Abstract Three species of mites, Acarus siro L., A. farris Oud. and A. immobilis Griffiths, were tested under optimum physical conditions (90% r.h. and 20°C). In most experiments hypopus production was estimated by comparing the number of protonymphs which developed promptly to tritonymphs with the number which developed to hypopi. Hypopus formation appeared to be favoured by malnutrition in the presence of a certain minimum quantity of B vitamins. The period of response embraced the active larval and active protonymphal stage and the response appeared to be cumulative. It scemed to remain reversible through the larval stage but at some point during the active protonymphal period it became irreversible. The developmental period of those protonymphs which became hypopi was much extended compared with the developmental period of those which proceeded directly to tritonymphs. It is concluded that this extension is associated with physiological and developmental changes peculiar to the hypopus and not with starvation. Acarus immobilis and Acarus farris, which have been shown to be mainly field species, produced more hypopi than the storage species Acarus siro, one population of which could not be induced to produce any hypopi. It is concluded that previously published records of the “Flour mite” motile hypopus from field habitats are probably based on the A. farris hypopus and the records of the “Flour mite” motile hypopus occurring in food stores are scarce because the storage species A. siro has a poor potential for hypopus formation or rarely encounters starvation conditions. Misidentifications, coupled with interspecific differences in hypopus production, may explain some of the conflicting results published by previous workers testing the same physical factors.