Host choice and fecundity are two factors that may contribute to the variation in flea counts observed when assessing the potential risk of flea-borne transmission of pathogens from rodents to humans. Using the black rat, Rattus rattus Linnaeus, as host the effects of age and sex on host choice and fecundity of the Oriental rat flea, Xenopsylla cheopis Rothschild, were examined experimentally at 25 degrees C and 80% rh. During the first two days of emergence from cocoons, female fleas dominated the sex ratio by 4:1 but from the third day onwards this switched to a male-dominated sex ratio of 4:1. The sex of the flea did not influence their host-seeking behaviour. Newly emerged fleas of both sexes were not influenced by the rat's presence and at seven days old both sexes demonstrated similar levels of attraction toward the rat host. The sex of the rat did not affect flea host-seeking behaviour. There was a 50-70% decline in the initial number of adult fleas during the first week after their release onto a rat host, and this decline was greatest on juvenile rats. Flea fecundity was also significantly lower on juvenile rat hosts but no differences due to the sex of the rat were observed. This experimental study supports the hypothesis that differences in flea count due to host sex, reported in field surveys, result from sexual differences in host behaviour and not from discriminatory host-seeking behaviour by X. cheopis. Differences in flea count due to host age may be affected by differences in X. cheopis fecundity, which may itself be mediated by host behaviour such as grooming.