Abstract.-I tested three assumptions of the Hamilton and Zuk hypothesis (1982), which suggests that the extravagant male plumage of many bird species allows females to choose mates that are resistant to the parasites exploiting the host population at a given time. By choosing such males as mates, females will rear offspring carrying the genes for resistance. I tested three necessary conditions for the Hamilton and Zuk model: (1) whether parasites affect the fitness of their hosts; (2) whether there is heritable variation in parasite resistance, and (3) whether the expression of the sexual ornament varies with parasite burden. The haematophagous mite Ornithonyssus bursa (Macronyssidae, Gamasida) sucks blood from their Bam Swallow (Hirundo rustica) hosts. Experimental manipulation of mite loads and partial cross-fostering experiments on Bam Swallows, where half of the nestlings in the brood were exchanged with nestlings from another nest, shows that parasite burdens and origin, but not rearing conditions, of Bam Swallow nestlings, affected their adult tarsus length and maximum body weight shortly before fledging. Mite loads of adult Bam Swallows at spring arrival were more similar to mite loads of their own offspring, whether reared in their own or in foster nests inoculated with mites, than to loads offoster offspring. Parent Bam Swallows with long tail ornaments had offspring with smaller mite loads in the partial crossfostering experiments. The amount of increase in male tail ornaments from one year to another was negatively related to experimentally manipulated mite loads of Bam Swallow nests during the preceding breeding season. In conclusion, the three assumptions ofthe hypothesis were supported by the experimental tests.