In African antelope and North American cervids, breeding males during the rut engage in less oral self-grooming, and harbor a greater density of ticks, compared with conspecific females and non-breeding males. The purpose of this study was to experimentally test the proposition that down-regulation of self grooming in some male bovids occurs via the direct or indirect action of testosterone. Domestic dairy goats (Capra hircus) were used as a model. In Experiment 1, comparative observations conducted on twelve gonadally intact male goats (bucks), nine males castrated at 3 weeks of age (wethers), and twelve intact females (does) supported the prediction that the grooming rate of intact males would be depressed relative to wethers and does. Bucks oral groomed at one-third and one-fourth the rate of wethers and does, respectively, and they scratch groomed half as much as does. There was no significant difference between wethers and does in oral or scratch grooming rates. Experiment 2 involved castration of eleven bucks from Experiment 1, followed by 2 months of observation. Similar to the pattern of other testosterone-dependent behavioral changes after castration in adult males, there was a good deal of variation in the individual grooming response of males to castration, with increases in grooming taking 2 to 8 weeks to be manifested in ten of eleven goats. Overall, castrated males oral groomed about 3 x above their intact rates, supporting the prediction that castration removes testosterone-mediated suppression of grooming. This is the first example of alteration of grooming behavior in males by gonadal androgen, and the first demonstration of enhancement of any mammalian behavior by removal of gonadal androgen. (C) 1998 Elsevier Science Inc.