During a tagging operation conducted in 1951 on the Alsea River, 1,142 adult silver salmon were tagged with plastic Petersen-type tags and stainless steel jaw tags. An attempt was made to determine whether tagged fish released in different apparent conditions suffered differential mortalities. Under the hypothesis that the recovery of the fish in the commercial fishery was independent of the apparent condition of the fish when it was released, a significant value of chi-square was obtained. However, it was concluded that this significance was a result of sampling differences and not actually due to differences in mortality. The number of fish taken in the commercial catch each week was calculated by dividing the number of pounds in the weekly catch by the average weight of the fish in the samples for that week. It was calculated that the commercial fishery took approximately 14,000 fish during the season, which was only about 15 per cent of the run available to the fishery. This is an extremely low fishing mortality and is considerably lower than had been expected. The number of tags taken each week by the commercial fishery was calculated by dividing the number of fish in the catch each week by the number of fish per tag in the samples for the week. It was calculated that 67 tags were taken during the season by the commercial fishery. This number of recoveries was 15.5 per cent of the 438 tags available to the fishery. With the cooperation of a number of boat moorages on the Alsea River, an estimate of the sport catch of silver salmon was made. It was calculated that the lower river sport fishery took approximately 3,000 adult silvers with an additional 500 fish being taken by the bank fishery. The weight of the total catch of 3,500 fish was calculated to be about 33,000 pounds. This weight formed 20.2 percent of the combined commercial and sport catches. It was calculated that 18 tagged fish were taken in the sport catch. A total of 44 numbered Petersen and jaw tags were recovered from both dead and live fish which had reached spawning areas. This number includes three tags recovered from the spawning grounds of adjacent streams. When the tagged fish were examined according to areas recovered, it was found that Drift Creek had a smaller number of fish per tag than any other area with one exception. Recoveries appeared to be distributed throughout the drainage from any particular time of tagging. Fish enter the lower river together, then separate to the various spawning areas in the system. Several fish moved from the tagging location in the Alsea River to adjacent streams. About 7 percent of the total recoveries in Table 12 were made from streams other than the Alsea River. The Petersen and jaw tag recoveries in the samples of the commercial catch were examined for selection of the Petersen tags by the nets. The percentage by size group in the catch of fish with each type of tag was about the same as the percentage by size group with each type of tag at the time of release. Large losses of tags in the fishery would have been reflected by a higher number of fish per tag on the spawning grounds. Since the gill-nets did not appear to be selective on either type of tag, the Petersen tags, jaw tags, and tag scars were combined in the calculations involving tag recoveries. After the fish moved beyond the fishing area, many still did not move rapidly to the spawning grounds. Several were taken in tidewater or in the vicinity of tidewater 48-58 days after they had been tagged. The average time between tagging and recovery on the spawning grounds was 63 days for live fish and 70 days for dead fish. Using the method proposed by Schaefer (1951a) for obtaining a population estimate from spawning ground recoveries, an estimate of the population available to the fishery was made from the number of tags recovered by sampling the commercial catch. It was calculated that approximately 90,000 fish were available to the fishery. About 89,000 fish were calculated to be available to the fishery when the equation N = nt/s was used, where N = the population, n = the sample size, t = the number of fish tagged, and s = the number of tag recoveries. The 95 percent limits of confidence for the population estimate derived from the equation N = nt/s were calculated using the method developed by Chapman (1948). The lower limit was found to be 63,736 fish, and the upper limit was 119,371 fish. Approximately 63,000 of the fish available to the commercial fishery continued the migration to the spawning grounds. Spawning ground tag recoveries provided an estimate of the population in the lower river which was in close agreement with the estimate of the population made from tag recoveries in the commercial fishery.