This report reviews the evidence for olfactory imprinting in salmon. The results of research performed by various investigators are remarkably consistant. The experiments conducted by these reseachers were done in the field instead of the laboratory and thereby provide direct evidence that salmon use olfactory cues for homing.
The olfactory hypothesis for salmon homing, first presented by Hasler and Wisby in 1951, is based on three beliefs: (1) because of local differences in soil and vegatation of the drainage basin, each stream has a unique chemical composiion and, thus, a distinctive odor; (2) before juvenile salmon migrate to the sea they become imprinted to the distinctive odor of their home stream; (3) adult salmon use this information as a cue for homing when they migrate through the home-streaming network to the home tributary.
There have been two proposed hypothesis to how salmon locate their natal stream. First in 1951, Hasler and Wisby proposed the "imprinting hypothesis." This hypothesis states that recognition of the home stream results from a relatively rapid odor learning process during a sensitive period called the smoltification process. The other hypothesis was proposed by Nordeng in 1971. He spectulated that smolts that migrate to the ocean, release population specific odors called pheromones. It's these pheromones, in the end, that are used by mature adults as cues to guide them back to their home stream. Evidence supporting both the imprinting and phermone hypothesis has accumlated over the last two decades (Hasler and Smoltz 1983; Smith 1985).
A. Eggs are laid each fall and hatch about 1 month after depositon in freshwater streams. Incubation of the eggs and much of the early developent occurs under the gravel. B. Hatchlings (alevins) reside in the gravel, absorbing nutrients from their yolk sac. C. Parr continue to live and grow in freshwater streams until the following spring, when they undergo a behavioral and physiological metamorphosis called smoltification that enables them to live in salt water. D. Smolts migrate downstream to begin life in the open ocean as sea-run salmon. E. Two to three years later, they become mature spawning individuals. F. Complete extraordinary migration, often over thousands of miles, to return to the exact tributary where they were hatched, to spawn and complete their life cycle.
When they are 1.5 years old, in April or May, coho salmon undergo a metamorphosis referred to as smolt transformation or smoltification. Morphological, physiological, and behavioral changes that preadapt young salmon to life in sea water occur during this period. During the smolt stage, their parr marks disappear, and they turn silver, which is a color adaptation for living in the ocean. Their osmoregulatory mechanisms begin adjustments that will enable survival in salt water. They cease territorial behavior and form schools numbering in the thousands, and embark on their seaward journey. Just prior to the smolt stage, the endocrine system undergoes major transitions. Smolt transformation is thought to be induced by thyroid hormones. The smolt stage is of critical importance to understanding the homing process because it is during this period that salmon "imprint" to some property of their natal tributary that serves later to identify it when they return as adults to spawn [Hasler and Scholz, 1983].
Adult salmon captured at Oak Creek, were transferred by boat to a point 3.2 km north of Oak Creek. Before releasing them, the researchers inserted an ultrasonic transmitter down the esophagus into their stomach to follow their movement.
Scholz and his collegues selected a release site assuming that fish released near the shore would follow the shoreline and intercept an intervening point where Oak Creek flowed into Lake Michigan. Morpholine was introduced into test area perpendicular to shore, extending from the mouth of Oak Creek to about 100 m offshore. This created a narrow band which acted as an "odor barrier" through which the fish had to swim. The morpholine concentration was kept at 5.7x 10-10 M. The water currents were measured with the use of various instruments to determine how long the chemical remained there. Before tracking, the fish were held for 3 to 17 days in tanks containing Lake Michigan water to reduce the possibility that they might react to morpholine only because they had been recently exposed to the odor.
Scholz and his collegues tracked 56 fish through this test area. Most of the salmon remained near the release point for about one hour before moving. In all cases (20 tracks) when morpholine was present in the test area, morpholine-imprinted fish stopped their migration and remained in scented area from 1 to 4 hours. The salmon's stay correlated roughly with the time it took for water currents to dissipate the chemical. When morpholine was not present in the test area, morpholine-imprinted fish moved through without stopping. On the other hand, 13 control (not imprinted) fish did not stop when morpholine was present. [Hasler and Scholz, 1983] See the results of this Lake Michigan Experiment. Illustration from Hasler and Scholz, 1983
In salmon, sex hormones play a role in modifying olfactory sensitivity and discrimination capability. The effect of sex hormones on olfactory sensitivity is interesting in light of the fact that olfactory sensitivity of salmon to their imprinted odor increases over the course of the spawning season during which there is increased circulating levels of sex hormones. See EEG response to Morpholine:(illustation from Hasler and Scholz, 1983)
Exposure to micromolar concentrations of PEA for as little as 10 days during a sensitive period for olfactory imprinting (parr-smolt transformation) correlated with a specific increase in the responsiveness of olfactory receptor cells to PEA. when tested 6-9 months later (Dittman, 1994). See behaviral responses to PEA: (illustration from Dittmann, 1994)
All these experiment show without a shadow of a doubt that the supernatural sense of smell of salmon is crucial in the maintance of their life cycle. It also demonstrates the great importance of the smoltification process in imprinting the natal steam's unique odor. The rate of return in these experiments was observed to be about 0.5%-2% (ocean runs) or 2%-5% (freshwater runs), with about 95% of the recovered fish captured in the stream of release, which compares favorably with homing in natural populations [Hasler and Scholz].
In contrast, transplanting fish after smolt transformation, resulted in poor recovery (.025%). These findings suggest that imprinting terminates soon after smolt transformation begins, thereby preventing fish from becoming imprinted to other tributaries during the coarse of their downstream migration.
Hasler and Scholz summarize that two important conclusions can be drawn from these transplantations studies. First, the memory of the home stream is not inherited. Second, homing is connected with a period of rapid and irreversible learning, i.e..,imprinting, of the cues that identify the home stream at the time the young salmon begin their downstream migration. After three or fours at sea, the salmon recall what they learned as smolts from their long term memory.
In addition, it was recently discovered Pacific and Altantic salmon use photoperiodic cues to initiate spawning migration by activating production of a sex hormone called gonadotropin. Plasma levels of gonodotropin increase in fish exposed to gradually decreasing daylengths. Gonadotropin stimulates differentation of gonads, induces production of sex steroid hormones, and induces migration.
Investigations into the imprinting have been applied to management applications. For example, information about the timing of imprinting or external indicators of imprinting are important for determining the time for stocking fish to insure homing to the stream of release.
"Surely any step toward saving salmon is worthwhile, for they are a gift from nature to man. We neither feed them nor take care of them in any other way. Yet, as young fish, they go out to sea where they grow fat on the ocean's rangeland and then deliver themselves, free of charge, to our back door for the catching and eating." [Hasler and Scholz, 1983].
Dodson, Julian J., et al. "Facilitative Effect of Preexposure on Heart-Rate Conditioning to an Olfactory Cue in Altantic Salmon (Salmo Salar)." Journal of Cmparative Psychology. 104 (1990) : 340-344.
Hara, Toshiaki J., et al. "Sibling Recognition and Olfactory Sensitivity in Juvenile Coho Salmon (Oncorhynchus kisutch)." Canadian Journal of Zoology 64 (1986) : 921-924.
Hasler, A.D. and A.T. Scholz. Olfactory Imprinting and Homing in Salmon. New York: Springer-Verlag Berlin Heidelberg, 1988.
Hassler, Thomas J., et al. "Homing of Chinook Salmon Exposed to Morpholine" California Fish and Game 76 (1990) : 31-35.
Maxwell, Jessica. "Swimming with Salmon." Natural History. September 1995: 29-39.
Morin, Pierre-Philippe., et al. "Changes in the Olfactory Function of Atlantic Salmon, Salmo salar, in the course of smoltification." Canadanian Journal of Fisheries and Aquatic Sciences 49 (1992) : 1704-1713.
Nevitt, Gabrielle A. "An Electrophysiological characterization of ciliated olfactory receptor cells of the coho salmon Onchohynchus Kisutch." Journal of Experimental Biology 166 (1992) : 1-17.
Quinn, Thomas P., et al. "Pacific Salmon Migrations and Homing: Mechanisms and Adaptive Significance." Tree 5 (1990) : 174-177.
Rehnberg, Bradley G., et al. "Olfactory Sensitivity during Parr and Smolt Developmental Stages of Coho Salmon." Tranactions of the American Fisheries Society. 114 (1985) : 732-736.
Rehnberg, B.G., et al. "Homing of Coho Salmon (Oncorhynchus Kisutch) Exposed to Morpholine." Aquaculture 44 (1985) : 253-255.
Scholz, Allan T., et al. "Imprinting to Chemical Cues: The Basis for Home Stream Selection in Salmon." Science 192 (1976) : 1247-1249.
