FISHERY BULLETIN: VOL. 86, NO. 4 



probably valid, to assume that all prey are equally 

 available to any predators captured in the immediate 

 vicinity a short time later. The amount of time dur- 

 ing which the prey subsequently remain equally 

 available to predators will likely vary from one situa- 

 tion to the next. If the prey are very mobile, prob- 

 ably some will eventually become less or more acces- 

 sible to predators than others. This possibility can 

 be eliminated or at least minimized by keeping the 

 experiment short enough to ensure that the prey do 

 not have sufficient time to segregate or move away 

 from the release site. The magnitude of this prob- 

 lem and thus the appropriate duration for each ex- 

 periment may be assessed by recapturing some of 

 the tagged prey after the release. The experiment 

 should be terminated when the characteristics of the 

 recaptured prey begin to diverge significantly from 

 those of the original prey population. 



Determining the prey preference of predators in 

 the wild is a concern to many biologists. All methods 

 of determining the prey selectivity of predators in 

 the wild are, and will likely continue to be, hampered 

 by the complexity of the related problem of deter- 

 mining the relative "availability" of prey to pred- 

 ators. The advantage of the method proposed here 

 is that it allows the investigator to control some of 

 the major variables that are known to affect the 

 availability of prey. The most important character- 

 istics of the prey (species ratios, abundance, size 

 ranges, etc.) can be determined before any preda- 

 tion occurs and in many cases can also be precisely 

 controlled. The predators remain free to feed on all 

 types of prey in the study area, but for the purposes 

 of the investigator, the choice of prey can effectively 

 be reduced to those with known characteristics and 

 origin. This is a major advantage when compared 

 with the more traditional approach of calculating 

 selectivity indices, as it eliminates the need to deter- 

 mine the relative abundances of prey in the field. 

 It also substantially reduces the ambiguity associ- 

 ated with interpreting selectivity indices for highly 

 mobile predators, where typically there is little or 

 no information available concerning the area 

 traveled by the predator during the feeding period 

 and thus what prey were actually available to the 

 predator. 



The specific goal of the two experiments reported 

 here was to determine if predators were size selec- 

 tive when preying on juvenile pink salmon during 

 the early sea-life period. The results indicate that 

 yearling coho salmon were the dominant predator 

 of juvenile pink salmon at two locations, one in 

 Masset Inlet and the other in Masset Sound, and 



that the average size of juvenile pink salmon con- 

 sumed by these predators was significantly less than 

 the average size of pink salmon that were released. 

 These results are consistent with those obtained 

 from two other independent approaches and suggest 

 this method may be a viable and cost-effective alter- 

 native for determining the prey preferences of pred- 

 ators in the wild. It may be particularly useful for 

 assessing prey preferences of predators feeding on 

 juvenile salmon near hatchery facilities in Canada 

 and the United States, where millions of juvenile 

 salmon are currently tagged and released each year. 



ACKNOWLEDGMENTS 



Suggestions and comments from an anonymous 

 reviewer are greatly appreciated. Robin LeBrasseur 

 and Owen Kennedy assisted with the supervision of 

 the field experiments and analyses of the stomach 

 samples. Bruce Patten, Lui Marinelli, Tom Poole, 

 Josette Weir, Rick Hobbs, Ted Carter, and Bob 

 Hungar helped to capture, tag, and sort the juvenile 

 salmon and assisted in the beach seining and pro- 

 cessing of samples collected in the various experi- 

 ments. Trans-Provincial Airlines generously per- 

 mitted unlimited use of their seaplane wharf and 

 other facilities at Masset, B.C. to conduct the sec- 

 ond experiment during 6 and 7 June 1985. 



LITERATURE CITED 



Anderson, V. L., and R. A. McLean. 



1974. Design of experiments: A realistic approach. Marcel 

 Dekker Inc., N.Y. 

 Brett, J. R. 



1965. The relation of size to rate of oxygen consumption and 

 sustained swimming speed of sockeye salmon (Oncorhynchus 

 nerka). J. Fish. Res. Board Can. 22:1491-1501. 



BURGNER, R. L. 



1962. Studies of red salmon smolts from the Wood River 

 Lakes, Alaska. In T. S. Y. Koo (editor), Studies of Alaska 

 red salmon. Univ. Wash. Publ. Fish. N.W. I(6)-251-316. 

 Cock, M. J. W. 



1978. The assessment of preference. J. Anim. Ecol. 47:805- 

 816. 

 Cochran, W. G. 



1977. Sampling techniques. 3d ed. John Wiley and Sons, 

 N.Y., 428 p. 

 Curio, E. 



1976. The ethology of predators. Springer- Verlag, N.Y., 250 



P- 

 Hargreaves, N. B. 



In press. Predation of juvenile pink (Oticorhynchus gorbuscha) 

 and chum (0. keta) salmon in Masset Inlet, B.C. Can. J. 

 Fish. Aquat. Sci. 

 Hargreaves, N.B., E. W. Carter, and R. J. LeBrasseur. 

 1987a. Beach seine catches of juvenile salmon and other fish 



770 



