Fish Vision - What do Salmonids See?
by Dave Wallbridge
Some time ago
ago I posted some comments on fish vision in the `Nymph
design` topic on an internet message board. As the
information given then was of a generalised nature and
knowing that fish vision is extremely species specific -
i.e. different species have widely different eyesight - I
decided to do some searching to see if I could find any up
to date information which related specifically to Trout, Sea
Trout and Salmon.
Here is a very brief summary of what I found,
most of which was kindly supplied either directly by the
researchers themselves or from publications which they
pointed me to.
It seems that
compared to ourselves, trout and salmon have pretty poor
vision - in fact our vision is about 14 times better at
resolving images than theirs is. This is more than a little
reassuring to me - it's good to know that even my ageing
vision must still be at least 12 times better than theirs!
From what I have gathered it appears that a trout's eye can
detect relative size, overall shape and general colour
pattern but, even at its sharpest focus which is about 2-3
inches from the fish's mouth, viewed objects will appear
As you might
expect of a surface feeding fish, the photoreceptors are
packed more densely in the lower part of the retina giving
maximum resolution to upward vision. Interestingly, because
of the unusual shape of the lens, they are able to focus
both distant and near objects simultaneously - like looking
through both parts of a pair of bi-focals at the same time.
There are two areas
where the trout's eye excels. The first is contrast -
anything which stands out against the underwater background
space light is easily seen - such as stripes, bars, and
particularly, circles and spots. The second is movement -
even remarkably small and rapid movements (like the gills
of a nymph) provoke an instantaneous response in the trout's
To me the above observations seem to validate the
impressionistic approach to tying imitations. If the fish
cannot see fine detail then there does not seem to be much
point in worrying about exact imitation. Until now I have
always tried to tie close copies, but from here on I will be
concentrating more on overall shape and size and
incorporating more mobile materials in my tying.
The trout's eye is capable of detecting colour over a
slightly broader bandwidth than our own, extending into the
far red to about 800nm but not into the Infra Red (we see to
about 700nm) and is very similar to ours - the retina having
both 'rods', which detect only monochrome differences and
are very sensitive, and 'cones' which detect colour,
however, unlike us, the rods and cones have their own
distinct operating conditions dependent upon light levels.
During the day
the sensitive rods are withdrawn below the surface of the
retina and shielded from bright light by dark pigments
whilst the cones move to the upper layer to give optimum
colour vision. At night, when the light level falls to
below about 1-foot candle, the reverse is true with the
cones withdrawn and the rods exposed. Even in the brightest
moonlight the fish will not see colour, only shades of grey.
Work done with
Brown and Rainbow trout shows that there are however two
periods of light adaptation within this cycle, these
coinciding with dawn and dusk - the main feeding times. The
changeover from cones to rods starts well in advance of
darkness and takes about 4-5 hours but the reverse change
from rods to cones can take even longer. This might explain
Falkus`s `second half ` observation that a large fly fished
deeply seems to work better. At this time on a short summer
night the fish's eye would be well into the change to
`daytime` vision with the sensitive rods partially
withdrawn, but no usable light for the emerging cones to
detect. Could it be that with its vision at such a
relatively low level, a large fly drifting across its nose
would be just about all it can easily see?
Trout and salmon can see all of the colours that we can, but
whilst our eyes are most sensitive in the green area of the
spectrum, the trout's eye can discriminate best in the
blue/green region. A number of behavioural experiments have
shown that not only will the fish show a preference for blue
under most background and light intensity conditions, but
that they are able to differentiate between small, subtle
differences of shade. Second in sensitivity to blue comes
red (about 10 times lower sensitivity than blue) then black,
orange, brown, yellow and green in that order, but what it
actually sees depends on a number of factors not the least
being the turbidity (cloudiness) of the water in which it
As white light
passes through a column of water it is progressively
absorbed - the deeper we go the less light can penetrate.
The longwave light (red) is absorbed first and is virtually
non-existent at around 12-15ft so that any red materials in
a fly will appear to be black at this depth. Orange survives
to about 25- 30 ft. and so on until at about 60-70 ft. only
blue light penetrates. These are approximate figures typical
of very clear water illuminated by bright sunshine- the
light penetration will obviously be less in dirtier water or
in low light conditions. As an example, in a rising spate
river with a high level of suspended solids, red light would
disappear only a few inches below the surface.
The light as it passes through water is reflected off any
small particles (or even large molecules) and is scattered
in random directions. The more animal and plant life, i.e.
the murkier the water, the greater the effect. Because of
this scattering effect objects appear indistinct and fuzzy
in anything other than shallow water - just as they do to us
in mist or fog. Short-wave light (blue) is affected the most
by this, long-wave (red) the least - this is why freshwater
fish generally have a colour response more red shifted than
fish of the clearer open seas.
UV or not UV
indicates that Salmonids have cones to detect UV light when
small, but as the fish grow these cones gradually disappear.
Their diet in their early period of life consists of
zooplankton and other small creatures that reflect UV light
but as the fish get larger they can no longer filter such
food with their gillrakers.
along with the change in habitat as the fish develops,
is given as the reason why
no UV receptors are found in fish above 2 years old.
Other studies however have shown that new temporary UV
receptors are created annually to coincide with the spawning
migration but that these are used to detect polarised light
as a navigational aid. This would mean that returning
Sea Trout and Salmon do have some ability to see UV but I
have yet to see any evidence that would lead me to believe
that this is used for prey detection. It would seem unlikely
as we know that at this time their appetite is suppressed
and also, if it was a useful tool for finding food it is
logical that they would retain it throughout their life.
Finally, the reason that I began the search in the first
place was to answer a question that came to mind when I made
the original postings. I had known that fish of the open sea
have a colour vision response blue shifted relative to
freshwater fish (as a broad generalisation) - but what kind
of colour response do Sea Trout and Salmon have - living
their early life in freshwater then migrating to and from
the sea? The answer is that when these fish migrate to sea,
they shift the spectral response of their colour receptors
towards the blue end of their spectrum. The process is then
reversed when they return to freshwater. The
to run more or less parallel with the changes
in the fish's physiology which enable it to tolerate the
change in water salinity and this can take many weeks (and
may be different for Salmon and Sea Trout).
How significant this is I do not know but it seems to tie in
neatly with the anecdotal evidence. On my local river, many
of the old hands swear by a predominately blue fly when
fishing the lower reaches for fresh run fish. If the new
arrivals colour vision was still blue shifted or in the
transition phase, this might be a contributing factor to the
success of this colour choice.
Most of the above information seems to simply confirm the
methods developed over the years by anglers, but although
some might argue that it is mainly of academic interest, it
has certainly made me think a little more deeply about fly
tying and presentation.
Sea Trout Articles