presented because the munljer of i-ecoveries was 

 small. 



Table 14. — Marked to unmarked ratios for hatchery returns 

 of 1961-brood chinook salmon by type of mark and age 

 of fish 



Origin 



Age (yeais) 



All 

 ages 



.- MaTkedlU7inmrked 



Ad-R.M - All study hatcheries 0.059 0.063 0.075 (1.106 0.067 



Ad-RV-RM... Kalama .016 .046 .092 .047 



Ad-LV-RM... Spring Creek .046 .038 .050 .062 .041 



Ad-KM Selected hatcheries 1 065 .067 .085 .081 .072 



1 Cascade, Oxbow, Little White Salmon, and Spring Creek Hatcheries. 



The increase in ratios witli age seems to indicate 

 that marking had some delaying effect on the age 

 of maturity. This effect was much greater for fish 

 marked at Kalama than for fish marked at 

 Spring Creek, although a similar type of mark 

 was used. From the standpoint of survival, the 

 ratios for all ages combined are smaller than 

 those at release (table 4), thus indicating lower 

 total survival for marked fish. The survival of 

 Ad-RM marked fish, for example, was (0.067/ 

 0.1193)100=56.2 percent of that for the un- 

 marked fish. 



If the differences between the ratios of marked 

 and unmarked fish at times of release and return 

 are interpreted as indicative of delayed maturity 

 or lower survival of marked fish, it is necessary 

 to assume that all unmarked fish returning to a 

 hatchery originated from that hatchery. It is 

 apparent that straying^ (provided the probability 

 of straying is the same for marked and unmarked 

 hatchery fish) reduces the marked to umnarked 

 ratio for any given year of return. Thus, the indi- 

 cated survival of marked fish relative to the sur- 

 vival of unmarked fish is too small. Also, if the 

 probability that nonhatchery fish stray into a 

 hatchery changes each year or is related to the age 

 of fish, then the dilution of unmarked hatchery 

 fish by nonhatcheiy fish in the return is dispro- 

 IJortionate from year to year and the change in the 

 marked to unmarked ratios with age of return is 

 impossible to interpret. 



Some of the hatcheries, such as Kalama Hatch- 

 ery, are located on streams in which natural 



spawning occurs. In these cases, fish which ai'e 

 spawned at the hatcheries and comprise what we 

 term "hatchery returns" are simidy a selected 

 sample (of various sizes in proportion to the total 

 return) of adults returning to the stream. 

 Marked to unmarked ratios would, therefore, be 

 difficult to interpret in terms of the effects of 

 marking on survival and age of maturity. 



Only 5 of the 12 hatcheries (Bonneville, Cas- 

 cade, Oxbow, Little White Salmon, and Spring 

 Creek) are on streams that do not support nat- 

 ural spawning po[)ulations. Marked (Ad-RM) 

 to unmarked ratios in the returns to four of these 

 hatcheries, Bonneville Hatchery excluded,'' are 

 presented in table II. The change in the ratio of 

 marked to unmarked fish with ;ige is not as pro- 

 nounced as for the 12 hatcheries combined. In 

 addition, the ratio, 0.072, for all ages is somewhat 

 larger than it was for all 12 hatcheries, indicating 

 that survival of marked (Ad-RM) fish relative 

 to unmarked fish was (0.072/0.1193)100 = 60.4 

 l^ercent. 



Even for the four selected hatcheries, however, 

 straying of nonhatchery fish into hatcheries is a 

 possible source of error. Unfortunately, a direct 

 measure of the extent of this straying is not avail- 

 able. General indications of straying of nonhatch- 

 ery fish are obtained from obserwations of marked 

 fish straying away from the hatcheries. Recoveries 

 of specially marked fish released at Elokomin, 

 Kalama, Oxbow, and Spring Creek Hatcheries are 

 presented in table 15. Of the recoveries of marked 

 fish at hatcheries and on spawning grounds, 71.4, 

 6.3, 38.0, and 4.9 percent of Elokomin, Kalama, 

 Oxbow, and Spring Creek fish were recovered at 

 places other than the release site. The higher per- 

 centage of strays from Elokomin and Oxbow 

 Hatcheries may possibly be explained by the fact 

 that 75 percent of the fish released at Elokomin 

 and 30 percent of the fish released at Oxbow were 

 the progeny of fish spawned at Spring Creek 

 Hatchery. 



Some indication that younger fish stray more 

 than older fish can be seen in table 15. Groves, Col- 

 lins, and Trefethen (1968) tagged two groups of 

 Chinook salmon at Spring Creek Hatchery and 

 released them into the main stem of the Columbia 



f" The term "straying" is uspcl here in the sense that fish spawn 

 or are spawned at a location other than their point of origin as 

 fingerlings. Indeed, straying into .a hatchery may be different from 

 straying into a natural spawning area where the fish are free to 

 leave again. 



^- Bonneville Hatchery is excluded becanse it is immediately 

 below Bonneville Dam. The potential for strayin.K may he greater 

 if the fish are delayed in their upriver iiassage at the dam. 



COXTRIBUTIOX OP COLUMBIA RIVER HATCHERIES TO FAIJ. CHINOOK SALMON HARVEST 



381 



