Adults of haematophagous Diptera, especially mosquitoes, are usually caught by using human or animal baits or in light or carbon dioxide traps. No single trapping technique will attract all species present in an area. For example, certain species are not attracted to light and very few ornithophagic species will be caught at human bait. But even when several species are caught by the same method it is most unlikely that they will be equally attracted, consequently their populations will not be equally sampled. This, however, is not always a serious limitation. In many virus isolation studies, for example, it does not matter whether mosquito vectors are sampled with the same efficiency, so long as sufficiently large numbers are caught. Sometimes, however, especially in ecological investigations, more representative samples of mosquito populations are needed. Because of the virtual impossibility of finding an attractant trap that will sample equally all species, it is best to use non-attractant traps. A disadvantage of these, however, is that because they catch mosquitoes only in their immediate area the numbers obtained are small unless mosquito populations are large. Whereas most attractant traps are heavily biased in favour of collecting unfed females orientated to host-feeding, non-attractant traps give more representative samples of both sexes and the different physiological conditions and age groups. However, it is important to remember that all the traps discussed in this chapter sample the aerial population, hence the numbers caught depends not only on population density but also on the flight activities of the individuals. Unfed females in fact usually comprise the greatest element of the catch, mainly because they are normally the most active. A non-attractant trap is unlikely to be completely free from sampling bias although every effort is made to minimise this. For example, the physical presence of a trap may promote visual responses, causing mosquitoes to be either attracted or repelled by it; similarly the hum of the motor in suction traps may influence the numbers caught. Despite these limitations aerial populations of insects can usually be more efficiently sampled, and with less bias, than most aquatic or terrestrial invertebrate populations. Suction traps probably give the most reliable estimates of both relative and absolute population size. For a better understanding of the total mosquito populations in an area the non-active resting populations should also be sampled (see preceding chapter).