nia Dungeness crab because of the evidence that 

 California catch statistics underestimate the number 

 of fish actually in the state during spring. These land- 

 ings provide a straightforward measure of brood sur- 

 vival that is independent of distribution and local 

 catchability. 



Even though the relatively low O.P.I, area landings 

 in 1961, 1962, 1963, 1977, and 1980 are comparable 

 with the predecline era, the general pattern of cor- 

 respondence in Figure 5 is consistent with an in- 

 crease in the coho salmon predation rate on Dunge- 

 ness crab megalops. Within the framework of the 

 predator-prey hypothesis, the association of low O.P.I, 

 area coho catches during the early years of the hatch- 

 ery era with reduced Dungeness crab survival would 

 indicate that a relatively small number of hatchery 

 coho can effectively surpress megalops survival. This 

 is particularly apparent when it is considered that 

 hatchery production was at a minimum during 

 the 1961-63 period and wild fish still dominated the 

 catch (Oregon Department of Fish and Wildlife 

 1982). 



The proposed impact of hatchery coho salmon on 

 the Dungeness crab resource is best explained by the 

 differences in the "functional response"^ of wild and 

 hatchery coho salmon. In controlled behaviorial ex- 

 periments, Glova (1978) found that hatchery fry 

 (43-88 mm) were largely nonterritorial, exhibiting 

 a stronger tendency to aggregate than the wild fry. 

 This behavioral pattern is believed to be the direct 

 result of the unnaturally high densities found in 

 hatchery operations. If adult hatchery coho retain 

 this behavior, the tendency for Dungeness crab 

 megalops to aggregate or "swarm" in coastal sur- 

 face waters (Lough 1976) would theoretically make 

 them more susceptible to predation (Eggers 1976). 

 Also a reduction in the number of "search images" 

 available to hatchery fish is believed to result in a 

 more homogenous diet (Sosiak et al. 1979). Under 

 these circumstances Dungeness crab megalops may 

 become a more important component of the hatch- 

 ery coho salmon diet. 



The apparent good survival of the 1961-66 year 

 classes in northern California (Fig. 5) suggests that 

 the majority of the hatchery coho salmon produced 

 during those years concentrated to the south of that 

 population during the period when Dungeness crab 

 megalops are most abundant. This supposition, 

 together with recently acquired evidence that the 

 central California Dungeness crab population is at 



least partially dependent on the recruitment of 

 southward drifting megalops (Hatfield 1983; Reilly 

 1983b), further suggest that the theoretical 

 predation zone critical to the central California 

 population lies somewhere in the region of strong 

 upwelling and high productivity between the two 

 populations (Fig. 1). Not surprisingly, commercial 

 fishermen have found coho salmon concentrated 

 either before or early in the season in this region. 

 The coho salmon stocks initally released during the 

 early 1960's may possess an inate affinity for these 

 waters. 



Northern California landings of the Dungeness 

 crab declined again during the 1970-71 season (Fig. 

 2). This period of low landings is apparently due to 

 poor survival of the 1967-71 year classes (Fig. 5), 

 which cannot be readily explained by an extended 

 period of warmer than normal water. The various 

 hypotheses to explain the northern California fluc- 

 tuations notwithstanding, it is possible that hatchery- 

 reared coho salmon began to limit Dungeness crab 

 survival in northern as well as central California, 

 concomitant with increased hatchery production^ 

 and/or environmental caused changes in distribution. 

 There is some evidence from coho tagging that sup- 

 ports this supposition. 



O'Brien (1973) reported that 17.3% of his return- 

 ed tags were found in Oregon and Washington hatch- 

 eries during the 1971 season, whereas in 1972 only 

 3% were found in the hatcheries. An exceedingly 

 strong 1972 Dungeness crab year class in northern 

 California (Warner 1984) is in direct contrast with 

 the very weak 1971 year class (Fig. 5) and is inversely 

 related to the small number of tags found in hatch- 

 eries during the 1972 season. The small percentage 

 of hatchery returns in 1972 suggest that there were 

 fewer hatchery coho available for tagging in the 

 northern California area during the 1972 season, and 

 this could indicate relatively poor survival of hatch- 

 ery fish throughout the O.P.I, area. It should be 

 remembered that hatchery-reared coho theoretical- 

 ly have a much larger effect on Dungeness crab sur- 

 vival than wild fish. 



Between 1972 and 1977 (Fig. 5), O.PI. area coho 

 survival and northern California Dungeness crab 

 survival became more erratic. The association of 

 relatively good Dungeness crab survival with good 

 coho landings in 1974 and 1976 may, however, only 

 indicate that coho were farther south than usual. 

 McLain and Thomas (1983) showed that both 1973 



^In predator-prey theory "functional response" is defined as the 

 relationship between the rate at which individual predators con- 

 sume prey and the density of that prey (Holling 1959). 



^The number of hatchery-reared coho salmon released in the O.P.I, 

 area increased from 7.5 million fish in 1960 to 60.8 million fish in 

 1981 (Oregon Department of Fish and Wildlife 1982). 



689 



