Field and laboratory observations in Sabah, East Malaysia on the proportion of Anopheles balabecensis balabacensis eggs hatching after holding in a humid atmosphere.

SEPARATION OF SPECIES OF THE ANOPHELES GAMBIAE COMPLEX BY SPERMATHECA SIZE 263 those of Ambursa B (33 ,m at wing length (Y) 266 mm to 31 ,um at 3 55 mm). These two scatter diagrams show no overlap and thus it is possible to distinguish Natsini A females from Ambursa B females by the size of the sperma- theca considered in relation to the wing length of the mosquito. Both the Natsini A and Ambursa B colonies originated from the Western Sokoto area of northern Nigeria and results indicate the potential use of the spermatheca as a taxonomic character in the An. gambiae complex. Undoubtedly selection will have occurred in maintaining colonies under insectary conditions, but it is unlikely that this will have affected six separate colonies so that the spermatheca size of females of the four An. gambiae B colonies is consistently smaller than the spermatheca of the females of the two An. gambiae A colonies. If these findings can be confirmed on material obtained directly from the field this will prove a useful and relatively quick method of separating species A and B of the An. gambiae complex, at least in northern Nigeria. It may be found especially convenient to make the spermathecal diagnosis in samples being dissected for other purposes. This method of separation can be used on alcohol preserved females, from which it is relatively simple to dissect out the spermatheca, allowing material to be preserved and examined at a later date if necessary. The other methods of separation can only be carried out with live material and in the case of chromosome determinations during only limited periods of the life cycle, or gonotrophic cycle. In many cases the method can be used on long- dried specimens after relaxing, providing the spermatheca has not been damaged during drying. Spermatheca size can be determined in the late pupal stage. ACKNOWLEDGEMENTS The author is indebted to Dr E. T. Burtt of the Zoology Department, University of Newcastle, for his invaluable suggestions and for discussions during the progress of this study, and to Dr G. Davidson of the Ross Institute of Tropical Hygiene, London, for providing the strains of An. gambiae A and B for the study. REFERENCES Chauvet, G., Davidson, G. & Dejardin, J. (1969) Cah. ORSTOM, ser. Ent. med., 7, 51 Chauvet, G. & Dedjardin, J. (1968) Cah. ORSTOM, ser. Ent. med, 6, 69 Coluzzi, M. (1964) Riv. Malar., 43, 197 Coluzzi, M. (1968) Parassitologia, 10, 179 Coluzzi, M. & Sabatini, R. (1967) Parassitologia, 9, 73 Davidson, G. (1962) Nature (Lond.), 196, 907 Davidson, G. (1964) Riv. Malar., 43, 167 Davidson, G. & Jackson, C. E. (1962) Bull. Wld Hith Org., 27, 302 Is4ail, A. H. & Hammoyd, E. I. (1968) Bull. Wld With Org., 38, 814 Field and Laboratory Observations in Sabah, East Malaysia on the Proportion of Anopheles balabacensis balabacensis Eggs Hatching After Holding in a Humid Atmosphere by N. RAJAPAKSA Anopheles balabacensis balabacensis is one of the most important vectors of malaria in South-East Asia, and is the main vector of malaria in Sabah, East Malaysia. It is a mosquito of the rain forest and its fringe. Interest in this species has been stimulated by the fact that the presence of Plasmo- dium falciparum strains resistant to all or most synthetic antimalarial drugs appeared in areas where this species is the sole or principal vector (Sandosham et al., 1963). The failure to interrupt malaria 2722F transmission in many areas where this species is a vector has also stimulated more fundamental re- search on its bionomics. Sabah is highly mountainous with forested hills and ridges occupying the major part of the country. The climate varies markedly from one valley to the next. Two monsoon rainy seasons occur, but many areas in the interior that are in the rain shadows of hill ranges have frequent dry spells lasting up to 2-3 months. 264 NOTES The breeding places of An. b. balabacensis are normally found in the shade under the vegetation amongst secondary vegetation near villages border- ing the jungle, or within the jungle itself. The favourite types of breeding place are buffalo hoof prints, buffalo wallows in partially dried stream beds, temporary pools on the margins of stream banks after a dry spell, seepages, blocked earth drains in rubber plantations, and wheel ruts. Breeding places may also be found in dense primary jungle far from villages near water holes where wild animals come to drink. Adults have been collected biting man in the evening at these locations in the jungle. Observations and results Careful observations were made of known breed- ing places in one of our study areas (Papar) to see when first-stage larvae appeared after dry potential breeding places were filled up with rainwater. In nearly all these previously dry breeding places, first-stage larvae appeared on the third or fourth day after the rain. Laboratory observations showed that An. b. balabacensis eggs took an average period of 2.5-3 days to hatch in water; this tallied with our field observations. The eggs of An. b. balabacensis in natural breeding places strand easily, as they do in the laboratory. These strands could be seen floating freely on the water surface, forming irregular patterns. In the laboratory it was noticed that first-instar larvae were often found in the breeding trays that had third- and fourth-instar larvae, although all the eggs placed in that particular tray came from the same female and were of the same age. It was later found that these larvae were a late brood from stranded eggs that hatched when fresh water was introduced to compensate for evaporation. Davidson (personal communication) reported that An. b. balabacensis eggs sent to him from Sabah hatched well 5 days after dispatch. The eggs were sent in plastic soap boxes containing moist filter-paper in layers between egg batches. Cheong & Santa Maria 1 reported that An. b. balabacensis eggs could hatch after a period of slow drying in a moist chamber. On day 0 they found an 83.6% hatch and on day 15 a 9% hatch. The rainfall in Sabah is unpredictable. Breeding places may dry up as quickly as they are filled with 1 Cheong, W. H. & Santa Maria, M. E. (1968) The effect of slow drying of Anopheles balabacensis balabacensis eggs on hatching. A paper presented at a Seminar and laboratory meeting, Singapore, 1968. rainwater. It is possible that newly laid eggs are stranded for days or weeks in such breeding places before the next shower. It would be an excellent way to tide over this unfavourable period if the eggs could withstand some degree of temporary desiccation. In an experiment to check whether stranded eggs could be found in known breeding places when they were drying up, 17 mud samples were collected from breeding places in Lingan Papar from which free water had disappeared. The samples were introduced into beakers and water was added. The mud was stirred with a glass rod so that any free floating matter could come to the surface. The samples were left overnight. In 3 of the 17 samples tested, a total of 7 anopheline eggs could be seen floating at the edge of the water surface. The eggs were carefully transferred with a camel-hair brush to finger-bowls containing water as used for laboratory rearing: 6 of the 7 eggs hatched within 3 days and went through all the aquatic stages to emerge as 2 female and 4 male An. b. balabacensis. In these earlier studies it was apparent that An. b. balabacensis eggs from mud in drying breeding places could hatch when water was added to the stranded eggs. To check the actual number of days that eggs remained viable when kept in a humid atmosphere, and to estimate the percentage hatch, we devised the following experiment. Both halves of plastic Petri dishes were covered with a 0.6-cm layer of cotton-wool; over these layers were placed two layers of filter-paper. The cotton-wool and filter-paper were moistened with distilled water so that the absorbent material soaked up enough water, without run-off when the dish was tilted. A known number of eggs on 1.2-cm-diameter filter-paper disks were placed on the bottom half of the Petri dish. The top half was then put in place and sealed tight with Cellotape. Altogether 23 such sets were made, containing a total of 10900 eggs. The Petri dishes containing eggs were kept at room temperature (average, 27°C). The eggs were from the P4 generation of the colony (Lingan Papar strain, Sabah) and from wild females from Lingan. One of the 23 Petri dishes was opened each day and the contents were transferred to trays containing tap-water. Daily counts of the numbers hatching were then made over several days. There were no delays in hatching beyond the third day following transfer, except in the day-8 batch when a few eggs hatched after the third day in the water. HATCHING OF ANOPHELES BALABACENSIS BALABACENSIS EGGS 265 Table 1. The percentage hatch of An. b. balabacensis eggs kept in humid atmosphere and wetted at various intervals of time No. of eggs Day of wetting No. of larvae [ % Hatch 200 0 172 86% 200 1 167 83.5 % 500 2 396 79.2 % 500 3 327 65.4 % 500 4 297 59.4 % 500 5 263 52.6 % 500 6 218 43.6 % 500 7 222 44.4 % 500 8 260 52 % 500 9 215 43 % 500 10 187 *37.4 % 500 1 1 163 32.6 % 500 12 170 34 % 500 13 57 11.4 % 500 14 63 12.6 % 500 15 66 13.2 % 500 1 6 71 14.2 % 500 17 14 2.8 % 500 18 17 3.4% 500 19 8 1.6 % 500 20 2 0.4% 500 27 3 0.6 % 500 28 J 0 0 % Hatches of over 50% were noted from day 0 to day 5; subsequently, the percentage hatch pro- gressively decreased, except on day 8 when slightly over 50% hatched. Viable eggs were found up to the 27th day (see Table 1). It was noticed that eggs laid by wild-caught females sometimes did not hatch for over 7 days, even in water. While checking these eggs in the laboratory for embryonation it was found that when slight pressure was applied on the coverslip the operculum popped out and live larvae were seen creeping out of the artificially hatched eggs. Discussion and conclusions Many vector insect species are known to produce eggs that can withstand desiccation but those reviewed by Bates (1949) are mainly found in the aedine and culicine groups. Although Holstein (1954) obtained hatching up to 16 days after holding An. gambiae eggs in damp clay and Breeland et al. (1970) found An. albimanus eggs would hatch up to 15 days after being held in saturated mud, it seems unique that the egg of An. b. balabacensis could remain viable out of free water for a period of nearly a month. An. b. balabacensis is a mosquito of the tropical rain forest where the rainfall is unpredictable. The breeding places are always shallow and temporary and are normally under shade. The forest floor is humid owing to slow evaporation from the spongy moisture-absorbent humus material and accumula- tion of dead leaves and vegetation. Hence, although breeding places lose their free water, the high humidity probably keeps the eggs viable for fairly long periods until the next rains come. These eggs were found towards the end of the rainy season and hence may possibly have been specially adapted to survive the oncoming dry period. The success of An. b. balabacensis as a malaria vector so widely spread throughout the forests of South-East Asia may be related to its adaptability to the unpredictable climatic conditions and to the ecosystem of the rain forest. Survival of this species in the egg stage during dry spells may be a means to tide it over the dry season when breeding places containing water are scarce. ACKNOWLEDGEMENTS The author acknowledges the encouragement given by Dr K. S. Jap, Director of Medical Services, Sabah, Dr Abdul Khalid Bin Sahan, Deputy Director of Medi- cal Services, and Dr C. 0. Innis, Principal Medical Officer (Health). Thanks are also due to Mr Chin Kui Foh, Mr Lazarus Badsun, and Miss Juliana Siol for technical assistance. REFERENCES Bates, M. (1949) The natural history of mosquitoes, New York, Macmillan, pp. 100-111 Breeland, S. G., Pletsch, D. J. & Quarterman, K. D. (1970) Mosquito News, 30, 81 Holstein, M. (1954) Biology of Anopheles gambiae, Geneva, pp. 72-75 (World Health Organization, Mono- graph Series, No. 9) Sandosham, A. A. et al. (1963) Med. J. Malaya, 18, 46-51 Scanlon, J. E. & Sandhinand, U. (1965) J. med. Ent., 2, 61-69