Home Again, Home Again

Chapter III - 4 

Home Again, Home Again 

I was talking to an angler one day and it didn’t take long before we were talking about salmon and the homing ability of salmon. Here is what we talked about.

Salmon are certainly famous for their long migrations and ability to return to their home stream. And it is widely understood that they use odor as a cue. The rest of that equation, however, is often overlooked. If salmon use odor as a cue to find their way home, they must first somehow learn that cue. That is called imprinting and it is not genetic. Salmon must first imprint on the odor or odors of their home stream and store that memory so they can use it as a cue to successfully return one to several years later.

How does that work? The salmon imprint on the odors from the complex mix of organic matter that is unique for their home stream. This must happen while they are juveniles and before they enter the ocean. They retain that memory until they need it to find their way back to their home stream. In the simplest interpretation, one would assume that the salmon merely apply that memory as they return and the home stream odor simply gets stronger and stronger as they get closer to their target.

Another version, which I believe is much more plausible and more widely accepted, goes like this. Young salmon, perhaps as young as alevin in the gravel, but certainly during the smolt stage, create a cumulative, sequential memory of odors they experience beginning in the home stream—and maybe even the home riffle—and continuing into the estuary. This has been likened to a “bouquet of odors.” Start with a bouquet of flowers which represents the home stream and take in the fragrance.  Now, as each flower dies, replace the dead flower with some different kind of flower. The odor of the bouquet changes sequentially as the flowers are replaced, just as the odor of the stream changes with the addition of each new tributary while the fish migrates downstream. As each tributary stream enters the river, the odor of their home stream becomes more and more dilute. Later, the returning fish uses the reverse sequence of the imprinted odors as a map to find their way to the spawning riffle of the home stream.

Think about taking a walk or ride through unfamiliar territory—a city, a forest, or a building. We store a memory of what we see, or even smell, perhaps from a bakery, coffee machine, or flowers. Later, we use those memories as cues to find our way back to our point of origin. It is well known that animals such as dogs and deer use odor as a cue. And there is evidence that other fishes and at least one salamander use the sense of odor to home as well.

Many scientists believe that salmon navigate across the ocean waters by using the sun for orientation like a compass or; they may be able to sense the earth’s geomagnetic field and use it as a navigational aid while they are in the ocean. Regardless of how they accomplish it, they need some navigational tool or tools to return to near shore until they can detect the first landmark on their map or bouquet of odors.

Why do salmon imprint and home anyhow? Why do they migrate to the ocean in the first place? Why don’t they just stay in freshwater? Because feeding and growth in the North Pacific is much better than in freshwaters. Salmon evolved to take advantage of both worlds. And thanks to their ability to return to their stream of origin, it is assured that they are returning to a place which has a history of successful spawning and rearing (the very stream where they were spawned) and not on a random search. Otherwise, how could they know if continuing upstream would be worthwhile?

Salmon are well known for their ability to return and find their home stream, perhaps even the same riffle from which they were hatched. But there is more. What is the opposite of homing? Getting lost or straying. Straying is homing’s little brother that is often ignored or overlooked. But it is critical to the survival of the population and the species. If salmon were perfectly successful in their homing, how would newly available habitat become colonized? Habitat exposed by a retreating glacier, for example. And if some well-used habitat is degraded, by a landslide, for example, the population would be lost unless some of the returning fish stray.

Straying is very hard to measure. In the first place, it is just hard to define because, although a fish may be caught in a non-home stream, is that because the fish is committed to spawn in that stream or did it simply make a mistake? Salmon often enter a non-home stream, detect that they have made a mistake, return downstream, and continue their migration. In addition, straying is a rare event, therefore, strayed fish are very hard to detect. Usually, straying is defined as a fish which actually spawns in a non-home stream. Many fish must be examined in a number of streams. Nevertheless, some studies suggest that the straying rate among salmon is about 1 to 5 per cent of a spawning population.

It is easy to watch schools of salmon swimming up a stream and be in awe of their homing ability but it is too easy to forget just how much this means to the fish and how challenging that migration really is.

Now I have a question for you to ponder. Just how do you suppose scientists figure this stuff out? If it was your task, how would you do it?

What do you think? Where would you start? What are the questions? How do we get answers?

Scientists approach a challenge like this by creating a hypothesis or informed guess and designing an experiment to accept or reject that hypothesis. Some experiments are performed in a laboratory setting and some are done in the field. Each is intended to answer one question at a time to eventually create a big mosaic step by step.

First, how do we know salmon home? Two ways. Install a weir on a salmon stream and collect some returning adult fish. Mark them with some sort of a tag, take them back downstream, and release them to see if they return to the weir. Even better, catch some smolts in the vicinity of a spawning area, give each an identifiable tag and release them. If you are studying coho salmon, install an adult salmon weir one year later and look for marked fish (sockeye and Chinook salmon return over a span of several years). This is verification that fish that originated from this stream emigrated as smolts and homed as adults.

Okay, and how did the scientists learn that odor is the cue? This is where it gets tedious. As early as 1653, Izaac Walton showed that Atlantic salmon returned to a home stream.  But it took until sometime in the early 1950s, when a scientist hypothesized that odor was the cue and he proposed that a laboratory experiment could test the hypothesis that different drainages have different odors. He followed a widely-used technique to train fish to respond to a particular stimulus and then tested them with a challenge. In this case, he placed a fish in a flow-through tank with two inlet pipes. Each pipe fed water from a different stream. When the fish approached pipe A, it was rewarded with a morsel of food. After the fish had been trained, he switched stream water A to pipe B to verify that the fish had learned to associate the reward with the stream odor and not the location of the pipe or some other cue, like a light or a color.

So fish have the capability to detect odors and different streams have different odors. But how does it work in a real stream? Well, that scientist and his hard-working graduate students designed more simple but elegant experiments to test adult salmon homing abilities. They found a small river with two tributaries of about the same size that formed a Y-shape and set up a weir on each branch. They captured and marked adult salmon which were sorted randomly into two groups. One group had their nasal passage plugged so they could not detect odors and the other group was left unmolested. All the fish were transported back downstream and released. You can guess what is coming. The fish that could sense odors mostly returned successfully to their stream of origin while those with plugged noses showed up in a random fashion. This demonstrated the importance of the sense of smell.

In a similar study, of two groups of marked fish, one group was blinded and all were returned downstream. Blinded and sighted fish homed at similar success rates. This demonstrated that the salmon did not rely on sight as a cue.

They didn’t stop there, but went back to the laboratory and performed an experiment similar to the one with the two stream waters, but they used only one water source. Water in one inlet was treated with a synthetic organic chemical called morpholine. The fish were able to learn to recognize that chemical. This showed that the cue was from an organic source.

Next, the scientists went to a salmon hatchery and dripped morpholine into a raceway filled with coho salmon smolts. After they were imprinted, the smolts were released in open water so they would not have the opportunity to imprint on any particular stream water. A year later, morpholine was dripped in a stream when adults returned. The returning salmon went up the stream that was treated with morpholine. When the morpholine drip was switched to a different stream, the salmon went up that stream.

You have probably noticed that when a tributary stream enters a bigger body of water, the inlet stream water hugs the bank for some time before it becomes fully mixed. Okay, this time, as the scientists dripped morpholine into an inlet stream, they monitored the extent of the mixing zone and they released an adult salmon that had been imprinted with morpholine and implanted with a radio transmitter. As they tracked the fish they discovered that it moved in and out of the interface of the mixing zone in a zigzag, searching pattern much like a dog tracks a scent across a field.

Finally, another graduate student did a bit of fascinating research. He anesthetized a salmon which had been imprinted with a particular home stream water, exposed the brain, and hooked very fine electrodes to the portion of the brain that processes odor stimuli. When he flushed the nasal passage with home stream water, that part of the brain showed elevated electrical activity.

Not all of this research was done by one scientist but most of the experiments have been replicated by other scientists and with other species of fish. And of course, not all hypotheses proved correct, but over time, all the pieces came together. Most of these studies spanned roughly a 35-year period. And it does not end there. Research hardly ever just stops. There are more and more clues and evidence that imprinting does not happen just during the smolt stage, but some imprinting may begin while the alevins are still in their redds.

There are many more details to this story, of course, but you can see the big picture and certainly appreciate the challenges and rewards of unraveling the secrets of how fish are able to do marvelous things.

References: Hasler and Scholz, 1983; Quinn, 2005.