This paper considers the available literature on the transmission of malaria by insects and concludes that, in contrast to the commonly held view (that implies mosquitoes are naturally vectors of malaria), it is more useful to consider that mosquitoes, like plants, normally express a variety of gene products, which together render the host resistant to infection. The consequences of this hypothesis upon current research are that when studying the passage of the malarial parasite through a competent vector it is relevant to ask either 'How have the natural innate defences of the insect failed?' or 'What mechanisms has the parasite used to overcome these defences?' At the population level, the hypothesis is consistent with the conclusions of Koella et al. that the evolutionary cost of maintaining defence mechanisms that can render the mosquito refractory (e.g. melanization) has prevented fixation of the necessary gene(s) in the insect population. We simply extend that concept by stating the innate and genetic defences that confer the natural (and sometimes incomplete) resistance to infection are of sustainable cost, with the consequence that the encoding genes may become highly prevalent or fixed in a population.