MADE AT THE PLYMOUTH LABORATORY. 271 
III. A Presatp Puatcer. 
In the memoir by Dr. MacMunn and myself on the Coloration of 
Fishes (Phil. Trans., 1894) specimens of flat fishes are described in 
which some portion of the upper side is abnormally destitute of 
pigment. Thus abnormalities consisting in the pigmentation of 
part or whole of the lower side are balanced by abnormalities con- 
sisting in the absence of pigment from areas on the upper side. I 
have recently received a living specimen of the plaice exhibiting 
the latter kind of abnormality. It was caught in the Hamoaze, 
and brought to the Laboratory on October 3rd. It is still living in 
one of the tanks. The anterior third and the caudal third of the 
upper or right side in this specimen are pigmented as in a normal 
plaice, the red spots having the usual appearance and position ; but 
the middle third is white like the whole of the lower side. The 
white unpigmented area is bounded by two definite irregular lines, 
the anterior passing transversely across the body just behind the 
pectoral fin, the other in the posterior region of the body. There 
is an isolated round patch of normal pigment within the white area 
dorsally. 
Mr. Holt, in the previous number of this Journal (p. 188, et seq.), 
gives reasons why my rejection of atavism as an explanation of the 
abnormal coloration of the lower side cannot be held to be valid. 
But I can see no reason why the principle which explains the 
occurrence of pigment on the lower side should not also explain its 
occasional absence on the upper. If it is atavism in the one case it 
is atavism in the other, and the occurrence of piebald plaice, or flat 
fishes white on both sides, is as good an indication that this family 
of fishes is descended from ancestors that were unpigmented on 
both sides, as the occasional presence of pigment on the lower side 
that they are descended from ancestors coloured on both sides. 
Thus it is obvious that atavism fails to explain both kinds of 
abnormality, whereas the explanation adopted by Mr. Bateson and 
myself applies equally well to either case. That explanation is 
that in certain cases one side, instead of developing normally, 
partially or completely imitates the other. It does not require 
much consideration to see that Mr. Holt’s reference to what he 
terms ambi-ciliation, tends to support my views, and not his 
own. For it is difficult to connect the varying conditions of the 
dermal armature in different kinds of flat fishes with an original 
ancestral condition. If the ciliated scales of the brill are ancestral, 
then the tubercles of the turbot are new, and vice versd ; but when 
