Résumé
Background: The M and S molecular forms of Anopheles gambiae s.s. Giles appear to have speciated in West Africa and the M form is now formally named An. coluzzii Coetzee & Wilkerson sp.n. and the S form retains the nominotypical name (abbreviated here to An. gambiae). Reproductive isolation is thought to be the main barrier to hybridisation; even though both species are found in the same mating swarms, hybrid fertilisations in copulae have not been found in the study area. The aim of the study, therefore, was to determine whether differences in circadian and/or environmental control over the timing of swarming in the two species contribute to reproductive isolation. Methods: The timing of male swarming in these species was recorded four nights per month over four years at five swarming sites in each of two villages. The timing of the start and end of swarming, and the concurrent environmental parameters, temperature, humidity and light intensity, were recorded for n = 20 swarms/month/ species. The timing of 'spontaneous' activity at dusk of individual An. coluzzii and An. gambiae males was videorecorded in an actograph outdoors for 21 nights. Results: Of the environmental parameters considered, swarming was most strongly correlated with sunset (r 2 > 0.946). Anopheles gambiae started and stopped swarming earlier than An. coluzzii (3:35 ± 0:68 min:sec and 4:51 ± 1:21, respectively), and the mean duration of swarming was 23:37 ± 0:33 for An. gambiae and 21:39 ± 0:33 for An. coluzzii. Accordingly, in principle, whenever both species swarm over the same marker, a mean of 15.3 ± 3.1% of An. gambiae swarming would occur before An. coluzzii males arrived, and 19.5 ± 4.55% of An. coluzzii swarming would occurred after An. gambiae males had stopped swarming. These results are consistent with the finding that An. gambiae males became active in the actograph 09:35 ± 00:22 min:sec earlier than An. coluzzii males. Conclusions: The timing of swarming and spontaneous activity at dusk are primarily under circadian control, with the phase linked closely to sunset throughout the year. The mating activity of these two species is temporally segregated for 15-20% of the swarming period, which may contribute to the observed reproductive isolation of these species in local sympatric populations.