To observe forage competition between deer and livestock, the forage selections of a tame deer (Odocoileus virginianus), a goat, a sheep, and a cow were observed under four range conditions, using both stocked and unstocked experimental pastures, on the Kerr Wildlife Management Area in the Edwards Plateau region of Texas in 1959. The animals were trained in 2 months of preliminary testing. The technique employed consisted of recording the number of bites taken of each plant species by each animal during a 45-minute grazing period in each pasture each week for 1 year. Results indicated moderate to heavy competition for browse and mast between the deer and goat in all pastures during all seasons, and between the deer and all three classes of livestock in the winter. Browse and mast comprised over 50 percent of the deer's diet, except in the spring and summer, and over 50 percent of the goat's diet in all seasons. Following a decrease in available browse, sheep become competitive with deer for forbs. Forbs formed an average of 68 percent of the deer's diet and 65 percent of the sheep's diet in the ungrazed control pasture in the summer; the animals generally were grazing the same species. The added factors of bearing and nursing fawns, and growing antlers, contribute to the nutritional stresses on deer in the summer, the period when extensive die-offs of deer most frequently occur in the Edwards Plateau region. Competition for grass between deer and livestock probably is important only when grass is in a succulent stage in the spring and no other forage is available in quantity. Continuous grazing by animals tends to remove the most palatable species and concentrates competition on fewer, less palatable ones. Observations from the numerous replications of feeding times and places allow the conclusion that this technique provides a reliable index to preferred foods and staple foods. In the Edwards Plateau region of Texas, most of the rangelands are grazed by two or more classes of livestock along with deer. Research on the Kerr Wildlife Management Area, near Hunt, Texas, indicates that deer when managed and harvested properly can provide economic returns comparable to those from domestic livestock. But a study of the economic feasibility of management for livestock and deer in combination revealed the need for more exact knowledge of their competition for forage. To provide such information, a study of the food habits of cattle, sheep, goats, and deer was initiated in May, 1959, and concluded in May, 1960. The Kerr Wildlife Management Area is a research facility of the Texas Parks and Wildlife Department located in Kerr County in the eastern portion of the Edwards Plateau. This region, encompassing some 26,000 square miles of central Texas, probably is unique in its combined productivity of deer and livestock in high densities (Hahn 1945, Taylor and Hahn 1947). Previous studies of its vegetation in relation to deer and livestock were reported by Buechner (1944), Taylor and Beuchner (1943), and Whisenhunt (1949). On the Kerr Area, particularly the pastures involved in this study, the prominent vegetal expression is a savannah of live oak, in which Ashe juniper (Juniperus ashei) and shin oak are conspicuous. Part of the Kerr Wildlife Management Area contains 10, 96-acre deer-proof pastures which are used for a long-term study of the influence of combined sheep, goat, and cattle grazing rates on deer production. Each species of animal obviously has a different influence on the welfare of the others and on the forage resource in general, but the experimental design had revealed only the gross influence of combinations of animals. Performances of deer in experimental pas-