COLLECTION AND ANALYSIS 

 OF DATA 



From the Catch 



The nature of the red salmon fishery in 

 Nushagak Bay determined the sampling scheme 

 used. Normally, several canneries operate in 

 this bay, but since fishing boats from each 

 cannery usually cover the same fishing 

 grounds, sampling was conducted from only 

 one cannery in any particular year, with a 

 few exceptions. Three main sampling loca- 

 tions were used: Alaska Packers Association 

 cannery at Clarks Point, Pacific American 

 Fisheries cannery at Dillingham, and the 

 former Libby, McNeil and Libby cannery 

 at Ekuk. 



Most of the catch is taken by drift gill nets. 

 During the days of sailboats, which prevailed 

 through 1952, fishermen operated in subareas 

 within the Bay, but inasmuch as no separation 

 is made in catch statistics between the sub- 

 areas, the entire Nushagak Bay has been dealt 

 with as a unit. Once a day, or more often when 

 fishing was good, the catches were delivered 

 either to stationary scOws or to roving power 

 scows. On high tide these catches were brought 

 into the cannery, where samples were taken. 



The elementary unit in this fishery is the 

 delivery from one fishboat. But, since boat- 

 loads were mixed aboard the receiving scow, 

 a sample was taken from the scow load as a 

 whole. Fish for the sample were collected 

 from the conveyer belt in a random fashion. 



A standard sample consisted of about 250 

 fish, of which the first 100 males and 100 

 females were measured, whenever available. 

 The sex ratio was determined from the entire 

 sample. The overall sex ratio for a fishing 

 period was determined from all the sex ratios 

 observed in the period. Also, scale samples 

 were taken from the first 20 males and 20 

 females in each sample. Lengths were meas- 

 ured from mideye to fork of tail. In 1946, the 

 first year of sampling, measurements were 

 taken with a steel tape. In subsequent years, 

 measurements were taken with several types 

 of measuring machines developed by W. F. 



Thompson. The latter method gave linear 

 measurements recorded directly on tape or 

 cards. 



In addition to the drift gill net fishery, there 

 was also a set gill net fishery, which gained 

 more importance during the later years. This 

 fishery was conducted primarily at Ekuk 

 Beach, Igushik Beach, Clarks Point, Combine 

 Flat, Coffee Point, and, in former years, Ralph 

 Slough (fig. 1). The usual sampling procedure 

 was to obtain a sample regularly from the 

 Igushik Beach fishery and one sample from 

 one other set net area, usually Ekuk Beach or 

 Combine Flat. Samples from the set net fish- 

 ery were treated in the same manner as those 

 from the drift net fishery. 



With the advent of power boats in 1953, 

 fishermen were not restricted to deliver their 

 catches to specific stationary receiving sta- 

 tions or to wait for a power scow to come by 

 to pick up their catches. They were free to 

 deliver to any company scow. This resulted 

 in considerably more mixing on the scows of 

 fish from all areas. Consequently, a sample 

 from a scow became more representative of 

 the entire catch. 



Ordinarily, sampling was continued with 

 the same intensity during the entire fishing 

 season. Occasionally, sampling was suspended 

 during the last fishing period of the season, 

 when the catches dwindled to only a fraction 

 of the catches from peak periods. The extent 

 of sampling done each year is given in table 1. 



Normally, there were eight or nine fishing 

 periods during the red salmon season in 

 Nushagak Bay. The periods were not all of 

 the same duration. During each period, all 

 measurements taken were combined into one 

 common unweighted length frequency dis- 

 tribution for each sex. The same was done 

 for the sex ratios. Since the total catch made 

 in each period was known, the number of 

 males and females caught could be easily 

 calculated. The seasonal length frequency 

 curve was obtained by weighting the length 

 frequency curve for each period by the catch 

 made In each period. 



