are being answered. For example, collecting resting mosquitoes will give a more representative idea of the population for sexes and the feeding states than a trap that will predominantly catch feeding or recently fed females. However, sampling resting mosquitoes is time-consuming and more suited to research projects than an NMCP, which would favor a less-intensive collection method such as passive window exit traps. For an in-depth review of methods see Service (1971, 1993).
Relative vector density is an adequate surveillance method for most NMCPs. Collections should ideally be assessed against a baseline and established before the onset of control activities.
A good example of how relative species density is informative is from the ongoing malaria control program on Bioko Island, Equatorial Guinea. Species density is measured through a series of window traps at sentinel sites; in year one the IRS program with a pyrethroid had a significant impact on the An. funestus population but not the An. gambiae population. This was shown to be due to knockdown resistance (kdr) in the An. gambiae population; when the insecticide changed to a carbamate it had a significant impact on both vector species (Figure 2-10) (Sharp et al., 2007b).
The sustainability of IRS is controversial. However, this mode of control has been linked to some of the world’s most successful malaria campaigns including the near eradication of malaria from Sri Lanka in the 1960s, South America in