doi:10.1006/anbe.2003.2072 o

tial 20 tat t o ted bse ati app th but not juveniles, also developed more saturated plumages in one of the years, but did not differ from controls in the other year. This is the first experimental demonstration that ectosymbionts (including mites) have a negative impact on the characteristics of the plumage developed and consequently can signal the healthiness of their hosts. 2003 Published by Elsevier Science Ltd on behalf of The Association for the Study of Animal Behaviour. In many species, individuals choose mates with well-developed ornaments (Andersson 1994). From this choice they can obtain ‘good genes ’ if such characters signal some heritable component of viability (Møller & Alatalo 1999). Alternatively, the benefits can be direct if more ornamented individuals are more able to defend or pro-vide resources for their mates or offspring (Møller & Thornhill 1998). Parasites can potentially affect both types of characters. According to Hamilton & Zuk (1982), bright coloration in birds could function as a reliable indicator of genetic resistance to parasites. Parasites can also reduce the quantity of resources allocated to repro-duction and increase the risk of infestation of the mate (Milinski & Bakker 1990; Hillgarth 1996). In the great tit, Parus major, and the cirl bunting, Emberiza cirlus, brightly coloured males are healthier than drab males (Dufva & Allander 1995; Figuerola et al. 1999a), a necessary condition for parasite-regulated sexual selection to work. Studies investigating the relations between parasite burdens and the expression of sexually selected traits at the intraspecific level have produced contradictory results, however, with positive, negative and nonsignificant relations between parasite abundance and ornament size or brightness (reviewed in Møller 1990). Experiments seem to provide a more con-sistent approach, but few have been done (Clayton 1991