FISHERY BULLETIN VOL. 



rivers (Berry*^; Cummings see footnote 5). The es- 

 capement of wild fish is also declining in the lower 

 Columbia River (Oregon Department of Fish and 

 Wildlife and Washington Department of 

 Fisheries 1976). 



Management agencies do not know how many 

 wild fish are caught in Oregon's troll and sport 

 fisheries, even though they strongly suspect that 

 the number of wild fish in the catch is smaller now 

 than it was 10 yr ago. If they knew the numbers of 

 hatchery and wild fish contributing to Oregon's 

 fisheries, they could recommend management 

 strategies according to the needs of the respective 

 stocks. Also, if they could distinguish hatchery 

 and wild fish caught in the fisheries at various 

 times and localities, they could determine the po- 

 tential for differentialjy harvesting the stocks. 

 Consequently, we elected to attempt separating 

 hatchery and wild coho salmon caught in the Ore- 

 gon ocean sport fishery by measuring and count- 

 ing characters of scales taken from adult fish. Our 

 objective was to determine the percentages of 

 hatchery and wild fish contributing to the Oregon 

 ocean sport fishery at different times and loca- 

 tions. 



METHODS 



Scales are a logical choice for separating hatch- 

 ery and wild coho salmon, since they have been 

 used to differentiate stocks of salmon in rivers 

 (Henry 1961), and for classifying mixed .stocks of 

 salmon caught on the high seas to continent of 

 origin (Mosher 1963; Tanaka etal. 1969; Anas and 

 Murai 1969). A review of the use of scales for 

 identification of stocks of salmon was given in 

 Major et al. (1972). Peck (1970) successfully dif- 

 ferentiated between hatchery and wild juvenile 

 coho salmon by using several scale characters. 



Scales from coho salmon of known hatchery ori- 

 gin ( identified by missing adipose fins), and scales 

 of unmarked salmon of unknown origin were col- 

 lected by personnel of the Oregon Department of 

 Fish and Wildlife from adult fish captured in the 

 ocean by sport fishermen from mid-June to mid- 

 September 1977. Samples were collected weekly 

 from eight coastal ports, listed from north to south: 

 Hammond, Garibaldi, Depoe Bay, Newport, Win- 

 chester Bay, Coos Bay, Gold Beach, and Brook- 



ings. We obtained scales from 178 adipose clipped 

 salmon and from 2,054 unmarked salmon (Table 

 1). 



Because few wild fish were available, we had to 

 use wild fish of several different brood years to 

 increase our sample. Geographic location, brood 

 year, and number of scale samples used in the 

 subsequent analysis are shown in Table 2. 



Two nonregenerated scales obtained from the 

 left side of each fish, one to four rows above the j 

 lateral line between the dorsal and adipose fins, ' 

 were mounted on gummed cards, and acetate im- 

 pressions were made by methods similar to those 

 described by Clutter and Whitesel (1956). A con- 

 scious effort was made to select the largest non 

 regenerated scales within each sample. Scale im- 

 pressions were read with the aid of a projector at a 

 magnification of 80 x. 



Based on Peck's (1970) analysis of scale charac- 

 ters of hatchery and wild smolts of coho salmon i 

 and on our analysis and observations of scale ' 

 characters from known wild adult fish and tho>i 

 from three Oregon hatcheries, we selected eight 

 characters that we believed were potentially use- 

 ful in separating hatchery from wild fish (Table 3). 

 Characters were selected based on the assumption 

 that the freshwater rearing environments for 

 hatchery and wild coho salmon are distinctly dif- 

 ferent and differences in scales between the two 

 groups would be manifest during this period. 



We selected for analysis preocean radius and 

 preocean circulus counts on the basis of results of 

 Peck ( 1970) and on data on weights of smolts being 



T.ABLE 1 —Number of adipose clipped and unmarked coho salm- 

 on sampled for scales at Oregon ports in 1977 Ports are arranged 

 from north to south. 



Table 2. — Ongin of scales of wild coho salmon, brood year, and 

 number of scale samples from each geographic location. 



"Berry, R, L. 197,5. Spawning surveys in coastal water- 

 sheds, 1974. Oreg, Dep. Fish Wildl. Coastal Rivers Inf Rep 

 75-4. P.O. Box 529, Tillamiwk, OR 97141. 



618 



