NetNews Usenet Archive 1992 #27/NN_1992_27.iso/spool/sci/bio/4213

Internet Message Format | 1992-11-24 | 7.2 KB

Path: sparky!uunet!olivea!spool.mu.edu!darwin.sura.net!jvnc.net!netnews.upenn.edu!lepomis.psych.upenn.edu!rowe
From: [e-mail] (Mickey Rowe)
Newsgroups: sci.bio
Subject: Re: More on color vision
Message-ID: <[e-mail]>
Date: 24 Nov 92 00:44:29 GMT
References: <[e-mail]><[e-mail]> <[e-mail]> <[e-mail]>
Sender: [e-mail]
Organization: University of Pennsylvania
Lines: 137
Nntp-Posting-Host: lepomis.psych.upenn.edu

In article <[e-mail]>
     [e-mail] wrote:

>[Some scoundrel named Mickey Rowe wrote:]
>>[color vision  widespread in teleost fish as well as in birds]
>
>Thanks, I wasn't sure how widespread it was in teleosts (I knew many
>teleosts had color vision but not if it was very common or not.)

Well, I suppose that could depend on how you define common.  Color
vision is more or less absent in fish that live in the deep ocean and
dark caves, but I think it's pretty common in most others.

>>>Mammals on the other hand usually don't have very good
>>>color vision -- especially nocturnal mammals.  Diurnal mammals, like squirrels
>>>and primates, especially fruit-eaters, tend to have very good color vision
>>>but a lot of the other mammals have some poor excuse for color vision or
>>>none at all.
>>
>>Actually squirrels don't have really good color vision,

I should have double checked before I went spouting off--I think that
I must have gotten that impression from the fact that squirrels are
dichromats.*  I'm probably exhibiting a bit of anti-mammal bias by not
being terribly impressed with an animal with only two visual pigments.
Squirrels do have mostly cone retinas, though, so they are probably
near the top of the line in terms of mammalian color vision.  (Big
fish in a small pond and all that :-) Of course, it may turn out that
squirrels are actually trichromats, but I'll get to that next.

*brief foray into nomenclature--chromacy in an animal has to do with
the number of lights *necessary* to recreate the sensory experience of
any arbitrary light.  Normal humans are trichromats--with three
suitable lights you can create the experience of *any* other visible
light.  That's why TV sets and color film can get away with three
phosphors and three pigments respectively.  Although it's not
necessarily true, generally the number of lights needed to reproduce
any other light in an animal's sensory world is the same as that
animal's number of cone pigments.  It's sloppy, but often people will
claim that, for instance, goldfish are tetrachromatic because they
have four cone pigments.  That may not be correct, because it depends
on the manner in which the signals from the photoreceptors are
processed.  Since I'm relying on a review paper at the moment, I'll
presume that the statement that squirrels are dichromats means that
they have two demonstrated cone pigments.  (I hope that's clear...)

>>but they do
>>seem to be better at making "color" discriminations than many other
>>mammals.  However this impression might be due in part to people
>>looking for color discriminations in the wrong wavelength bands.
>>Within the past couple of years, evidence of UV detection has been
>>discovered in three different rodents (mice, rats and gerbils if I
>>remember correctly).

I've since checked.  Evidence of UV sensitivity has been found in
mice, rats, gerbils and gophers.  By inference, you might conclude
that UV sensitivity is common in rodents, so I wouldn't be surprised
to find that squirrels are trichromatic.  I don't have any references
on hand that say what part of the spectrum the two known classes of
squirrel cone are sensitive to yet.  I'm in the process of tracking
down a reference that will probably tell me, so I'll decide when I
find that information whether or not I want to go out on a limb and
predict trichromacy in squirrels.

>>Apparently it's difficult to get cats to make discriminations based on
>>"color" even though they have all the necessary ingredients for it.
>
>What do you mean "necessary ingredients"?  Do they have a high
>concentration of cones?  I'm curious...

The main thing I had in mind was more than one cone class, but one of
my main references at the moment (Goldsmith, T.H. (1990).
     "Optimization, Constraint, and History in the Evolution of Eyes",
      Quart. Rev. of Biology, 65(3):281-322.) has this to say about
cats and rats:

     Rats and cats, both of which have been examined with some care,
     do not learn to make color discriminations easily and possess few
     cells that seem to be involved in color vision."

So no, cats don't seem to have a high concentration of cones.  The bit
about the rat should be taken with a slight dose of salt, though,
since the above statement was written before the UV sensitivity was
discovered.

>>>A general rule of thumb for figuring out if an species has color vision is,
>>>do they eat brightly colored things (fruits, flowers) or communicate with
>>>each other with color in some way (bird plumage, lizard dewlap
>>>displays)?  In other words would color vision be useful for them in some way?
>>
>>Careful on the plumage thing... I take it you've heard of the pitohui
>>by now?  Bright coloration also signifies "I'm poisonous; don't eat
>>me!"  Since distinctive coloration is nearly ubiquitous in poisonous
>>animals (and plants?) one might argue that color vision is useful to
>>just about *any* animal.
>
>I was especially careful to say "communicate with EACH OTHER with color"
>for exactly this reason.

Ok.  I was just trying to clarify a point about drawing conclusions
from an animal's coloration.  The second part of my paragraph drifted
onto a related tangent, though, which is that color vision is useful
even to drab animals that sometimes need to recognize what other
organisms might have to "say" with *their* coloration.

>Obviously there are many animals that are
>brightly colored for entirely different reasons than communication with
>each other -- most notably, bright coloration to advertise toxicity, and
>bright coloration as some form of camouflage (e.g. green caterpillars on
>leaves).

And of course there's coloration in non-toxic animals that mimic the
coloration of animals that are toxic.  But I'm digressing again :-)
(not to mention speaking in fragments :)

>At any rate, I only meant the "do they need color vision" argument as a
>rough rule of thumb, to get people thinking about the costs & benefits of
>color vision.  Clearly, to really demonstrate color vision you need to do
>a lot more than see if the animal has colored feathers or eats fruit.

Agreed.

[Stuff about pitohui deleted because I don't know whether or not
they're monomorphic either...]

[Stuff about "monomorphic" birds being dimorphic in regions of the
spectrum to which we're not sensitive deleted--The only thing I have
to add is that I'm not sure that I'd heard about that before, thanks!]

There was a little bit more in Kathleen's post asking about
polarization vision, but I'm going to put off responding to that
momentarily.  This has gotten long, and I fear that if I don't answer
soon the conversation will die.  (Also I can nudge Dave Cameron now
and see if he'll talk about polarization vision so that I don't have
to :-)

>Kathleen