With the opening of the 1953 beaver trapping season in Michigan, Conservation Department field personnel began receiving reports of dead beaver. From the number of reports it was evident that an epizootic had occurred among these animals in the western counties of the Upper Peninsula. Trappers found carcasses floating in flowages and along stream margins, and in several instances sick and weakened beavers were found on land short distances from water. Approximately 180 dead or sick beavers were reported during the trapping season. The greatest concentration of carcasses occurred on the headwaters of the East Branch Presque Isle River in Gogebic County. Obviously, these scattered, voluntary reports received by the Game Division represent only a small fraction of the total number of animals that must have died. Conditions that prevailed in flowages not readily accessible to trappers were, of course, not known. Also, as subsequent field observations have shown, the activities of scavengers quickly scatter and destroy the remains of dead beavers. Concurrent with this situation in the Upper Peninsula, trappers in adjoining counties in northern Wisconsin were reporting a die-off of beavers (Knudsen, 1953). Gross pathological examinations of carcasses by L. D. Fay in some instances disclosed lesions characteristic of tularemia, whereas in many carcasses there was no obvious diseased condition. Attempts to isolate Pasteurella tularensis from carcasses, or from water and mud collected from flowages where die-off had occurred, were unsuccessful. Laboratory animals were taken into the field and inoculated with freshlycollected water to insure infection. Post mortem materials from three fresh carcasses were sent to Dr. William Jellison, U. S. Public Health Service Laboratory, Hamilton, Montana, for examination. He, too, was unable to demonstrate tularemia. However, during the trapping season three serologically-confirmed cases of this disease occurred among Michigan trappers in the Upper Peninsula. There were several other suspected human cases but blood samples for serological confirmation were not taken. In all three diagnosed cases, the trappers had found and skinned either dead or sick beavers. Thus this evidence, in spite of the negative bacteriological findings, was highly suggestive that tularemia was responsible, at least in part, for beaver mortality. Field studies made during the summer of 1953 by the authors showed that die-off was still in progress. It appeared that muskrats were also being affected as carcasses of these animals were found in a flowage where beavers had died. However, the rapid decomposition of carcasses during warm weather, which made it nearly impossible to collect material suitable for bacteriological examination, together with the prompt consumption of them by scavengers, especially bears, made it difficult to determine 1 Paper #3 Tularemia Project, School of Natural Resources; in cooperation with the Department of Bacteriology and School of Public Health, University of Michigan, and the Michigan Department of Conservation; supported by Research Grant E688, National Institutes of Health and Faculty Research Grant #914, Horace H. Rackham School of Graduate Studies, University of Michigan. This paper has been revised since its acceptance for publication to include 1955 observations.