Distinctive Hydrocarbons Among Giant Honey Bees, the Apis Dorsata Group (Hymenoptera: Apidae) Da Carlson, Dw Roubik, K Milstrey
Total Page：16
File Type：pdf, Size：1020Kb
Distinctive hydrocarbons among giant honey bees, the Apis dorsata group (Hymenoptera: Apidae) Da Carlson, Dw Roubik, K Milstrey To cite this version: Da Carlson, Dw Roubik, K Milstrey. Distinctive hydrocarbons among giant honey bees, the Apis dorsata group (Hymenoptera: Apidae). Apidologie, Springer Verlag, 1991, 22 (3), pp.169-181. hal- 00890905 HAL Id: hal-00890905 https://hal.archives-ouvertes.fr/hal-00890905 Submitted on 1 Jan 1991 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Original article Distinctive hydrocarbons among giant honey bees, the Apis dorsata group (Hymenoptera: Apidae) DA Carlson DW Roubik K Milstrey 1 US Department of Agriculture, IAMARL, PO Box 14565, Gainesville, FL 32604, USA; 2 Smithsonian Tropical Research Institute, APDO 2072, Balboa, Panamá; 3 University of Florida, Department of Entomology and Nematology, Gainesville, FL 32611, USA (Received 10 April 1988; accepted 11 February 1991) Summary — Cuticular hydrocarbon pattern (CHP) analysis was performed on giant honey bees (the Apis dorsata group) including: 1), those occasionally given species status-Himalayan honey bees, Philippine honey bees, Sulawesi honey bees; 2), those separated since the Pleistocene- common A dorsata of the Indian and Asian lowlands and islands on the continental shelf (India and Sri Lanka, Thailand and Sumatra); and 3), giant honey bees of Borneo and Palawan, potential step- ping-stones to the Philippines and Sulawesi. Four groups were found among giant honey bees by this CHP analysis. Most distinctive were those of Palawan and Nepal. The widespread lowland Apis dorsata differed very little among mainland and island populations, whereas those of Borneo, Sulaw- esi, and Philippines proper formed a single group. Those of the Himalayas appear to have diverged from A dorsata. Apis dorsata / Apis laboriosa / systematics / hydrocarbon / gas chromatography INTRODUCTION The giant honey bees, formerly subgenus Megapis (Ruttner, 1988) apparently con- Taxonomic problems having evolutionary tain cryptic species, notably Apis laborio- implications and applied importance sa, the Himalayan honey bee (Maa, 1953; abound in honey bees (Ruttner, 1988; Sakagami et al, 1980; Roubik et al, 1985; Roubik, 1989). Geographic races or sub- McEvoy and Underwood, 1988). New spe- species, although often purely artificial in cies have also been confirmed for other taxonomic work, have special value in Asian honey bees. Apis andreniformis (F studies of organisms such as honey bees Smith, 1858), proved to be distinct from because they may lead to recognition of the widespread "little honey bee" A florea bionomic characteristics that differ among (Wu and Kuang, 1986; Roubik, 1989). populations (Mayr, 1970; Ruttner, 1988). Also, a distinctive Saban honey bee, Apis * Correspondence and reprints : DW Roubik, Smithsonian Tropical Research Institute, APO Miami 34002-0011, FL, USA koschevnikovi (Buttel-Reepen, 1906), is merly Celebes; A ’binghami’ from Minahassa; found in northeastern Borneo and shares Dumoga), the Philippine Islands (A ’breviligula’ from Los Banõs, and its habitat with the closely similar Apis ce- Laguna: Leyte: Visca). Samples of Apis dorsata were collected from rana et al, 1988; et al, (Koeniger Tingek much of its geographic range and several is- 1988). These last 2 species (A koschevni- lands and included: 1), Palawan Islands (collect- kovi was called Apis vechti) not only have ed by SF Sakagami and T Inoue); 2), Sabah, distinctive male genitalic structure but also Borneo (collected by B Sutton); 3), New Delhi, different times of mating flight activity- India (DWR); Chaing Rai, northern Thailand both features that would prevent interspe- (DWR); Sri Lanka; Monaragala (DWR), and Su- western Indonesia From cific mating. Recently, McEvoy and Under- matra; Padang, (DWR). 4 to 13 bees were assayed from each giant hon- wood no differences in the (1988) reported ey bee group (table I), ie a total of 41 bees. All male endophallus of Himalayan honey were collected at flowers or had flown to lights bees and the common giant honey bee at night, making it likely that they were older Apis dorsata, but believed they could be bees. This is important for cuticular hydrocarbon distinct species. The Himalayan honey study, since younger bees, those that do not bee and other populations of giant honey leave the nest or forage, have different chemi- cal traits than older worker bees (D Carlson, un- bees are still under study because we do published observations). not know which races and which represent The identification of 3 classes of cuticular hy- are species. As a step toward definitive drocarbons-alkanes, alkenes, and methyl- analysis, we studied the giant honey bees branched alkanes-was made solely on the ba- using a chemoanalytical technique that sis of gas chromatograph retention times. Fol- can be applied to dead pinned museum lowing the methods of Carlson and Service specimens without damaging them- (1980) lipids were extracted from individual bees immersion in and then cuticular hydrocarbon pattern analysis during overnight hexane, fractionated by liquid chromatography using and Bolten, 1984; Francis et al, (Carlson small columns made from disposable 0.4 x 5.0 Based 1985; Lockey, 1988; Smith, 1988). cm pipettes packed with silica gel. The hydrocar- on the analysis presented here, we con- bon fraction of each extract was eluted with hex- clude that 4 natural groups and at least 2 ane, then subjected to gas chromatography. A species may exist. These hypotheses are varian Model 3700 flame ionization instrument was used, fitted with a DB-1 fused silica corroborated by a biogeographic analysis. capil- lary column (15 m length x 0.32 mm internal di- ameter; J & W Scientific, Folsom, CA). An OCI-3 glass on-column injector (Scientific Glass Engi- MATERIALS AND METHODS neering Inc, Austin, TX) was used in the split- less mode with the column oven at 40 °C. It was then temperature-programmed from 40 °C to The names of candidate giant honey bee spe- 320 °C, at a 20 °C increase per min. The sam- cies first suggested by Maa (1953) will appear ples were evaporated to dryness and taken up within quotation marks throughout the present in 50 μl of hexane solvent just before the injec- text. Pinned specimens of worker bees are now tion of one μl. The gas chromatograph was cou- in the collection of DWR or SF Sakagami, and pled through a 760-series interface and a Nel- were supplied from the collections of CK Starr son Analytical system to an IBM PC/XT (Philippine bees), B Sutton (Borneo bees), SF computer, an Epson FX80+ printer, and a Hew- Sakagami (Palawan, Nepal, Sulawesi bees), or lett-Packard 7470A plotter for data output. The were collected by DWR. These specimens in- column had ≈ 60 000 theoretical plates with a cluded workers collected up to 18 yr before our C13 n-alkane standard. The samples were the analysis. The bees, identified by SF Sakagami, equivalent of 2% of the amount obtained from were from: central Nepal (A ’laboriosa’ from Kal- one bee. Each was injected together with alkane li La, Drandikhola, Trubkin Kharka, Tukucha standards for determination of Kovats retention Palpa and Tharepati-Melemchi), Sulawesi (for- indices (KI) (Kovats, 1966). KI values were con- verted to equivalent chain lengths (ECL) by di- RESULTS viding by 100. The bee specimens used here were undamaged by hydrocarbon extraction. Five ratios were derived for each specimen by Visual dividing the recorded integrated areas of 2 se- comparison lected alkenes (2 475 and 2 675 retention times), and 3 methyl alkanes (2 735, 2 935, Alkanes, alkenes and internally methyl- 3 the of the 2 700 135) by area alkane peak. branched alkanes in each sample con- The areas were calculated for the peaks printed tained 23 to 36 carbons (fig 1 a-f). In gen- out by the plotter, as shown in the following re- sults. eral, the bees exhibited variation in 4 sets of triple peaks, with the center peak always Precautions were taken to screen for the ex- an n-alkane of or 31 carbons. In traction technique and possible effects of other 25, 27, 29 a series of alkenes was hydrocarbons found on the bodies of worker addition, larger bees. Pollens on the bodies of Apis mellifera found, having 34- to 37-carbon (outside of the load carried in the corbicula) "backbones" (Ki 3 365 to 3 665). A series were found to contribute insignificant amounts of alkadienes with the same ’backbone’ (KI of to that extracted from hydrocarbon compared 3 343 to 3 643) was also found in the de- a bee The of as specimen. quality hydrocarbons scendants of African honey bees in Vene- shown by GC patterns did not appear to differ appreciably between honey bees extracted zuela (Carlson and Bolten, 1984). Each of overnight in hexane, compared to nest mates the bee groups we examined was evaluat- that were placed in hexane for 10 min. ed as follows: Apis dorsata neo bees (fig 1c and fig 2) were substan- tially similar to those of the Philippine bees and but unlike those of the Asian The 5 worker bees from Palawan exhibited Sulawesi, mainland or Palawan. The CHPs of bees consistent CHPs, comprised primarily of 4 from Thailand, Sri Lanka, and India were sets of triple peaks with a smaller fourth similar to each other Indian bee, peak (fig 1 a). The Kovats retention indices very (see and to an A dorsata worker from of 2 675, 2 700 and 2 735 here correspond fig 1b), Pakistan (Francis et al, 1985).