Abstract
Sixty-four mares (27 foaling, 37 barren or maiden), mainly Finnhorses, were subjected to treatment with 14.5 h of light and 9.5 h of darkness, starting at the beginning of December. The onset of cycling in non-foaling mares was estimated by weekly serum progesterone determinations. All of the non-foaling mares cycled in the middle of March. They started to cycle on average in the middle of February, 11.1 weeks after the beginning of the trial. There were statistically significant differences in relation to breed (Finnhorses started to cycle 2 weeks later than warm blooded, p less than 0.02) and in relation to age (brood mares with mean age 10 years, started to cycle 10 days later than those 3 years old, p less than 0.03). However, there was no statistical significant relation to previous lactation, although lactating mares lactating up to the previous autumn started to cycle 7 days later than dry mares (p less than 0.15). Artificial insemination of 14 mares in the 2nd oestrus of the year, in March, resulted in 12 foalings in the subsequent year (86%). The following winter, all pregnant mares (N = 27) were exposed to the same kind of light treatment, starting on 1st December. The 1st mare foaled at the end of January. The time from foaling to 1st post partum ovulation was significantly longer (17.0 days) in foalings taking place within 10 weeks from the beginning of the light treatment period, than in foalings occurring after more than 10 weeks of lighting (12.1 days) (p less than 0.01).