July 08, 2022
This article is a bit of a reboot of an article that originally appeared in Fly Fisherman in 1990 on fish vision entitled “How Trout See” by Gordon Byrnes. The science has progressed, so Don Stazicker updated the original. You can read the original "How Trout See" here.
You cast accurately with just enough slack to defeat drag, and your fly drifts perfectly over the big trout. The fish’s pectoral fins flick out, a sure sign that it’s seen your fly, and the trout rises up toward your imitation. Success appears certain until, 3 inches short of your fly, the fish stops and drifts a short distance downstream before sinking back to its lie. A refusal.
What went wrong? Your fly and drift were obviously good enough to cause the fish to rise from its lie and approach to within a short distance of the fly. But once it got close, something caused it to reject your fly. It could have been the fly’s behavior, slight micro-drag, or something about the tippet might have put the fish off. These factors may be invisible to you, but they are all visible to the fish at close range. Apart from making effective slack-line casts with long, fine tippets, there is little more we can do about these problems. However, there are things we can potentially do about our fly patterns.
Trout Use Search Image
Trout feeding selectively use a search image, which becomes reinforced with repeated successful encounters. If our artificial fly matches the search image, it is taken. If not, it is refused.
When you get a refusal, the fly brings the fish up for a close inspection and then, when the fly is a few inches away, it does not match the search image and is rejected. Something about the appearance of the fly may have prompted the refusal. We can easily spot our Parachute Adams among the natural duns we are trying to imitate when it is over 30 feet away. We assume that the trout, a predominantly visual hunter, will be at least as good as we are at telling the difference between real and fake flies. But we are wrong to assume this.
Color Vision of Trout
Scientific testing has shown that trout have color vision, and has even measured how faint a color can be and still trigger a response from the fish.
It is often said that we can’t know what a fish sees, but that’s also true of other humans. When I view the Stars and Stripes, I see what I call red, white, and blue. I have no way of knowing if your brain interprets the colors you call red, white, and blue in the same way as mine. But it doesn’t matter as long as both of us can consistently recognize those colors.
We cannot know how the trout’s brain interprets the signals sent from its eyes via the optic nerves to the brain, but by studying the structure of the eye and experimentally testing the fish’s responses to a variety of optical stimuli, we can get a good idea of what the trout’s eyes see, and how they respond to it. When a trout is feeding selectively to size 16 olive mayfly duns, and my size 16 olive Compara-dun fools it into taking, that’s as much confirmation as I need. That the fish’s brain interpreted both flies as purple doesn’t matter as long as the trout consistently sees them that way.
But what if trout can see a wider range of light wavelengths than we can—what about ultraviolet (UV)? Recently, fly tiers have used materials treated with dyes to reflect UV light. Do they more accurately imitate the trout’s prey? Humans don’t have UV-sensitive cones in our eyes, therefore we cannot tell if either the natural we are trying to imitate or our fly-tying materials are reflecting it. As a result, we cannot know if we are using the materials correctly. However, it’s unlikely to affect our catches because the balance of scientific opinion is that adult trout do not use UV reflected light to detect prey, as they lose the UV-sensitive cones they had when they were small fry.
Visual Acuity of Trout
Remarkably, a trout’s eyes can focus from a few inches to infinity in a narrow arc of vision in front of and slightly above the fish, while focusing at infinity in all other directions allowed by its 180-degree field of vision. Thus, the fish can employ its binocular vision to focus on prey approaching from the front, while retaining distance focus in the backward, downward, upward, and lateral directions. This enables the trout to detect and focus on its prey and simultaneously spot approaching predators.
Trout have better low-light vision than humans, and brown and rainbow trout can respond to a light level half of that required by brook and cutthroat trout.
Trout must detect, recognize, and intercept prey drifting with the flow before the current carries it beyond reach. As a result trout are particularly good at detecting movement and contrast.
The intensity of light passing through water is attenuated rapidly—more so if the water is turbid due to suspended matter. The presence of inedible particles makes the detection of prey more difficult. It’s like driving into a snowstorm at night and reaching out through the car window to grab a passing butterfly.
We see the results onstream. Trout are less demanding of exact imitation in fast, turbulent currents, but may be maddeningly choosy in slow, clear water.
Where the water is turbid or contains inedible particles, our flies must stand out from both the background and from the nonedible items. Moving the fly might achieve this, or fly tiers might use hot spots or flash to make our flies catch the fish’s attention. You may think that increasing the fly size would make it stand out, but this rarely works. A fly that is too large will no longer conform to the trout’s prey image and will not trigger a strike.
So far so good, but there’s a weak spot in the trout’s visual performance. A rainbow trout’s visual acuity is 1/14 that of the human eye. Other salmonid species are likely to have similar visual acuity.
The American Optometry Association defines visual acuity as the clarity or sharpness of vision.
What appears to be in sharp focus to a human will appear blurred to the trout.
This does not mean humans can see things that the trout cannot. It only means that we can see them in more detail. Watching a trout feeding on size 30 aphids or intercepting an ascending size 24 midge pupa, you realize they can detect tiny items, especially if they are moving characteristically.
In 1990, Fly Fisherman published a seminal article on fish vision entitled “How Trout See” by Gordon Byrnes, then a Navy physician specializing in optometry. He comprehensively explained many aspects of the trout’s vision in terms that non-optometrists could understand.
Angling authors frequently cite this article, particularly mentioning the section about visual acuity because Gordon had included pictures of flies as trout would see them at distances of 3, 6, and 12 inches. He had “altered the focus a calculated amount” to simulate the detail available to the trout at these distances.
The Byrnes research shows that at 3 inches, where the trout’s visual performance peaks, the detail available to the fish is significantly worse than we would see if the fly was at our best close viewing distance. This is not because the fly is too close for the fish to focus on it. It’s just that it doesn’t have the density of photoreceptor cells in its retina to produce a more detailed image. Viewed at 12 inches, the fly is a blur. Fine detail is gone, but size and shape are still visible.
Is this the reason why a trout will rise to inspect our fly and then refuse when it has viewed the fly at close range? It is probably one reason why our obviously artificial flies fool trout at all. At a distance, the primary characteristics, overall shape and size of the fly, possibly its color, and its behavior signal to the fish that here is something that may well be edible. Once the range is closer, and the fly is viewed at 3 inches, the fish can determine whether to take or refuse the fly. This decision depends on secondary characteristics, the finer details of the fly, including a better appreciation of its color, which is more easily assessed when the fish is close to the fly, and the fly occupies a greater proportion of the visual field.
Hold your thumb at arm’s length against a bright sky. It is a virtually black silhouette. Now bring your thumb closer to your eye, and its color is much easier to determine. Once the fish get close, the fly is not a silhouette and color becomes a factor. This is another cause for refusals at close range.
If fine detail matters when the fish is close to the fly (and it must, otherwise last-minute rejections would not occur) then surely we should put as much detail into our flies as possible.
Well, yes and no. The fish cannot see the same detail that we can, so it might be convinced by what to us would appear a crude, poorly detailed image. If we leave out crucial details, whatever they may be, then we are more likely to get a refusal.
It’s also true that the presence of certain details can cause refusals. I have often had trout swim 5 feet to inspect my gold beadhead scud, only to stop dead at around 3 inches and refuse to take the fly. The gold bead was the negative aspect of the fly. I determined this because I changed to a pattern identical in every respect except it had a black bead, and caught the same fish. This took place on a river that receives a lot of pressure, and where many anglers use gold beads. The first fly looked good enough when the fish saw it at 5 feet. It came over to investigate further, but once it could see the fly more clearly, it didn’t like it.
Knowledge of the trout’s lower visual acuity doesn’t seem to have stopped us tying flies with lots of potentially ineffective details. Maybe people doubted the pictures.
Thirty-two years after the Byrnes article, our modern computing power, and measurements of trout eyes and its modulation transfer function—an optical measurement used to test lenses of all types—can reveal how much detail a trout can see at varying distances.
Recently, Phil Connor, who lives near Lake Taupo on New Zealand’s South Island, featured an article on his website, ontheflynz.com, where he used a program called AcuityView, developed by Dr. Eleanor Caves and Sönke Johnsen, to process pictures of artificial nymphs. These modern images appeared similar to the photos in Byrnes’s original article. Connor kindly processed some of my own images to show what a trout would see with flies at 3, 6, and 12 inches (these images are shown below). While there are some differences, Gordon Byrnes was pretty close with his “calculated blurring” 32 years ago.
These pictures can give some guidance on what to include in our artificial flies to make them maximally effective.
For instance, translucent backlit structures such as wings and shucks appear in full color. Contrasting segmentation and ribbing are clearly visible at 3 and 6 inches, but disappear at 12 inches. Dubbing is visible as individual colored fibers only at 3 inches, and even then, not clearly. Color vibrancy decreases with distance and increased blurring. The so-called “light pattern in the surface film” of the Upside Down Dun is very blurred at 12 inches and is unlikely to initiate a rise. Contrasting stripes in the wet fly wing, thought to attract sensitive trout, are invisible at 12 inches and barely visible at 6 inches.
What does all this mean for fly design? The primary characteristics of shape, size, color, and behavior are essential in getting the fish to come close enough to our flies to resolve fine details. Fine detail is wasted if the fish can’t see it.
Include as much detail as you wish in your flies, while remembering that hyper-realistic flies, which humans find hard to distinguish from real insects, don’t excel at catching trout. Including complex, exact detail is time-consuming, expensive, and often results in a rigid, lifeless, ineffective fly.
Many simple patterns work well, and an easy way to find out how simple a fly can be and still be effective is to take some of your successful patterns and remove features from them until they cease to work effectively. The results will surprise you; they’ll certainly make your flies quicker to tie. When you look at the elegant simplicity of the original Sparkle Dun wing, body, and shuck, just three materials used to tie a devastatingly effective fly, it shows just how overly complex some flies have become.
A well-tied fly with plenty of detail can be a thing of beauty. Just don’t fool yourself that complex detail is necessary. Some of my flies include features the trout probably don’t notice and that do little to increase the fly’s effectiveness, but they do look good.
Don Stazicker is a British writer and guide, coauthor of the e-book Trout and Flies - Getting Closer, which examines the trout’s interaction with natural and artificial flies and how to use that knowledge to fly fish more successfully.
H. Rahmann, G. Jeserich, and I. Zeutzius (1979).
Ontogeny of Visual Acuity of Rainbow Trout under Normal Conditions and Light Deprivation.
Behaviour, 68(3/4), 315-322.
E.M. Caves and S. Johnsen (2018).
AcuityView: An R package for portraying the effects of visual acuity on scenes observed by an animal.
Methods in Ecology and Evolution (9), 793-797.
G. Mueller (2013).
What a Trout Sees: A Fly-Fishing Guide to Life Underwater.
Guilford, CT: Lyons Press.