Responses of migrating adult sockeye salmon (Oncorhynchus nerka) to population-specific odours

Transcription

1 Responses of migrating adult sockeye salmon (Oncorhynchus nerka) to population-specific odours C. GROOT, T. P QUINN', AND T. J. HARA~ Department of Fisheries and Oceans, Fisheries Research Branch, Pacific Biological Station, Nanaimo, B.C., Canada V9R 5K6 Received August 9, 1985 GROOT, C., T. P. QUINN, and R. J. HARA Responses of migrating adult sockeye salmon (Oncorhynchus nerka) to population-specific odours. Can. J. Zool. 64: The ability of homing sockeye salmon (Oncorhynchus nerka) to discriminate between two populations of their own species on the basis of odours was tested in a Y-maze choice apparatus and by electrophysiological recordings from the olfactory rosette. The tests were performed with adult sockeye captured at the entrances to Great Central Lake and Sproat Lake on Vancouver Island, British Columbia, Canada. In behaviour experiments, Great Central Lake sockeye salmon were significantly attracted to chemical traces of their own population but no such preference was evident for the Sproat Lake sockeye salmon. The electrophysiological experiments indicated that Sproat Lake sockeye salmon discriminated between water conditioned by their own population from water conditioned by Great Central Lake sockeye. Great Central Lake sockeye salmon failed to show differential responses to the two population waters. The experiments revealed that the olfactory system of migrating adult sockeye salmon was sensitive to amino acids and a bile acid, but that the major active components in water responsible for olfactory stimulation were bile-acid-like components. We conclude that adult sockeye salmon can discriminate between population specific odours, but that differences exist between stocks. GROOT, C., T. P. QUINN et R. J. HARA Responses of migrating adult sockeye salmon (Oncorhynchus nerka) to population-specific odours. Can. J. Zool. 64: La capacit6 qu'ont les saumons sockeye (Oncorhynchus nerka) anadromes de distinguer deux populations de leur propre espkce a l'odeur a it6 mise a 1'6preuve par l'utilisation d'un labyrinthe en Y et par l'obtention d'enregistrements 6lectrophysiologiques au niveau de la rosette olfactive. Les saumo?s qui ont servi a l'exp6rience 6taient des saumons adultes captur6s a leur entr6e dans le lac Great Central et le lac Sproat dans 1'Ile de Vancouver, Colombie-Britannique, Canada. Au cours des exp6riences sur le comportement, les saumons sockeye du lac Great Central ktaient attirks de fa~on significative par les kmanations chimiques de leur propre population, mais cette prkfkrence ne s'est pas manifestke chez les saumons du lac Sproat. Les enregistrements 6lectrophysiologiques indiquent que les saumons du lac Sproat peuvent faire la diffkrence entre de l'eau conditionnke par leur propre population et de l'eau conditionnee par des saumons du lac Great Central. Les saumons du lac Great Central ne semblent pas faire le diffkrence entre les eaux des deux populations. Les exp6riences ont montrk que le systkme olfactif du saumon sockeye adulte migrateur est sensible aux acides aminks et a un acide biliaire, mais les principales substances actives prksentes dans l'eau et responsables de la stimulation olfactive sont des substances apparentkes a l'acide biliaire. 11 semble donc que les saumons sockeyes adultes soient capables de distinguer des odeurs spkcifiques aux populations, mais cette capacitk varie d'un stock a l'autre. [Traduit par la Revue] Introduction Two hypotheses have been proposed to explain the role of olfaction in homing of salmonids. The "imprinting hypothesis," (Hasler 1966; Hasler and Scholz 1983) holds that specific odours emanating from rocks, plants, soil, and other riverine features are learned by juvenile salmon during crucial developmental stages. Exposure to these odours releases upstream swimming in adult salmon (Johnsen and Hasler 1980). Nonolfactory mechanisms are assumed to guide salmon from oceanic feeding grouds to coastal water. The "pheromone hypothesis" (Nordeng 1971, 1977; reviewed in detail by Stabell 1984) states that the homeward migration of adult salmonids from salt water is initiated and directed by population-specific pheromone trails laid down by juveniles on their way to sea prior to the adult migration. In the final stages of adult migration, pheromones released by resident young in freshwater localities also guide adults. Additionally, fish at the rear of the homeward run may be guided by pheromones from adult conspecifics ahead. The response to pheromones is assumed by Nordeng (1977) to be inherited, but conclusive evidence is presently lacking. If pheromones are part of the total bouquet of odours that juvenile 'present address: School of Fisheries, WH- 10, University of Washington, Seattle, WA, U.S.A ~resent address: Department of Fisheries and Oceans, Freshwater Institute, Winnipeg, Man., Canada R3T 2N6. salmon learn prior to emigration, then there is no basic difference between the two hypotheses (Selset and D~ving 1980; Horrall 1981 ; Hara et al. 1984), except that one refers specifically to olfactory components from conspecifics in the learning process. Nordeng (1977) proposed the "pheromone hypothesis" as a result of migration studies with Atlantic salmon (Salmo salar), anadromous char (Salvelinus alpinus), and trout (Salmo trutta). Positive responses to population-specific odours of smolts were subsequently reported by Selset and Doving (1980) for mature anadromous char. Nordeng (1977) also suggested that migration data of Pacific salmon (genus Oncorhynchus) support the pheromone hypothesis. Initial experimental tests of this hypothesis using coho salmon (0. kisutch) indicated attraction to the odours of juvenile conspecifics, but failed to demonstrate populationspecific responses (Quinn et al. 1983) or preference for pheromones over home water lacking pheromones (Brannon et al. 1984). However, recent experiments have shown that juvenile coho salmon can distinguish siblings from nonsiblings (Quinn and Busack 1985) and can distinguish between populations based on chemical cues (Quinn and Tolson 1986). The objective of this study was to test whether migrating adult sockeye salmon can discriminate between populations on the basis of odours. We conducted behavioural experiments using a Y-maze choice apparatus and made electrophysiological recordings from the olfactory rosette to test responses to popula-

2 GROOT ET AL. 927 FIG. 1. Map of Vancouver Island showing the locations where sockeye salmon were captured during their migration into Great Central and Sproat lakes and were tested at the Rosewall Creek Hatchery and the Pacific Biological Station. tion-specific odours. We also discuss the implications of the findings for interannual variability in juvenile and adult sockeye salmon migration routes in coastal waters of British Columbia. Material and methods Experimental animals We used adult sockeye salmon returning to Great Central () and Sproat () lakes for both the behaviour and electrophysiological experiments. These neighbouring lakes in central Vancouver Island drain through the same river (Somass River) and inlet (Alberni Inlet) to sea (Fig. 1). We chose these populations because (i) studies using parasites for stock identification have shown that these sockeye populations home very accurately to their respective lake systems (Margolis 1982), (ii) the adult sockeye of both lakes return comparatively early (mid-june to early August) and therefore their movements overlap with the outmigrating smolts (which leave between early April and early June) in the nearshore area, and (iii) fishways near the outlets of both lakes give easy access to migrating adult sockeye. Twenty-five to 30 adult sockeye were captured weekly and transported to the Rosewall Creek experimental hatchery (90-min journey) (Fig. 1). Great care was taken to minimize the time out of water to reduce stress. Water in the transportation tanks was cooled to about 3"C, which greatly reduced the activity of fish during transport. At the Rosewall Creek Hatchery sockeye from the two lakes were held in separate 245 cm diameter fiberglass tanks in groups of Five fish from each group were selected at random and kept in a 180-cm fiberglass tank to provide the odours for the experiments. All tanks were supplied with 8OC well water, which may be considered devoid of fish odours. Each fish was tested singly and only once and after the test held in a separate tank. At the end of a weekly testing period all sockeye were returned to their respective lake systems. For the electrophysiological experiments, sockeye were transported from the Rosewall Creek Hatchery to the Pacific Biological Station at Nanaimo, British Columbia, as required (Fig. 1). Behaviour experiments The choice test apparatus consisted of two painted plywood troughs 488 x 122 x 122 cm (Fig. 2). One-half of each trough was divided lengthwise into two equal compartments, 60 cm wide and 244 cm long. Water flowed in at the head of each compartment at a rate of 90 Llmin. A gate at the end of the trough maintained a water level of 30 cm. The odour sources were siphoned out of the tanks holding five sockeye each from Great Central and Sproat lakes and from a tank of unconditioned well water, which served as a control (blank water). Odour or blank water was introduced in the outside comer of each compartment at 10 Llmin. (Fig. 2). Any paired combination of the two odour sources and blank water could be supplied by positioning the siphon nozzles into holders in the appropriate troughs. The divided part of each trough was covered with a black plastic screen over a distance of 245 cm during the test. The screen left a gap of 30 cm at the head of the trough to allow some light to enter the compartments so that the behaviour of the fish near the water inlet sources could be easily observed. The behaviour experiments were conducted between July 11 and August 10, Sockeye were introduced by dip net into the lower central compartments after the tanks had been flushed for 10 min. During the 30-min tests, records were kept of the time fish entered each compartment, nosing (lifting nose or head out of water as far as operculars), jumping (moving body at least halfway out of water), and fluttering (rapid swimming movements against the walls). All experiments were performed during daylight hours between 1000 and Sockeye from each lake were exposed to two odour combinations: their own population against blank water and their own population against the other population. For each test series the odour sources were switched after every trial. At least 40 sockeye from each lake were tested in each experimental pairing of odours. After introduction into the experimental troughs, the sockeye were often inactive on the bottom for a few minutes before surfacing to gulp air and swim about. At this time, the behaviour recordings started and

3 CAN. 1. ZOOL. VOL. 64, 1986 FIG. 2. The experimental setup for the behaviour tests showing the two choice (Y-maze) apparatus and the three head-tanks used to supply any paired combination of two population-specific odours and blank water. the duration of immobility was noted. Fish that were inactive on the bottom for 25 min or longer were not included in the analysis but were used to assess motivational differences between stocks. The four levels of responses recorded (entrances, time spent, noses, and jumps in left and right compartments) were analyzed with the nonparametric Wilcoxon matched-pairs signed-rank test (Ferguson 1971). Electrophysiological experiments The olfactory responses of migrating adult sockeye salmon to odours were examined by recording the electro-olfactogram (EOG) using essentially the same technique described previously by Evans and Hara (1985). The fish ( cm in length) were immobilized by intramuscular injection of Flaxedil (gallamine triethiodide, 4 mg/kg body weight) after brief anesthetization with tricaine methane sulfonate (1:10000). Each fish was secured in a holding apparatus placed in a flow-through trough and the gills were perfused with water (19.5OC; 1.5 Llrnin). Following removal of the overlying skin and cartilage of the nasal sac, an olfactory rosette was perfused (18 ml/min) with water from a continuous flow delivery tube. EOG responses were recorded with an Ag-AgC1 electrode (WPI Type EH- 1s) filled with 3 M KC1 via a saline gelatin-filled (8%) capillary pipette (tip diameter, km), amplified with a DC preamplifier (Grass 7P1), and recorded on a pen recorder (Grass 7B Polygraph). An indifferent electrode of the same type was placed on the surface of the skin adjacent to the olfactory cavity. Constant volumes of stimulus solutions were drawn into disposable glass pipettes, 0.5 ml of which was injected via an automatic stimulatory apparatus into a stream of water perfusing the rosette. The final test solutions were freshly prepared just prior to each application with the same water used to perfuse the gills and nares. Water used to perfuse the gills and olfactory rosette and to prepare amino acid and bile acid solutions was dechlorinated Nanaimo city water piped into the laboratory through a plastic plumbing system. Four types of tests were conducted: (i) responses to selected amino acids and bile acid, (ii) responses to waters conditioned by and adult sockeye salmon, (iii) responses to waters conditioned by and sockeye smolts, and (iv) reponses to adult fish odours while cross-adapting to amino acids or bile acid. For cross-adaptation experiments the water delivery apparatus was connected via a three-way stopcock to a bottle containing a perfusing chemical suspended in a water column. A constant flow of the perfusing chemical to the olfactory rosette was thus obtained through a gravity feed system. Adult sockeye salmon test waters were collected from the same odour-source tanks with five fish each that were used for the behaviour experiments at the Rosewall Creek Hatchery. The waters with and sockeye odours were transported in 20-L plastic pails to the Pacific Biological Station the day before the electrophysiological experiments. For the smolt waters, 200 g of sockeye smolts (42 and 33 smolts), caught during their emigration from Great Central and Sproat lakes in May, were held for 10 min in 6 L of Rosewall Creek Hatchery well water. The volume of each sample was then diluted to 18 L and divided into three containers and kept frozen until the time of testing. The experiments were conducted between July 7 and August 5, Results Behaviour experiments Sockey salmon from Great Central Lake were attracted to water conditioned by adults of their population ( water) over unconditioned well water (blank water) (Table 1, A). Significantly more time was spent in the channel containing

4 GROOT ET AL. TABLE 1. The four levels of responses (entrances, time spent, jumps, and noses in left and right compartments) of and sockeye salmon in paired choice tests when exposed to odour combinations of their own population against blank water and their own population against the other population Sockeye origin Odour combination Date Entrances Time(s) Jumps Noses N (A) (B) (C) (Dl blank Aug 2-8 July 25- Aug. 2 Aug 2-9 July July 3 1- Aug. 1 NOTE: N indicates the number of fish included in the analysis. Levels of signific 0.001; **, p; < 0.01; *, p TABLE 2. The mean time (seconds) before and sockeye salmon started to swim after introduction in the choice apparatus Own odour Own odour Sockeye Choice against against other origin condition blank water population Sockeye 185 (42) 91 (40) Sockeye 260 (41) 309 (40) NOTE: The numbers of fish analyzed are in parentheses. sockeye odours, and more entrances and nosing occurred on this side as well. However, the numbers of jumps in the two channels did not differ. The 42 salmon were tested over two time periods, July and 2-8 August. In general, the preference for water was stronger in the earlier tests than in the later ones: 56% vs. 54% of entrances, 63% vs. 55% of time spent, 83% vs. 57% of nosing, and 75% vs. 58% of jumps. When water was paired with Sproat Lake () water, fish displayed a significant preference for the chemical traces of their own population (Table 1, B). The 41 salmon in this series were tested from 11 to 19 July and 25 July to 2 August, but unlike the results with blank water, preferences for water over water were somewhat stronger in the second set of tests. In contrast to the strong responses of sockeye, Sproat Lake fish did not enter, jump in, or spend more time in the channel when it was paired with blank water (Table 1, C). However, more nosing occurred in the channels. When and water were paired, the Sproat Lake sockeye displayed,ante of the nonparametric Wilcoxon matched-pairs signed-rank test; ***, p < no preference. Some attraction to water was indicated in the first sets of vs. blank water and vs. water in the amount of time spent in the two channels, but no attraction was evident in later tests (Table 1, D). The Sproat Lake sockeye were generally less active than the fish. After introduction into the experimental troughs they remained on the bottom significantly longer than fish before moving (Table 2; ANOVA, p < 0.01) and more sockeye remained for longer than 25 min or did not move at all within the 30-min test period than fish (9.9% vs. 0%). sockeye also started to move sooner under the vs. test condition (average 91 s) than under vs. blank (average 185 s) odour choice conditions (Table 2; ANOVA, p < 0.05). There was no significant difference in the times to first movement for sockeye in the two odour combinations (Table 2; ANOVA, p > 0.05). Electrophysiological experiments The response magnitude of the EOG to amino acids (Larginine, L-cysteine, and L-serine) and taurocholic acid increased nearly exponentially with a logarithmic increase in the stimulus concentration (Fig. 3). Threshold concentrations of injected samples for these chemicals ranged between and M. sockeye salmon showed significantly lower responses to adult water than water (Fig. 4). The EOG response levels of fish to adult waters of the two populations were essentially the same. sockeye demonstrated significantly lower EOG activities to water conditioned by smolts than smolts (Fig. 5). However, there was no significant difference in the responses of fish to water conditioned by smolts of the two populations. EOG responses of

5 0 L-ARGININE r L-CYSTEINE A L-SERINE CAN. f. ZOOL. VOL. 64, SOCKEYE -SOCKEYE TAUROCHOLIC ACID LOG MOLAR CONCENTRATION FIG. 3. Dose-response relationships of adult sockeye salmon electro-olfactograms (EOG) to three amino acids and one bile acid. Each point is the mean response of 3 to 5 fish. The response magnitude is normalized to 100% for lo-' M L-serine. - ADULT WATER -ADULT WATER ADULT WATER ADULT WATER 1, L t L t.l - o 0'1 0' CONCENTRATION OF FISH WATER (DILUTION) FIG. 4. Dose-response relationships of and adult sockeye salmon electro-olfactograms (EOG) to varying concentrations of waters conditioned with and adult sockeye salmon odours. Each point is the mean response of 4 to 5 fish. The response magnitude is normalized to 100% for lo-' M L-serine. sockeye to water from their own population were nearly eliminated when the olfactory organ was cross-adapted to lo-' M taurocholic acid. However, the reduction was less (60%) after cross-adaptation to lo-' M L-cysteine. Discussion The behaviour experiments showed that sockeye were significantly attracted to chemical traces of their own population, but no such preference was evident for the sockeye. The homing migration of Sproat Lake sockeye, which begins during the 1 st week of June and ends by the middle of August, is generally 2 weeks earlier than the Great Central Lake population (Morley 198 1). For a period of at least 6 weeks, both populations mingle while migrating into Barkley Sound, Alberni Inlet, and the Somass River before ascending the Sproat and Stamp rivers (Fig. 1). By using freshwater parasites as - SMOLT WATER 0 - SMOLT WATER - SMOLT WATER -SMOLT WATER oll-i- - 1 L I 1 1, I Ol CONCENTRATION OF FlSH WATER (DILUTION) FIG. 5. Dose-response relationships of and adult sockeye salmon electro-olfactograms to varying concentrations of water conditioned with and smolt odours. Each point is the mean response of 3 to 5 fish. The response magnitude is normalized to 100% for lo-' M L-serine. biological tags, Margolis (1982) determined that these two populations accurately return to their own lake system. The straying rate is much less than 5% (Quinn 1985). This indicates that the Sproat Lake fish have no difficulty identifying home water when they reach the junction between Sproat and Stamp rivers (Fig. 1). The behaviour experiments were conducted towards the end of the migration period of the sockeye and this could have affected their behaviour owing to lower migratory motivation. Although the sockeye were less active in the behaviour tests than the fish, we consider this assumption unlikely, since the sockeye were still actively migrating when captured. Another possibility is that Sproat Lake sockeye do not use pheromones for directing their migration towards the home lake system, but rely on other riverine odours which they learned as juveniles (Hasler and Scholz 1983). The electrophysiological experiments revealed that the olfactory system of migrating adult sockeye salmon is sensitive to amino and bile acids. The demonstrated levels of sensitivity are generally comparable to those found in rainbow trout (Salmo gairdneri), Arctic chair (Salvelinus alpinus), and coho salmon (Oncorhynchus kisutch) fingerlings (Hara 1982; Hara et al. 1973; T. P. Quinn and T. J. Hara, unpublished data, Department of Fisheries and Oceans, Pacific Biological Station, Nanaimo, B.C.). Unlike coho salmon fingerlings and rainbow trout (Hara et al. 1984), however, the response magnitude of sockeye to taurocholic acid was generally lower than to amino acids. Contrary to the behavioural experiments, the results of electrophysiological experiments indicated that sockeye discriminated water conditioned by their own population from that by the other. It should be remembered that the discrimination properties of the olfactory system are a property of the system as a whole, and that the EOG response is a summated activity of receptor neurons (Persaud and Dodd 1982). Although the evidence does not demonstrate unequivocally that the two populations of homing sockeye salmon respond to population-specific odours, it does indicate that homing behaviour of different stocks may be influenced by these chemicals. Cross-adaptation experiments with L-cysteine and taurocholic acid suggested that the major active components in water

6 PERCENT FRASER SOCKEYE USING NORTHERN PASSAGE Oueen YEARS FIG. 6. Map showing the migratory routes of adult sockeye salmon on their way to the Fraser River and graph of the percentage of sockeye salmon approaching the Fraser River through the northern passage from 1953 to responsible for olfactory stimulation were primarily bile-acidlike substances, with minimal contributions by amino acids. In general, however, little progress has been made in identifying the chemical substances in home water. In a study on sockeye homing to Great Central Lake, Idler et al. (1961) reported positive olfactory responses to home water but not to other waters. The home water for these experiments was collected in the lake and would have contained both environmental odours and pheromones. In their attempt to isolate and identify the home stream odour substance, they determined that the active component is volatile, water soluble, neutral, dialyzable, and heat labile. Fagerlund et al. (1963) continued these studies and found that recombining the volatile and nonvolatile fractions of the home water induced a stronger response than did the volatile fraction alone. This indicated that a nonvolatile material also played a part in the olfactory response of sockeye. They concluded that home water apparently contains a "symphony" of odours to which sockeye salmon respond, and not a single substance. Bodznick (1978) found that sockeye fry discriminated between natural waters that differed only in calcium concentration and suggested that calcium is part of the odour bouquet that sockeye use to identify natural waters. Stabell et al. (1982) indicated that the active components of population-specific odours may be contained in skin mucus and intestinal fluids. However, Fiskness and Dgving (1982) failed to demonstrate population-specific responses to intestinal con- tents by Atlantic salmon. While salmonids are sensitive to amino acids and bile acids (Hara et al. 1984), no fish pheromones have yet been chemically identified (Liley 1982). Hara and MacDonald (1976) considered it unlikely that amino acids are involved in recognition of home-stream water but the present study suggests that bile acids may be major active components for recognition of population-specific odours. Bodznick (1975) found that the relative EEG responses of Adams River sockeye tested at three points during the last 550 krn of their migratory route home changed from location to location. Similar changes in olfactory responses during the course of migration were also reported by Cooper and Hasler (1973, 1974) for coho salmon (0. kisutch). Bodznick (1975) suggested that these changes might be related to changes in GROOT ET AL sexual maturity and endocrine status of the fish. Dgving et al. (1974) reported that odours from different populations of char elicited differential EEG responses. Quinn and Tolson (1986) have recently found that juvenile coho salmon from two hatcheries on Vancouver Island showed different choice responses to stock-specific odours in a Y-maze. These varied 80 reactions of different salmonids may be due to genetic back-,, ground or different cultural conditions giving the fish a different sensory experience. Thus, the olfactory responses of migrating salmon can be affected by physiological and endocrinological conditions, 40 a which change over time, and by stock-specific differences. 30 These factors could have played a role in the behavioural,, responses of the Great Central and Sproat Lake sockeye, causing one population to respond to pheromones and the other not. Results to date indicate that coho and sockeye salmon can potentially recognize stock-specific odours, but is there evidence that they actually use pheromones during migration? The 2-20 million sockeye salmon returning yearly from the ocean feeding grounds to the Fraser River in British Columbia, Canada, offer a natural experiment, since they face a giant Y-maze situation when they approach Vancouver Island. They can either take a northern route through Queen Charlotte Sound and Johnstone Strait or travel south through the Strait of Juan de Fuca to reach their home river (Fig. 6) (Royal and Tully 1961 ; International Pacific Salmon Fisheries Commission (IPSFC) ; Groot et al. 1984). Studies on the migration patterns of Fraser River juvenile sockeye salmon in the Strait of Georgia have shown that the majority travel north through Johnstone Strait towards the ocean (Groot et al. 1984; and unpublished data, Department of Fisheries and Oceans, Pacific Biological Station, Nanaimo, B.C.). The pheromone hypothesis would predict that most adults return via the northern passage, retracing the path of the smolts that left the Strait of Georgia through the northern route prior to the adults' arrival (Groot et al. 1984). The ratio of adult sockeye migrating north and south around Vancouver Island has varied greatly over the 32 years ( ) for which information is available (average diversion rate via the north 32%, range 5-85%) (Fig. 6) (IPSFC ). In 1983, 85% of adult sockeye salmon returning to the Fraser River entered the Strait of Georgia via the northern route (IPSFC 1984). Since during this same year we (C. Groot and K. Cooke, unpublished data, Department of Fisheries and Oceans, Pacific Biological Station, Nanaimo, B. C. ) established that all smolts also migrated out northward toward the ocean, the findings are consistent with the pheromone hypothesis. However, in 1982 and 1984 the northern diversion rates for Fraser River adult sockeye salmon were, respectively, 25% and 31% (IPSFC 1983, 1985). These low diversion rates were inconsistent with the pheromone hypothesis because surveys in the Strait of Georgia indicated that most, if not all sockeye smolts migrated northward from the Fraser River in those years (Groot et al. 1984, and unpublished data, Department of Fisheries and Oceans, Pacific Biological Station, Nanaimo, B.C.). We therefore conclude that the interannual variability of the Fraser River adult sockeye migrations around Vancouver Island does not support the hypothesized role of pheromones in directing the migrations of salmon in coastal waters. We assume with Tully et al. (1960) and Royal and Tully (1961) that oceanographic conditions in coastal areas determine the migratory routes of the sockeye salmon.

7 CAN. I. ZOOL. VOL. 64, 1986 Acknowledgments - Arnold, J. J. Dodson, and W. H. Neill. Plenum Press, New York. We gratefully acknowledge the assistance of Ken Cooke with PP HAER, A. D Underwater guideposts. University of Wisconsin both the behaviour and electrophysiological experiments and of Press, Madison. Ken Lurid, Craig and Graeme for their HAER, A. D., and A. T. SCHOLZ Olfactory imprinting and with testing and transporting the salmon. We thank the homing in salmon. springer-verlag, Berlin. Directors of the Pacific Biological Station at Nanaimo and the HORRALL, R. M, Behavioral stock-isolating mechanisms in Freshwater Institute at Winnipeg, Drs. R. Beamish and B. Great Lakes fishes with special reference to homing and site Eales, respectively, for making this cooperative study possible. imprinting. Can. J. Fish. Aquat. Sci. 38: Sherry Greenham expertly prepared the illustrations. During the IDLER, D. R., J. R. MCBRIDE, R. E. E. JONAS, and N. TOMLINSON. later stages of this project, T. P. Quinn was supported by Olfactory perception in migrating salmon. 11. Studies on a NSERC strategic grant G laboratory bio-assay for homestream water and mammalian repellant. Can. J. Biochem. Physiol. 39: BODZNICK, D The relationship of the olfactory EEG evoked by naturally-occurring stream waters to the homing behavior of sockeye salmon (Oncorhynchus nerka). Comp. Biochem. Physiol. A, 52: Calcium ion: an odorant for natural water discriminations and the migratory behavior of sockeye salmon. J. Comp. Physiol. 127: BRANNON, E. L., R. P. WHITMAN, and T. P. QUINN Responses of returning adult coho salmon to home water and population-specific odors. Trans. Am. Fish. Soc. 113: COOPER, J. C. and A. D. HAER An electrophysiological approach to salmon homing. Fish. Res. Board Can. Tech. Rep. No Electroencephalographic evidence for retention of olfactory cues in homing coho salmon. Science (Washington, D.C.), 183: D~VING, K. B., H. NORDENG, and B. OAKLEY Single unit discrimination of fish odours released by char (Salmo alpinus L.) populations. Comp. Biochem. Physiol. A, 47: EVANS, R. E., and T. J. HARA The characteristics of the electro-olfactogram (EOG): its loss and recovery following olfactory nerve section in rainbow trout (Salmo gairdneri). Brain Res. 330: FAGERLUND, U. H. M., J. R. MCBRIDE, M. SMITH, and N. TOMLINSON Olfactory perception in migrating salmon Stimulants for adult sockeye salmon (Oncorhynchus nerka) in home stream waters. J. Fish. Res. Board Can. 20: FERGUSON, G. A Statistical analysis in psychology and education. McGraw-Hill, New York. FISKNES, B., and K. B. D~VING Olfactory sensitivity to groupspecific substances in Atlantic salmon (Salmo solar L.). J. Chem. Ecol. 8: GROOT, C., L. MARGOLIS, and R. BAILEY Does the route of seaward migration of Fraser River sockeye salmon (Oncorhynchus nerka) smolts determine the route of return migration of the adults? In Mechanisms of migration in fishes. Edited by J. D. McCleave, G. P. Arnold, J. J. Dodson, and W. H. Neill. Plenum Press, New York. pp HARA, T. J. (Editor) Chemorecepion in fishes. Elsevier, Amsterdam. HARA, T. J., Y. M. C. LAW, and B. R. HOBDEN Comparison of the olfactory response to amino acids in rainbow trout, brook trout, and whitefish. Comp. Biochem. Physiol. A, 45: HARA T. J., and S. MACDONALD Olfactory responses to skin mucus substances in rainbow trout, Salmon gairdneri. Comp. Biochem. Physiol. 54: HARA, T. J., S. MACDONALD, R. E. EVANS, T. MARUI, and S. ARAI Morpholine, bile acids and skin mucus as possible chemical cues in salmonid homing: Electrophysiological re-evaluation. In Mechanisms of migration in fishes. Edited by J. D. McCleave, G. P. INTERNATIONAL PACIFIC SALMON FISHERIES COMMISSION (IPSFC) Annual reports for IPSFC, New Westminster, B.C. JOHNSEN, P. B., and A. D. HAER The use of chemical cues in the upstream migration of coho salmon, Oncorhynchus kisutch Walbaum. J. Fish Biol. 17: LILEY, N. R Chemical communication in fish. Can. J. Fish. Aquat. Sci. 39: MARGOLIS, L Parasitology of Pacific salmon-an overview. In Aspects of parasitology-a Festschrift dedicated to the Fiftieth Anniversary of the Institute of Parasitology of McGill University, Edited by E. Mecrovitch. McGill University, Montreal. pp MORLEY, B Adult sockeye escapements to Sproat Lake and Great Central Lake (Somass River System, Port Alberni, B.C.) in Can. Manuscr. Rep. Fish. Aquat. Sci. No NORDENG, H Is the local orientation of anadromous fishes determined by pheromones? Nature (London), 233: NORDENG, H A pheromone hypothesis for homeward migration in anadromous salmonids. Oikos, 28: PERSAUD, K., and G. DODD Analysis of discrimination mechanisms in the mammalian olfactory system using a model nose. Nature (London), 299: QUINN, T. P Homing and the evolution of sockeye salmon, Oncorhynchus nerka. In Migrations: mechanisms and adaptive significance. Edited by M. A. Rankin. Contrib. Mar. Sci. 27, Suppl. QUINN, T. P., E. L. BRANNON, and R. P. WHITMAN Pheromones and the water source preferences of adult coho salmon Oncorhynchus kisutch Walbaum. J. Fish. Biol. 22: QUINN, T. P., and C. A. BUSACK Chemosensory recognition of siblings in juvenile coho salmon (Oncorhynchus kisutch). Anim. Behav. 33: QUINN, T. P., and G. M. TOLSON Evidence of chemically mediated population recognition in coho salmon (Oncorhynchus kisutch). Can. J. Zool. 64: ROYAL, L. A., and J. P. TULLY Relationship of variable oceanographic factors to migration and survival of Fraser River salmon. Calif. Coop. Oceanic Fish. Invest. Rep. No. 8. pp SELSET, R., and K. B. D~VING Behavior of mature anadromous char (Salmo alphinus L.) towards odorants produced by smolts of their own population. Acta Physiol. Scand. 108: STABELL, 0. B., R. SELSET, and K. ETTEN A comparative chemical study on population-specific odorants from Atlantic salmon. J. Chem. Ecol. 8: STABELL, 0. B Homing and olfaction in salmonids: a critical review with special reference to the Atlantic salmon. Biol. Rev. 59: TULLY, J. P., A. J. DODIMEAD, and S. TABATA An anomalous increase of temperature in the ocean off the Pacific coast of Canada through 1957 and J. Fish. Res. Board Can. 17: