The aim of this study was to clarify further the relationship between dietary keto-carotenoids (canthaxanthin and astaxanthin) and reproductive performance in female rainbow trout. Three experiments were undertaken in three successive breeding seasons. In addition to a control diet, ®sh were fed a canthaxanthin-supplemented diet (200 mg canthaxanthin kg A1 feed, designated as 6mC200) for 6 months prior to spawning (experiment A); the same diet as in A but for either 3 months (3mC200) or 6 months (6mC200) before spawning (experiment B); diets with one of two levels of astaxanthin supplementation, 50 mg (6mA50) or 100 mg (6mA100) kg A1 feed, or a diet with 100 mg canthaxanthin kg A1 feed (6mC100) fed for 6 months (experiment C).There was no signi®cant in¯uence of carotenoid supplementation on either the frequency of maturing females or the date of maturation. The number of ova per kg of female body weight averaged 2700 and did not vary signi®cantly among ®sh fed the dierent diets. Across experiments A, B and C there was no signi®cant dierence in egg and larval survival among ®sh fed carotenoid-supplemented and control diets. In experiment A, 6mC200 females produced smaller eggs than controls but this result was not con®rmed in experiments B and C. In general, eyed egg yield appeared partly dependent upon egg size. Alevin weight was also correlated with egg weight. The growth test conducted on ®ngerlings from experiment B failed to provide any evidence of an eect of feeding carotenoid supplemented diets to the female parent. Pigment analyses conducted on alevins revealed that canthaxanthin fed to the female parent was transferred into the eggs and therefore to the larvae, although canthaxanthin was metabolized within a few weeks after hatching. KEY WORDS