A comparison was made of the relation between mean number of Xenopsylla cheopis per infested rat, Rattus rattus and R. norvegicus, and the prevalence of infestation in four climatic zones in Southeastern United States. An increase of the mean number of fleas per infected rat, accompanied probably by increased activity of both rats and fleas in spring, induces an increased prevalence of infestation. A decrease of the mean number of fleas per infested rat, presumably most pronounced in habitats least tenable to fleas, induces a decline in prevalence. Two periods of increase and decrease occur in the northern-most and southernmost zones and only one each in the central zones. The second annual peak in the northernmost zone, most pronounced, seems to be induced by movement of rats into artificially warmed buildings in the fall. These are related to reported peaks and low points of prevalence of human typhus and presumably would be related to those of plague if it were present.
Present Address: State Natural History Survey, Urbana, Illinois.