This study is based on 159 female white-tailed deer (Odocoileus virginianus) collected during the winters of 1961, 1962, and 1963 from, in order of decreasing habitat quality, the Turtle Mountain, Duck Mountain, and Whiteshell areas of Manitoba. Conception rates were high for adults and yearlings but relatively low for fawns. Ovulation rates and average numbers of fetuses per pregnant doe were high for adults from all areas but the ovulation rate was significantly higher for adults from Turtle Mountain. Both rates were high for yearlings from Turtle Mountain but relatively low for yearlings from the Whiteshell and Duck Mountain. Considering all age-groups, fertility was highest in Turtle Mountain, followed by Duck Mountain, and lowest in the Whiteshell. The rate of ova loss among adult does from Turtle Mountain was significantly higher than among adults from ihe other areas. This may be due, indirectly, to "flushing." Kidney fat indices showed that deer from all areas lost condition at the same rate during winter. A general comparison of winter environment between the Whiteshell, which was closed to hunting, and the Upper Peninsula of Michigan indicates that lower temperatures in the Whiteshell cause deer there to suffer more from malnutrition during average winters than do deer in the Upper Peninsula and hence to have relatively higher numbers of stillbirths and postnatal losses of fawns. Lighter snowfall in the Whiteshell allows deer to cover more of their summer range in winter than do Upper Peninsula deer, thus helping to explain why over-utilization of browse in the Whiteshell is not the serious problem it is in the Upper Peninsula, Winter weather appears to be sufficient to prevent deer irruptions in the Whiteshell and to hold maximum density well below that reached in the Upper Peninsula. It is postulated that similar controls function in the rest of Manitoba.