One hypothesis assumes that sexual ornamentation has evolved to reveal an individual's health and vigour. Therefore, choosy mates may use ornamentation as an indicator of the presence and effectiveness of genes for resistance against parasites (Hamilton & Zuk (Science, Wash. 218, 384 (1982))). However, the connection between parasites and sexual ornamentation may be more direct: the different characters of the ornamentation could reveal the contribution of each sexual hormone to the whole hormone mix that induced the ornamentation. As androgens and oestrogens are known to weaken specifically parts of the immune system (Grossmann 1985), and several parasites require a specific aspect of the immune system to be countered effectively, the hormone mix may reveal the actual use of an animal's immune system which depends on the presence and burden of, or even susceptibility to, different parasites. Therefore detailed information about a host's parasites could be available by studying its sexual ornamentation. Breeding tubercles, the sexual ornamentation of many fish, are induced by several androgens and at least one oestrogen (Wiley & Colette 1970). I studied four characteristics of the ornamentation of male roach (Rutilus rutilus) and found that two of them could be used to discriminate between males that are infected with either Diplozoon or nematodes, the two most severe parasites found. Furthermore, the number of parasites of each group correlates negatively with the expression of one or both of the other ornamental characters. A female roach could potentially decode a male's ornamentation to gather a sort of clinical picture of him and use this information in her choice of mate.