Inappropriate lighting has a negative impact on the welfare of many captive animals; photoperiod, illuminance and spectrum all influence the suitability of a light source or regime. The lighting under which a pig is raised can influence an animal’s ocular, physical and neural development, as well as its behaviour. Although there are no published data for the UK pig industry, photoperiod and illuminance are thought to vary widely within pig housing. In England, the legal requirements for pigs housed under artificial lighting are: (i) that lighting with an intensity of at least 40 lx must be provided for a minimum period of 8 h per day (The Welfare of Farmed Animals (England) (Amendment) Regulations 2003; S.I. 2003 No. 299; following EC regulation: Council Directive 2001/88/EC) and (ii) that animals kept in buildings must have an appropriate period of rest from artificial lighting (TheWelfare of Farmed Animals (England) Regulations 2000; S.I. 2000 No. 1870; DEFRA, 2003), though the duration of this period is not specified. Additionally, adequate lighting (whether fixed or portable) must be available to enable the animals to be thoroughly inspected at any time (ibid). One UK farm assurance scheme (Freedom Food, 2003) further requires that housed pigs must have access to an area that provides a minimum illuminance of 50 lx for a continuous period of at least 8 h and a continuous dark period of at least 6 h; either of these temporal minima may be reduced to correspond with seasonal changes in daylength. The research described in this paper stresses the importance of identifying the needs and preferences of domestic animals rather than imposing conditions based on human perception and preference. The preferences of growing pigs for different illuminances and indirectly for photoperiod were determined experimentally. This experiment investigated the preference of juvenile pigs for illuminance, and indirectly photoperiod, at two ages. The animals were equally familiar with all illuminances prior to testing. Four groups, each of four pigs, occupied a four-compartment preference chamber in which a different illuminance was applied to each compartment: minimum (2.4), 4, 40 and 400 lx. Illuminances were rearranged every 2 days to avoid positional bias. The pigs significantly preferred the dimmest illuminance (mean occupancy 7 h 20 min per 24 h, backtransformed data) and spent the least time in the brightest (4 h 49 min per 24 h), with an intermediate and similar length of time spent in the other illuminances (6 h 25 min and 5 h 25 min in 4 and 40 lx, respectively, F3, 127 = 8.93, P < 0.001). The most common behaviours of the pigs when in the darkest compartment were resting and sleeping. The EU directive 2001/88 requires a minimum illuminance of 40 lx for pig production; this illuminance was neither aversive nor strongly preferred by the pigs. Our findings also suggest that pigs should be provided with an appropriate period of rest at an illuminance of 2.4 lx for at least 6 h per day. The only active behaviour affected by illuminance was defecation; the pigs preferred to defecate in the brighter illuminances. Spatial provision of minimal illuminance could potentially improve pig welfare by providing a preferred light environment for resting and also by creating a resting area distinct from dunging areas, thus improving hygiene.