"Tillamook Bay chum salmon are caught commercially by gill-nets, both set and drift, and from 1928 through 1949 the landings have averaged 819,689 pounds per season. More chum salmon are caught on Tillamook Bay than on the rest of the Oregon coastal rivers combined. These fish enter the ocean only a few weeks after hatching and return to their native streams to spawn, apparently in the third, fourth, or fifth year of their life. There are numerous factors such as predation, fishing intensity, success of spawning, which may have some effect on the production of Tillamook Bay chum salmon. There was no apparent relationship between total chum landings and lumber production 4 and 8 years previous. It is possible that a relationship between these two factors may exist for some period other than 4 or 8 years previous, but this was not investigated further at this time. Numerous factors, mainly various combinations of stream flow data, were correlated with the size of the commercial landings of chum salmon in an attempt to determine a "predictor" of future abundance. An apparent good relationship, with b = 2.05944 (P = 01), between total chum landings and the minimum stream flow occurring between January 15 and March 20 three years previous was discovered. The regression equation obtained is y = -493.6355 + 2.05944x. The regression of the size of the commercial landings of chum salmon on three variables was computed. These three variables were: (1) minimum stream flow occurring between January 15 and March 20 three years previous; (2) highest air temperature occurring during January or February 3 years previous; and (3) the maximum stream flow occurring between November 1-10 four years previous. The regression equation obtained is y = 346.4554 + 97.3148x[subscript 1] + 66.1037x[subscript 2] 77.8188x[subscript 3]. The probabilities attached to the results of all the regression functions which were computed are biased because of the manner in which the data were selected to obtain the best relationship. Furthermore, although an apparent good relationship was determined between minimum flows and the size of the commercial landings, it must be realized that these results have some very definite limitations to their actual values. In addition to the biased nature of the probabilities, is the fact that these results have been derived from a very limited amount of data. Also, these analyses have been based on some very important assumptions which, if not true, would almost completely invalidate the results obtained. These assumptions are: (1) catch = a constant X abundance; (2) all the fish mature at 4 years of age; (3) these data can be fitted to a linear model along with the other general assumptions about regression data; (4) fishing intensity has remained relatively constant (implied in 1); and (5) methods of recording flow data have remained consistent from year to year."--Summary.