Polistes wasps (Hymenoptera), Treubia,24: 73-82. Sonan, J.,1929, Research on Taiwan wasps. Trans. nat. Hist. Soc. Formosa, 19: 136-1.49. Starr, C.K., 1987, A colony of the hornet Vespa philippinensrs (Hymenoptera: Vespidae), Insectes soc., 34'. 7-9; 1989, More on Vespa tropica subspecies. Sphecos, 18: 6-7. Starr, C.K. & Jacobson, R.S., 1990, Nest structure in Philippine Hornets (Hymenoptera, Vespidae, Vespa spp.). Jap. J. Ent.,58'. 125-1,43. Vecht, J. van der, 1957, The Vespinae of the Indo-Malayan and Papuan areas (Hymenoptera, Vespidae), Zool. Verh., Leiden,34: 1-83; 1959, Notes on Oriental Vespidae, including some species from China and Japan (Hymenoptera, Vespidae), Zool. Meded. Leiden, 36:205-232. Wilson, E.O. & Brown, W.L., 1953, The subspecies poncept and its taxonomic applications, Sys/. Zool., 2: 97-1.1.1. Yamane, Sk., 1974, On the genus Vespa (Hymenoptera, Vespidae) from Nepal. Kontytt, 42:29-39. Yamane, So., 1977, Collecting techniques for Vespa nests, Seibutsukyozai, 12: 42-59. University College of Ripon and York St John, Lord Mayors Walk, York YO3 7EX. Iuly 3rd,7990.