The observations reported in this paper were made in a small settlement which exemplifies the conditions met with in an endemic malarious area in the savanna country of Eastern Colombia. The studies were aimed at obtaining comparative information about the mosquito density in the savannas and the samples obtained in house captures. Direct information on mosquito density was obtained by means of daytime captures of resting mosquitoes made with a portable cage and with evening captures of flying mosquitoes made with nets fixed to a moving vehicle.
The results of such captures were widely different from those obtained from house captures. It was proved that this was mainly due to host preference and, to a certain extent, to the clearing of grass around the houses. Of the anopheline mosquitoes found in the area, A. darlingi showed a clear preference for man when given the choice, under identical conditions, between man and domestic animals such as donkey and calf. A. darlingi proved to be more able to detect a bait at a distance than the other anophelines, thus rendering the small clearings round the houses ineffective against this species. A. argyritarsis behaved in the opposite way, being little attracted by man and easily deterred by grass clearance. A. pessoai showed an intermediate behavior between these two species.
Measuring the mosquito density in the savannas by the direct methods mentioned above, which are not influenced by any host preference, it was found that, of the anophelines present in the area, A. pessoai was by far the most abundant. A. argyritarsis had a much higher density than the house captures would lead one suspect. A. darlingi, on the other hand, proved to have a very low density, utterly disproportionate to the considerable numbers caught in human dwellings.
As in previous observations, we found that the culicines were about ten times more abundant in the savannas than the anophelines. They were little attracted by man and hardly entered the human dwellings.
Mosquito density within a 100 m. radius of houses was approximately the same as in the open savanna at half a kilometre from any human dwelling, but a relatively higher number of anopeheline females was caught near the houses than away from them. Though the sample in this case is small, this seems to indicate that the anophelines brought into contact with the human dwellings tend to remain in their neighbourhood.
Engorged females of both anophelines and culicines were more abundant in the vicinity of houses than away from them. The average number of engorged females found in these experiments was several times higher than the average number found previously in uninhabited areas.