An intensive study of 2061 field records involving 1609 individuals of the Common Garter Snake (Thamnophis sirtalis), Ribbon Snake (Thamnophis sauritus) and Butler’s Garter Snake (Thamnophis butleri) has been carried out with special concentration on a large mixed population near Ann Arbor, Michigan. Snakes captured in the field were individually marked, measured, and released immediately at the point of capture, and extensive data on their food, habitat, and movements on the area recorded. The rates of growth on the three species all follow the same pattern of change. Growth is rapid in young snakes, then decreases steadily, the decrease being faster in males. First year young are readily distinguishable from older snakes until a year old, but the large group of older ones cannot be separated into age categories. Evidence from growth and reproduction of recaptured individuals indicates that both sexes in all three species may mature by their second spring. Food gathered by forced regurgitation from snakes in the field shows differences in the feeding habits. Butler’s Garter Snake feeds almost entirely upon earthworms with leeches being the only other identifiable food recorded. Amphibians from 90 per cent of the food of the Ribbon Snake, although fish and caterpillars were also taken. The Common Garter Snake eats 80 per cent earthworms, 15 per cent amphibians, and catches a few mammals, birds, fish, caterpillars,and leeches in addition. Differences in the selection of size and types of prey decrease the competition for food between the three species and between small and large individuals of the Common Garter Snake and Ribbon Snake. The food taken by a snake varies with season and availability of the prey. Garter snakes do not feed extensively when they first emerge from hibernation but they reach their feeding peak in early summer and it continues until cool weather in the fall. The food records indicate that garter snakes have little individual food preference but that within the range of the food preference of the species they will take any type available. The fish, birds and mamals eaten by the Common Garter Snake were those which were abundant and readily available. Laboratory feeding experiments generally confirmed the field records with a few exceptions. Butler’s Garter Snake will eat fish and amphibians in the laboratory, yet none were recorded from this snake in the field. Some forms, such as mice and caterpillars, eaten in the field were not taken in the laboratory by the Common Garter Snake. In spring and fall the three species occur together in high concentrations near hibernation sites. In summer, they show differences in habitat preference, Butler’s Garter Snake being most abundant in dense grasses and sedges near water, whereas the Ribbon Snake prefers areas with bushes near a body of water. In the grassy area near water and out in the water, the Ribbon Snake shows a restriction to bushes while Butler’s Garter Snake is seldom found in them. The Common Garter Snake occupies the habitats of the other two species as well as a variety of others and shows less restriction to the vicinity of water. Butler’s Garter Snake was the only species not observed in woods. The Ribbon Snake is the most frequent climber, and during July and August, 61 per cent of all records of this species were from bushes. This climbing tendency may be an escape from high ground temperatures. Observations at other places verified the results found at Cherry Hill. The habitat preferences show definite correlation with the food preference of each species. Movements of garter snakes were shown to be limited, and the term "activity range" was proposed to replace "home range" in reference to limited movement by snakes. Individuals recaptured over long periods averaged only slightly farther from the point of original capture than those recaptured after a short interval. Recaptures indicated a general restriction to one part of the area for individuals captured three or more times and 92 per cent tended to return toward the point of original capture. The activity range of between 2 and 3 acres was influenced by the habitat preference of the species and was shown to be three dimensional. Extensive marking of snakes in peripheral populations of the three species indicated very little movement to or from the Cherry Hill area. Populations of the three species on the Cherry Hill area were estimated to be 482 Common Garter Snakes, 477 Ribbon Snakes and 121 Butler’s Garter Snakes. Densities of the three species varied with the habitat type. The breeding potential varied between the three species, that for the Common Garter Snake being the largest and for Butler’s Garter Snake the smallest. Though closely associated at Cherry Hill the snakes showed no indications of any intolerance; rather, the young at least, of all three species hibernate together. The mortality rate was estimated to be approximately equal for the three kinds. Analyses of the data for the three species of garter snakes indicate a reduction in competition for food and space and help to explain why they are able to exist closely associated in large numbers at Cherry Hill.