Understanding fishing fleet dynamics is important when using fishery dependent data to infer the status of fish stocks. We analyzed data from mandatory catch reports from the commercial lake trout (Salvelinus namaycush) fishery in Michigan waters of Lake Superior during 1929–1961, a period when lake trout populations collapsed through the combined effects of overfishing and sea lamprey (Petromyzon marinus) predation. The number of full-time fishermen increased during 1933–1943 and then decreased during 1943–1957. Addition of new fishermen was related to past yield, market prices, World War II draft exemptions, and lost fishing opportunities in Lake Huron and Lake Michigan. Loss of existing fishermen was related to declining lake trout density. Large mesh (≥ 114-mm stretch-measure) gill net effort increased during 1929–1951 because fishermen fished more net inshore as lake trout density declined, even though catch per effort (CPE) was often higher in deeper waters. The most common gill net mesh size increased from 114-mm to 120-mm stretch-measure during 1929–1957, as lake trout growth increased. More effort was fished inshore than offshore and the amount of inshore effort was less variable over time than offshore effort. Relatively stable yield was maintained by increasing gill net effort and by moving some effort to better grounds. Because fishing-up caused yield and CPE to remain high despite declining lake trout abundance, caution must be used when basing goals for lake trout restoration on historical fishery indices.