Effects of photoperiod on sex steroids and gonad maturation in Arctic charr

Four groups of Arctic charr (age 3+), previously exposed to natural photoperiod, were subjected to different photoperiod regimes from February 20: Simulated natural photoperiod (SNP; Tromsø 70°N), continuous light (LD 24:0) followed by short day (LD 6:18) from either May 1 (L_Feb–May) or June 25 (L_Feb–June), or continuous light (LD 24:0) throughout the experiment (L_Continuous). In females, peak levels of estradiol-17β and testosterone were 6 and 10, 6 and 7, and 3 and 5 weeks advanced, in L_Feb–May, L_Feb–June and L_Continuous, respectively, whereas in males plasma levels of 11-ketotestosterone peaked 8 weeks earlier in L_Feb–May and L_Feb–June than in SNP and L_Continuous. Median ovulation time in L_Feb–May and L_Feb–June was advanced by 10 weeks compared to SNP and L_Continuous, and median spermiation time (first observation of running milt in each individual) in L_Feb–May, L_Feb–June and L_Continuous was advanced by 10, 7 and 5 weeks, compared to SNP. A switch from long to short day early in the reproductive cycle (L_Feb–May and L_Feb–June) apparently resulted in stronger synchronization of both ovulation and spermiation. Ovulation in L_Feb–May and L_Feb–June occurred over 3 weeks (L_Feb–May) and 8 weeks (L_Feb–June), as compared to 10 weeks in the SNP group, whereas all males (except one individual) began spermiating over a period of 3 weeks (L_Feb–June), or less (L_Feb–May), as compared to 7 weeks in the SNP group. Lack of a short day stimulus (L_Continuous), on the other hand, resulted in larger variation in the timing of final maturation, with ovulation and spermiation being spread over 15 weeks. Low egg survival in L_Feb–May (17% versus 60% and 97% for L_Continuous and SNP) indicate that a long day signal of 10 weeks early in the reproductive cycle may be too short to ensure good egg and sperm quality. Photoperiod manipulation did not affect the proportion of maturing fish.