Spatial Aspects of Imprinting and Homing in Coho 

 Salmon, Oncorhynchus kisutch 



T. P. Quinn, E. L. Brannon, and A. H. Dittman 



ABSTRACT: Analysis of seven years of coded 

 wire tag data revealed that juvenile coho salmon, 

 Oncorhynchus kisutch. released from two hatcher- 

 ies in the Lake Washington watershed return al- 

 most exclusively to their hatcheries of origin. To 

 determine if they learn the characteristics of more 

 than one water source prior to seaward migration, 

 coho salmon were reared in one of three hatcheries 

 and were released from it or, after transportation, 

 from a release site farther downriver. The locations 

 to which adult salmon returned indicated that they 

 had learned both the characteristics of their release 

 site and the hatchery where they had been held 

 prior to release. Salmon transported around much 

 of their migratory route returned primarily to their 

 release site, indicating that they needed to learn 

 sequences of odors during their seaward migration 

 in order to home in a complex river system. 



The majority of salmonid fishes that survive to 

 adulthood return to their natal site to spawn 

 (Foerster 1936; Shapovalov and Taft 1954; Arm- 

 strong 1974; Swain 1982; Quinn and Fresh 1984; 

 Berg and Berg 1987; Quinn and Tallman 1987; 

 Quinn et al. 1987). The prevalence of homing in 

 species with highly variable patterns of fresh- 

 water residence and anadromy (Rounsefell 1958) 

 suggests that the process by which the fish learn 

 the characteristics of their natal environment is 

 flexible. 



Coho salmon, Oncorhijnchus kisutch, exposed 

 to an artificial odorant prior to downstream mi- 

 gi'ation as smolts are attracted to that odor at 

 maturity (Scholz et al. 1976; Hasler and Scholz 

 1983). These and other results led Hasler and 

 Scholz (1983) to hypothesize that salmon imjirint 

 only once, immediately prior to downstream mi- 

 gration. However, there is also evidence that 

 wild coho salmon move considerable distances 

 within watersheds before migrating to sea 

 (Peterson 1982). Adult coho return to the site 

 where they emerged from gravel nests as fry, 

 not the site where they resided as smolts ( Lister 

 et al. 1981). Sockeye salmon, 0. nerka, also typi- 



T. P. Quinn and A. H. Dittman. School of Fisheries WH-10, 

 University of Washington, Seattle, WA 98195. 

 E. L. Brannon. College of Forestry. WildUfe, and Range 

 Sciences, University of Idaho, Moscow, ID 83843. 



cally home to tributaries of lakes experienced 

 only as embryos or fry, not to the lake and its 

 outlet experienced as smolts (see references in 

 Quinn et al. 1987). 



Transportation of juvenile salmon and trout 

 within river systems has had mixed effects on 

 homing. In some cases, fish captured during sea- 

 ward migi'ation, trucked to the lower Columbia 

 River, and released, generally returned to the 

 upriver rearing site (Ebel et al. 1973; Slatick et 

 al. 1975). On the other hand, displacement from 

 a hatchery to a release site downriver has often 

 resulted in returns to the release site (Jensen 

 and Duncan 1971; Vreeland et al. 1975; Cramer 

 1981; Vreeland and Wahle 1983). 



It is thus unclear whether salmon learn the 

 chemical characteristics of a single site at a spe- 

 cific developmental stage ("imprinting" by 

 smolts: Hasler and Scholz (1983)) or if they learn a 

 sequence of olfactory landmarks (Harden Jones 

 1968; Brannon 1982). By displacing smolts sea- 

 ward, we can create gaps in their migratory ex- 

 perience as a way to examine the spatial aspects 

 of olfactory learning. Specifically, we conducted 

 two experiments in which salmon were released 

 at their rearing sites or at a site downriver. The 

 locations to which these salmon returned were 

 compared among experimental gi'oups and also 

 compared to data from previous years on homing 

 and straying within the watershed. 



MATERIALS AND METHODS 



Data Analysis on Homing in 

 the Lake Washington Watershed 



There are two major hatchery sources of coho 

 salmon in the Lake Washington watershed (Fig. 

 1); the University of Washington's (UW) hatch- 

 ery and the Washington State Department of 

 Fisheries' hatchery on Issaquah Creek (Iss). We 

 inspected the Washington Department of Fish- 

 eries and University of Washington data bases 

 on coded wire tagged coho salmon and identified 

 salmon recovered at these two hatcheries for 

 return years 1979-85 to determine the extent of 

 straying within this system. 



Manuscript accepted May 1989. 

 Fishery Bulletin, U.S. 87: 769-774. 



769 



