316 
pupil is commonly seen to be widely 
open in bright sunlight. But the wide 
pupil by itself would emphasize chro- 
matic aberration, and we have to look 
elsewhere for a means of reducing its 
effect. A possible answer to the prob- 
lem lies in the pigmented oil droplets 
which have been known for many 
years to be present in the cones of 
birds; and these droplets are so situ- 
ated that light passes through them to 
reach the light-sensitive terminal seg- 
ment of the cone. They are generally 
red and yellow in diurnal and color- 
less in nocturnal species, the propor- 
tion varying both with the species and 
with position in the retina, and a very 
plausible explanation of their distri- 
bution has been developed by Walls 
and Judd. The relevant point to the 
present argument is that the cones of 
the fovea apparently always contain 
yellow droplets, and only yellow drop- 
lets; so here we have a device which 
seems capable of absorbing the blue 
and to a lesser extent the red ends of 
the visible spectrum, and so suppress- 
ing the chromatic fringes which would 
otherwise mar the image.’ The pres- 
ence of the yellow droplets does not, 
of course, mean that the bird is color- 
blind, even for foveal vision, for it is 
not an absolute freedom from chro- 
matic aberration which is the target. 
The absorption of the blue and red is, 
presumably, nicely calculated to per- 
mit of full use of the foveal cones in 
resolving detail with the least possible 
sacrifice of hue discrimination. 
We may conclude that birds with 
the keenest vision are potentially ca- 
pable of an acuity two to three times 
that of man. 
COLOR VISION 
The mechanism of color vision is 
still so littlke understood in man that 
5 It is still an open question whether the 
yellow macular pigment does or does not 
extend over the central part of the human 
fovea. If it does we have here another inter- 
esting example of convergence between the 
human and avian eye. 
ANNUAL REPORT SMITHSONIAN INSTITUTION, 1948 
no purpose would be served by a gen- 
eral discussion here. The fact which 
seems to be established by the best 
experimental work is that there is very 
little difference between the color 
vision of birds and man. The ratio 
of luminosity to wave length for pure 
spectral colors, and also the number 
of distinguishable hues in the spec- 
trum, is not significantly different, 
which is a little surprising since the 
work of Watson (1915) and, later, 
Lashley (1916) was done on the ban- 
tam, whose retina is rich in cones with 
red oil droplets. It must be presumed 
that there are still enough cones with- 
out colored droplets in the retina to 
register blueness unimpaired. No 
method has been developed for assess- 
ing foveal color vision alone in birds. 
Although the limits of color vision 
seem to be the same for man and birds, 
it is likely nevertheless that color sense 
plays a more emphatic part in the 
visual sensations of diurnal birds than 
of man simply because of the density 
of cones in the peripheral retina. If 
a man looks at a distant part-colored 
object, such as a flag flapping against 
the sky, at first directly and then at an 
increasing angle, he finds that at 20° 
the individual colors can no longer be 
distinguished and the hue of the flag 
is uniform though inconstant. At 30° 
there is no sensation of hue whatever 
and the flag is just a dark moving 
object. This loss of color definition 
must be attributed to the dilution of 
cones with rods in the periphery of the 
human retina, and consequently it will 
not occur in the most completely 
diurnal birds, which will see hue as 
well as detail in its full emphasis over 
the whole visual field. 
MOVEMENT PERCEPTION 
If the eye can be said to have a 
primitive or intrinsic function, that 
function is undoubtedly the perception 
6 Despite this work, many students of bird 
behavior are convinced that birds do not see 
as far toward the violet end of the spectrum 
as we do. 
