Reindeer live in arctic and subarctic regions, where ambient temperature can vary f rom 4 5 ° C i n mid-winter to + 3 5 ° C in mid-summer. In many reviews on food intake it is concluded that food intake increases i n cold and decreases i n hot environments (Baile & Forbes 1974, A r n o l d 1984). A seasonal decrease i n voluntary feed consumption of ad libitum fed reindeer during winter has been reported earlier (Ryg & Jacobsen 1982). However , increased feed intake of reindeer during cold winter periods is reported by reindeer herders feeding reindeer in practice. O u r objective was to determine the effects of ambient temperature on voluntary feed intake of reindeer. Eight adult female reindeer and eight halfyear-old reindeer calves (4 males and 4 females) were fed w i t h pelleted feed concentrates (contained 9 % ash, 10 % crude protein (CP) , 15 % crude fibre (CF) , 7 % ether extract (EE) and 59 % N-free extract ( N F E ) on dry matter basis) f rom the beginning of N o v e m b e r in 1987 unt i l the calving period in M a y 1988. Af ter the calving the reindeer were changed to a high protein feed (contained 10 % ash, 20 % C P , 13 % C F , 7 % E E and 50 % N F E in D M ) . The females and their newly born calves (4 males and 4 females) were fed together unt i l mid-December 1988, w h e n the calves were weaned and excluded f rom the experiment. The yearlings were also fed w i t h the same feed at the same time. In the middle of September 1988 the female and male yearlings were separated to prevent the mating. F r o m the middle of December 1988 the feeding of reindeer continued w i t h the winter feed unt i l next calving period in M a y 1989. The reindeer were fed twice daily ad libitum. The offered feed and the daily feed remains were weighed. The ambient temperature was measured daily w i t h a thermograf. The reindeer were weighed weekly .