Search

During 1972–74, 13 865 female mosquitoes of 25 species from eight genera were collected at the Kisumu heronry using light-traps, chicken-baited traps and battery-driven aspirators. A total of 3298 of these were engorged and 2169 were subjected to precipitin testing in order to determine their food sources. A high percentage of feeds on birds was recorded for Coquillettidia karandalaensis (Wolfs) (96%), Culex poicilipes (Theo.) (82%), and C. univittatus gp (65%); Mansonia uniformis (Theo.), M. africana (Theo.) and C. antennatus (Becker) were found to feed on birds occasionally. Results from catches in CDC light-traps set at various heights in trees with nesting birds indicated that C. univittatus gp, C. poicilipes and Aedomyia africana Nev.-Lem. were active in the upper levels of the canopy. Further analysis of blood-meals from birds indicated that C. univittatus gp mosquitoes were feeding principally on ciconiiform birds. Any mosquito which feeds regularly on birds and man must be considered as a potential vector of arboviruses; the results indicate that the most important vectors at Kisumu are probably C. univittatus gp mosquitoes.

Mosquitoes flying at low levels over open farmland were sampled by means of electrical suction traps. These were set up at nine levels from ground level up to 6 m. From the vertical profiles obtained it was possible to recognise three patterns of behaviour: (1) a low-flying group with relatively very high densities below 1 m, comprisingMansonia (Man-sonioides)spp.,Aedesspp. and some species ofAnopheles; (2) an intermediate group with densities rather evenly distributed at the lower levels but declining above 2–4 m, comprisingA. funestusGiles,A. gambiaeGiles andCulex neaveiTheo.; (3) a high-flying group with catches at 6 m greater, or much greater, than at 1 m, composed ofC. antennatus(Becker),C. thalassiusTheo. andC. poicilipes(Theo.). For all species, catches after 23.00 h showed an increase in the proportion of mosquitoes taken in traps at the lower levels, this being most marked at ground level and 0·5 m. No influence of either moonlight or wind speed could be detected to account for this. Biting catches on human baits showed a generally similar pattern to suction-trap catches, although differences between baits at 1-m intervals at the higher levels were less than with unbaited traps.

The mosquitos of two villages on the Kenya coast were studied over a period of two and a half years, which included years of very light, average and very heavy rainfall. A survey was made of the species occurring in the bush and in the houses, and in addition 24-hour biting catches were done and window-trap catches were examined to obtain information on the movement of species in and out of houses.Sixty five species were taken in net catches in the bush but only 19 were taken regularly; five (Taeniorhynchus uniformis (Theo.) and T. africanus (Theo.), Aëdes woodi Edw., Culex invidiosus Theo. and C. guiarti Blanch.) were abundant. In the houses, two species, C. pipiens fatigans Wied. and Aë. aegypti occurred regularly, the former being abundant and the latter not uncommon. Anopheles gambiae Giles and A. funestus Giles were seasonal; A. funestus appeared only after exceptionally heavy rain. The in-going window trap attracted only an occasional mosquito seeking shelter during the day and only six species were taken. Thirty species were found in the out-going trap but only seven occurred regularly. A. gambiae, A. funestus, T. uniformis, T. africanus and C. p. fatigans were most active during the hours of darkness, with the main activity in the four hours before dawn. Aë. aegypti had two waves of activity, in the four-hour periods before dawn and before sunset, respectively; the sunset peak was higher and the main activity probably occurred between 18 and 19 hr., as previously recorded for this species in Mombasa.Thirty five species were taken biting in the 24-hour catches in situations designated bush, house and compound, respectively. The biting cycles of A. gambiae, Aë. aegypti and C. p. fatigans have been compared in the three situations. The biting rhythm of each species was much the same in the different environments but the time and magnitude of the main peaks varied. In the case of Aë. aegypti, which probably includes two different forms, there is a marked difference in the magnitude of the evening peak in the bush and house. The biting cycles in the bush are given for A. squamosus Theo., T. uniformis, T. africanus, Aë. fryeri, Aë. metallicus (Edw.), Aë. albicosta (Edw.) and C. antennatus (Becker) and their occurrence in the other two situations is noted. The cycles of T. uniformis and Aë. fryeri agree with previous findings, but that of T. africanus differs from other cycles recorded for this species by having the main biting after midnight. The remaining species were rare, but one catch of 274 specimens of A. funestus was recorded.Resting mosquitos were more numerous in the second half of the year from August to December, a period of light rainfall and high temperatures. Biting mosquitos were more commonly taken, especially in the bush, in the second quarter of the year, during the main rainy season. The biting activity of some species increased in the houses in the second half of the year, when there are high winds, and biting in the bush was minimal.

In the February number (1928) of this Journal, a most interesting and valuable paper was published by Mr. F. D. Golding, Government Entomologist of Nigeria, on the bionomics of cotton-stainers (Dysdercus). In a footnote on page 319 of this paper it was suggested on my authority that three of the seven species mentioned by him as occurring in Nigeria were probably wrongly identified, since no representatives of these three species were present in the Nigerian material in the British Museum. Since then I have been able to examine Mr. Golding's collection and can state definitely that the specimens of Dysdercus nigrofasciatus, Stål, and D. migratorius, Dist., taken in Nigeria, were correctly identified. D. intermedius, Dist., however, does not, so far as is known, Occur in Nigeria. To the species mentioned by Golding as occurring in that country must be added D. antennatus, Dist., which however is probably generically distinct from typical Dysdercus species.