Physiologie Neuronaler Systeme

Dept. of Physiology, Med. School, University of Vienna


3)Two main changes in mammalian photoreception

As it turns out when we compare mammalian eyes with those of reptiles and birds
there are two major differences right at the beginning of the visual process:

3.1) Mammals have retinas dominated by rod photoreceptors

.....(with some IMPORTANT EXCEPTIONS - see below).

The rod system (rods and their specific set of retinal interneurons) expanded and was linked ("plugged")
into visual pathways which had been designed primarily for cone vision.

This is not so surprising if we remember the primary lifestyles of the first mammals.
For a long period of time (ca.50 Mio years) the early mammals adopted a nocturnal lifestyle.
They could do this because they became able to keep their body temperature
independent of the environment (homoiothermy).
Reptiles and other animals have to warm up to become active.
This is limiting their to warmer to temperate climates or daytime hours.
The small early mammals however could shift their activity to night times while at the same time
avoiding most of the sharp eyes of diurnal reptilian predators.

How then could it happen that dinosaurs declined while mammals expanded
with many new families of species across the earth ?

Current discussion favours the impact of an asteroid as the initial event
for the decline of the dinosaurs (others volcanoes or interstellar dust).
Maybe several conditions created a "cocktail of causes" (see article in "Beagle").

In any case such catastrophies may also have had another important consequence:
a layer of ashes may have dimmed down the available light dramatically
(similar to the scenarios for a nuclear war).
Then clearly animals adapted to low light should have done better, in particular when they were homoiothermic.

Animals with vision depending on direct light at high intensities may have been slowed down
or even blindfolded. Even for the immediate survivors with cone-cominant visual system
this conditions will have reduced the daily and annual activity periods immediately and drastically
and: there was no time to evolve better adaptation.
Thus the dinosaur visual system may have turned into a major handicap perhaps contributing to their
However: other groups that were adapted to daylight vision (like turtles) did survive too.
So no simple answer is at present at hand for the decline of these large reptiles.

This diagram gives a simplified visualization of this speculative correlation between the darkening events at the turn to the Tertiary and the rise of the mammals.

In any case: the rod-dominant eyes of early mammals may have been one of the features
for making them the "unexpected" winners of this critical turn of evolutinary history,
The "period of mammals" lasts until present day.

Here are two typical examples of mammalian photoreceptor layers.

This is the retinal sensory surface (matrix, mosaic - whatever metapher you like)
from a mammal with ancient features (see further below) : the Opossum.
Dispersed among the dominant lattice of rods a few shorter cones are hidden in the flat notches .
The blue-green colorizing is intended to indicate the wavelengths (ca. 500 nm)
that is most likely absorbed by the rod photopigment (Rhodopsin).

This is a small patch from cat photoreceptors. By histochemical incubation (NBT ) the rod photoreceptors,
covering the major part of the surfaceare, are darkly labeled. Consequently the unlabeled cones stand out like stars on a night sky.

There is always a trade-off for specialization:

* To achieve maximum light sensitivity the signals of the rod photoreceptors converge onto
relatively few interneurons and are pooled along ganglion cell pathways.
Thus a particular ganglion cell does cannot locate the exact origin (position)
of a light spot within its receptive field.

This precludes the rod system from obtaining high acuity as well as from spectral discrimination.
These features require separate pathways of information originating from a single or a few cells (cones) at most and weighted processing.
Our anthropocentric view (sic!) with its daylight activity pattern makes this seem like a big handicap.
However sophisticated spatial and color discrimination probably makes not much sense anyway
for nocturnal creatures compared to the advantage of optimally exploiting available light for roaming around.
TIP : Watch a pet hamster, mouse or rat. They don't bother doing their business in the dark
relying mostly on their ears, nose and whiskers, but if the light is suddenly turned on
they may become scared and "freeze". It's usually no good to be seen when you are that small.

3.2 Mammals have a reduced number of spectral cone subtypes

Also the second reptilian feature - multitude of polymorph spectral cone types -
was dis-continued allong the mammalian line.

In most mammals only two cone photopigments are present -usually in two respective cone types :
short wavelength sensitive cone is antagonized by a long wavelength sensitive cone system.

The trichromatic system of OldWorld primates may be seen as a refined variant (see on next page )

2 Cone Pigments

(Cone Types)


Absorption Curves of photopigments




Antagonistic Processing in Retina and Brain



This basic set is called the Ancient (sub)system of color vision (Mollon).

Thus most mammals are dichromats, usually being only able to discriminate between bluish and greenish components.
For many animals like rabbits or hamsters red objects appear dark and not so interesting (probably also for the bulls in the arena).

These animals have a "neutral" point in the blue-green area of the spectrum.
They cannot distinguish thes hues from certain shades of grey:

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .<-- |--> neutral point

Simulated spectrum of a Blue/green (Protanope) dichromate

Mammals - with exceptions of some primates - have a (relatively simple) dual cone system.

Reptilian set of photopigments

Rhodopsin, UV-, Blue, Green, etc.

Basic set in mammals:

Short wavelength and Middle (Long) wavelength sensitive opsins

Apparently the genetic information for expressing this variability was lost forever
during the "Mammalian Dark Ages". This even holds for species
with highly visually oriented behaviour such as the cheetah.

This cheetah lives in the

former Emperor's

Zoo at Schönbrunn Castle,


Even there it gives agood example of blending into the environment by the fur's camouflage colour and patterning.


However, mammalian radiation still has led to astounding variability. The following will illustrate these principles.

Photoreceptor diversity -
Mammals make the best out of the limited set
inherited from the nocturnal ancestors
by varying their proportions and topographic patterns.

The disappearance of the dinosaurs enabled the mammals to radiate quickly.
Rapidly evolving groups started to re-conquer the various habitats including the diurnal niches previously dominated by reptiles.
the features of a cone system - good spatial and color vision -
(for some groups at least)
became more important again.

This is a small sample of Guinea Pig photoreceptors (high magnification).
In this retina cones would be discernible without further marker on a wholemount preparation.
The cones are also somewhat more frequent than in cat but
rods (pale small diameter) are still the majority.
From electrophysiology etc. we know that these cones carry two types of photopigments (Blue and Green).
As in most mammals consistent morphological differences are not available to distinguish these basic cone subtypes
(as e.g. in turtle -> above).

In this experiment the cones are further labeled bluishly by Formazan precipitate.
These experiments with stimulus related NBT-incubation resulted in rather inhomogeneous labelings.
Today we know that this was a likely consequence of a dorsoventral opsin-gradients
and varying amounts of opsin co-expression.

Up to present day only three mammalian groups again developed retinas dominated by cones:





in prep.

Most mammals however have typical "Duplex retinas" with a dominant rod populations
Recently the apparent simplicity of the dichromatic scheme has been shown to carry some more surprises:
  • It was shown that the "Blue" cone receptors in some rodents such as mice is actually UV sensitive.
    The reason for this shifted spectral positioning is unclear.

  • Besides rod/cone regionalization surprisingly also
    reciprocal regionalization of spectral cone subpopulation has been found
    in species such as rabbit or wallaby (see also opossum below).

  • Furthermore "Opsin- Coexpression" - presence of two photopigments within the same cones
    of many rodents makes normal color discrimination seem obselete in favour of broadening the spectrum.

Pages on these topics are in preparation.

For recent review see:
.To Top of Page |
Ahnelt P.K., Kolb,H. (2000)
The mammalian photoreceptor mosaic - adaptive design.
Prog.Ret.Eye Res. ,19:711-777

Next page: 3) Primates: back to Trichromacy