We studied survival, growth, exploitation, and production from five consecutive matched plantings of brook trout (Salvelinus fontinalis) and rainbow trout (Salmo gairdneri) (8.5-9.5 inches long) in East Fish Lake, Montmorency County, Michigan, 1958-1962. Stocked fish were given varying fin clips for later recognition. Population numbers at several intervals between introductions were determined by Petersen-type estimates. Angler exploitation was tabulated from a complete creel census. Rainbow trout survival was about 98 percent from stocking in October to the following fishing season in April, whereas brook trout survival averaged only 49 percent. About one-third of brook trout deaths occurred between 15 October and the date of ice formation (approx. 15 December). Brook trout stayed in shallow water along shore more than did rainbow trout. Anglers caught 86 percent of the stocked rainbow trout, but only 39 percent of the brook trout. For each pound of trout stocked, anglers caught 3.59 lb of rainbow trout, but only 0.76 lb of brook trout. In addition to the better return on a poundage basis, rainbow trout provided a fishery throughout the angling season, whereas nearly all brook trout were caught during the first 10 days. The brook trout grew well (average increment 0.9 lb/yr), but the rainbow trout averaged 1.3 lb increment per year. Possibly a greater poundage return on rainbow trout would accrue if the beginning of the angling season could be delayed to take advantage of the early summer growing season. Fish managers generally maintain trout in lakes by stocking hatchery fish, because trout do not reproduce in most lakes. A distinct advantage in periodic stocking is that the manager has control over fish density. We can manage trout lake-fisheries most efficiently with an understanding of mortality, growth, and production. This report deals with results from comparable plantings of brook and rainbow trout over a 5year period (1958-62) in East Fish Lake, Montmorency County, Michigan. There have been many studies of trout growth, but few studies of population numbers (and weights) throughout the life of a brood year or a given stocking. Natural mortality is difficult to measure, because the investigator usually has information only on angler exploitation, but lacks corresponding total mortality data. Seasonal differences in natural mortality of trout were observed in a general way by 1 Contribution from Dingell-Johnson Projects F-21-R, F-27-R, and F-30-R, Michigan. 682 Alexander and Shetter (1961), Latta (1963), and Eipper (1964), but we need to understand the seasonal pattern more precisely. Our objectives in this study were to determine, from October releases of rainbow and brook trout, matched for size and numbers: (1) the relative survival and growth of the species; (2) the fraction of each species later creeled; (3) the temporal distribution of the catch; and (4) the biomass of fish flesh produced by the two species. We wish to acknowledge the assistance of past and present members of the Hunt Creek Trout Research Station staff in the collection and tabulation of data. The fine cooperation of the trout fishermen in reporting their angling results is appreciated. G. P. Cooper, M. H. Patriarche, and W. E. Schaaf reviewed the manuscript and offered many helpful suggestions.