Short or Long Day Light Regimes May Not Affect Reproductive Performance in the Sow O Ha¨lli, A Tast, M Heinonen, C Munsterhjelm, A Valros and OAT Peltoniemi Department of Production Animal Medicine, University of Helsinki, Ma ¨ntsa ¨la ¨, Finland Contents The objective of this study was to evaluate the effect of short or long day light regimes on the reproductive performance in the sow. The study comprised approximately 1300 sows and their 3400 breedings in three commercial sow pools, in which a batch farrowing of 40 sows was employed. Each batch remained in the farrowing unit for 8 weeks under either a short day (10 h light, 14 h dark) or a long day (14 h light, 10 h dark) light regime. After weaning and transportation to the central unit, all sows were kept under the long day regime until they were moved back to the farrowing unit. Production results for each individual sow were recorded. The effect of the two different light regimes on the farrowing rate (FR) and on the weaning-to-oestrous interval was evaluated using logistic and Poisson regression models, respectively. The light regime affected none of the parameters significantly. The FR in June, July and August did, however, drop below that in December, January and February (OR 0.7, p < 0.05). The FR was affected by the sow pool (p < 0.01). Parity two sows exhibited a lower FR (OR 0.5, p = 0.05) and higher incidence of delayed first oestrus (IRR 2.7, p < 0.01) than did older sows. This study indicates that despite an artificial light regime, sows may still react to changes in season. In conclusion, use of simple light regime to obtain complete control over reproduc- tive performance appears to be difficult. Introduction In regions where the length of daylight varies consider- ably between seasons, seasonal infertility in pigs is a well-recognized phenomenon. Some of the most common manifestations of seasonal infertility in gilts and sows are delayed puberty, a longer interval from weaning-to-oestrous and a decreased farrowing rate (FR) (Love et al. 1993). The photoperiod is known to be the most important external regulating factor in seasonal breeders (for review, see Karsch et al. 1984). In neuroendocrinolog- ical control of seasonality, melatonin is considered a chemical messenger that conveys the information coded in a photoperiod to the reproductive functions of an animal. Both wild boars (Sus scrofa) and domestic pigs react to changes in the length of daylight in terms of melatonin secretion (Tast et al. 2001a). Yet, the effect of the circadian melatonin rhythm and its implications in reproduction remain partly unknown in pigs (Paterson et al. 1992; Bassett et al. 2001). The season and photoperiod are clearly capable of compromising certain production parameters in pigs. Paterson and Pearce (1990) claim that long days inhibit the attainment of puberty in gilts. In sows weaned under a long photoperiod, the weaning-to-oestrous interval (WOI) lengthens (Prunier et al. 1994). At least in Finland, a drop of 10% units in the FR is frequently observed during the period of seasonal infertility (Pel- toniemi et al. 1999). All factors mentioned above could significantly affect reproductive performance in sows. Impaired fertility has dire consequences for the economy of swine production. Knowing the importance of the photoperiod in the regulation of seasonal infertility, it is worthwhile to attempt to control the production losses with light regimes that can be fitted in the production cycle of the sows. In Finland, one previous attempt to improve fertility has applied short and long light regimes. In this previous trial, short and long day groups showed no significant differences in their reproductive perfor- mances. Researchers claimed that the short day period should have lasted longer, at least 8 weeks instead of the 6-week period used (Tast et al. 2005). An adaptation period of up to 16 weeks had been used in successful trials in sheep (Kennaway 1988; Lincoln 2002). In the present study, we aimed to correct the duration of the short day period by extending it to 8 weeks. The goal was to establish a higher reproductive performance by using a simple light regime. We hypothesized that the short day treatment before breeding would promote a higher reproduction rate than would the long day treatment. Materials and Methods Animals, sow pools and animal husbandry The study was carried out in three sow pools (Dalin et al. 1997; description for a sow pool) located in southern and western Finland from June 2003 to December 2004. Approximately 1300 sows and their 3400 breedings were included in the study. The mean parity of sows was 5.6 (SD 2.5) with a minimum value of 2 and a maximum of 12. The animals were crosses between Yorkshire and Finnish Landrace. Each sow pool consisted of a single central unit, which managed the breeding and housed the pregnant sows, and four satellite units in which farrowing took place. The strict batch farrowing of 40 sows was practised in all sow pools. Altogether, each sow pool owned 440 sows. Each batch of sows was transported to the satellite unit 2 weeks before the anticipated farrowing, and the sows spent altogether 8 weeks in the satellite unit. Immedi- ately after weaning, the sows were transported back to the central unit for breeding. Beginning on the second day after weaning, oestrous detection was carried out twice daily in a fence-line contact with mature boars. The sows were either inseminated with a mixture semen from several boars provided by a commercial artificial insemination centre (Cooperative Breeding Service, Reprod Dom Anim 43, 708–712 (2008); doi: 10.1111/j.1439-0531.2007.00976.x ISSN 0936-6768 Ó 2008 The Authors. Journal compilation Ó 2008 Blackwell Verlag