Visual Target Discrimination in Sharks — Tester and Kato 
469 
targets from a distance of at least 5 ft and 
sometimes at about 10 ft, supporting the his- 
tological conclusion of high sensitivity. 
Regarding form discrimination, Sutherland 
(1962) reports that no particular difficulties in 
discrimination between squares, circles, and 
triangles have been encountered in most animals 
that have been tested, including octopuses, min- 
nows, sticklebacks, pike, and a variety of higher 
animals. However, he points out that the angle 
of rotation of the figure was frequently im- 
portant; for example, he found that with octo- 
puses, a normal square (with horizontal base 
as in our tests) and an equilateral triangle were 
easier to discriminate than a diamond (square 
rotated through 45°) and an equilateral tri- 
angle. 
Clark (1959) successfully trained two large 
lemon sharks ( Negraprion brevirostris ) to asso- 
ciate a 16 inch square white target with food. 
Three nurse sharks {Gm glymo stoma cirra- 
tum ), however, failed to make a strong associa- 
tion. Clark (1961, 1963) also trained lemon 
sharks to distinguish between a square and a 
diamond, and between a plain white square and 
one with vertical stripes, but was unable to 
train them to discriminate a square from a cir- 
cle even with the large targets used. 
Our blacktips and greys readily discriminated 
between rectangles oriented at 90° to each 
other. However, in other test situations involv- 
ing circle vs. triangle and square vs. triangle, 
only two of five sharks provided positive re- 
sults. The shark’s difficulty in form discrimina- 
tion may be attributed to poor retinal resolu- 
tion, or possibly to differential ability in learn- 
ing, which, in turn, may be related to our 
methods. Hobson (1963) suggests that form 
discrimination may not be utilized by grey 
sharks in their natural environment. In feeding 
tests, he found no significant discrimination 
between whole baitfish (suitably slit to provide 
good olfactory stimulation) and decharacterized 
baitfish (heads and fins removed). 
Clark (1961, 1963) trained lemon sharks to 
distinguish between a white and a red circle, 
and a white and a red square. As in our ex- 
periments, the luminosity factor was not elimi- 
nated. 
In our tests with blacktip and grey sharks, 
some subjects were able to distinguish, but 
with apparent difficulty, red and yellow, and 
possibly green and blue also, from grey targets. 
As indicated above, it still remains a question 
as to whether the sharks were responding to 
differences in brightness or to hue. The colors 
were chosen for maximum chroma and, to the 
human eye, presented a vivid contrast with grey 
when viewed through water. Although the il- 
lumination was somewhat low, measuring 28 
ft-c at the surface and 11 ft-c at the level of 
the targets, there was enough light to allow 
color vision, at least for animals with cone-rich 
retinas that have demonstrated the ability to 
distinguish hues. For humans, 0.01 ft-c is suf- 
ficient for photopic vision (Moon, 1961). 
Walls (1942) reports several workers’ findings 
that the minnow Phoxinus laevis matches human 
ability in regard to the illumination level at 
which they can perceive hues. John (1964), 
utilizing schooling responses of Astyanax mexi- 
canus, found a cone threshold in the order of 
0.001 ft-c. 
Eyes of blacktips and greys kept in the shark 
house were nearly in a completely light-adapted 
state: the pupils were almost slits, and the 
tapeta were nearly completely occluded by dark 
pigment. It is possible that a small increase in 
illumination might have raised the sharks’ visual 
ability, as it would certainly have done for ani- 
mals with duplex retinas. However, optimum 
light conditions for blacktip and grey sharks, 
with all-rod retinas, may not necessarily be the 
same as those of animals with cone-rich retinas. 
There was no noticeable difference in learn- 
ing rates or ability between the simultaneous 
and successive techniques employed. It has been 
shown (Sutherland, 1962) that the former 
method is more advantageous if very small dif- 
ferences, such as neighboring shades of grey, 
are to be discriminated. Using the simultaneous 
technique, shades of grey differing by 3 Munsell 
units were distinguished by a grey shark but not 
by a blacktip. Using the successive technique, 
shades of grey differing by 2 Munsell units 
were distinguished by a second blacktip. 
Difference Between Species 
No consistent differences were found in the 
visual capabilities of blacktip and grey sharks. 
