Forty-six Simmental × British Friesian bull calves were allocated to six treatment groups. In four groups (each of eight animals) half the animals were given a prime injection of gonadotropin-hormone-releasing-hormone (GnRH) either as the decapeptide or as an octapeptide (residues 3 to 10) conjugated to egg albumen. Prime injection times were at 3, 4, 5 or 6 months of age. All animals were boosted with a GnRH conjugate, similar to that used for the prime injection, at 8 months. At 8 months, six other animals were surgically castrated while the remaining eight were left as untreated bull controls. Weight gain and consumption of a barley-based diet offered ad libitum were recorded for individual animals. Blood samples were taken at a minimum of fortnightly intervals and the serum analysed for antibody titre against GnRH, testosterone and insulin-like growth factor 1 (IGF-1). Animals were slaughtered at 12 months and chemical analyses performed on the dissectible material of the 10th rib for protein, lipid, ash and water content. Greater antibody titres and a longer period of low serum testosterone were achieved with the octapeptide conjugate compared with the decapeptide. Serum IGF-1 slowly decreased following both surgical- and effective immuno-castration. There were no significant differences in food intake between the groups. Both steers and the more responsive immunocastrates had higher fat (P < 0·01), lower protein (P < 0·05) and water concentrations (P < 0·01) in tissues from a rib sample joint compared with untreated bulls. As immuno-responsiveness decreased there were indications of compensatory changes in body composition. The technique may be applicable during periods of, for example, mixed grazing for bulls.