254 
SCANDINAVIAN FISHES. 
brilliancy. The pretty network is much less distinct; 
the central collections of pigment which were previously 
punctiform, have now increased in extent. But on the 
body, which is thus almost dark, the emerald spots on 
the middle of the sides still glitter with undiminished, 
or, if possible, still greater brilliancy. We now see 
that these spots are composed of nothing more than 
chromatophores exactly like those which form the selli- 
form spots, but subjected to extraordinarily strong 
contraction. It is the immense number of glittering 
disks that are here collected within a comparatively 
small space, that gives these spots their iridescent lustre, 
which, when the chromatophores are expanded and the 
glittering disks distributed over a larger surface, passes 
into the dull gleam of the selliform spots, the top of 
the head and the belly. 
“If we now place our darkened Goby on a, light 
bottom of straw or sulphur yellow, for example, after 
about an hour the fish is as light as it was dark before, 
and the selliform spots are invisible. Under the mi- 
croscope we observe the following peculiarities. The 
black chromatophores are contracted, the yellow ex- 
panded. As the former are far more numerous than 
the latter, we may sometimes see the yellow bottom of 
the vessel shine through at certain parts of the body, 
where only contracted chromatophores occur, especially 
at the tail. This phenomenon is due to the thinness 
of the body. In the selliform spots there now appear 
numerous, yellow, expanded pigment-cells; but in addi- 
tion to these the network of pale, lustrous colour is 
sharply marked. 
“Let us now move the fish, which is partly of a 
yellow colour and partly transparent, showing the yellow 
bottom, over some red substance. Instantaneously it 
assumes a transparent, reddish tint, and the selliform 
spots, which were invisible on the yellow bottom, be- 
come distinct for a moment. There is no other change in 
the degree of expansion of the chromatophores than that 
the yellow, as well as the black, are contracted, though 
not so strongly; while the few red chromatophores on 
the upper surface of the body, which previously looked 
like dots varying from brownish red to black, now begin 
to form pretty projections and large processes of a red- 
dish brown or cherry colour. The rest of the body is as 
transparent as before. The selliform spots are visible, as 
they contain no chromatophores, and therefore stand 
out on the red colour which shines through from the 
bottom on which we have placed the fish. 
“If we now change the substratum from red to 
blue and from blue to light green, we see that each 
time we change the colour, the fish for a moment 
assumes the corresponding tint, on account of its trans- 
parency. Exactly the same results may be obtained by 
setting a dark fish for a somewhat longer period, on 
a red, blue or green bottom. The adaptation seems to 
take place most slowly on a green bottom. If a spe- 
cimen is kept for some time on a light (white or light- 
coloured) bottom, even the handsome, lustrous, lateral 
spots grow pale, evidently on account of the expansion 
of the chromatophores. Last of all the black caudal 
spot disappears, the pectoral spot having already van- 
ished. The handsome stripes on the fins gradually fade. 
The male becomes exactly like the female, and in this 
dress it is quite impossible to distinguish either of’ them 
by the colour from a young Gobius niger of the same 
size that has been kept on a light bottom. But if we 
take the two species out of their narrow glass vessels 
and again put them among the plants in a large aqua- 
rium, all the characteristic differences of colour between 
them soon reappear. Gobius niger never possesses the 
selliform spots or either of the caudal and pectoral spots 
that distinguish Gobius flavescens. The faintly lustrous 
chromatophores, which are also present in the former, 
are there irregularly sprinkled in small groups over 
the surface of the body, and, when contracted, form 
irregularly scattered, small, bright dots, but are never 
arranged with the same regularity as in Gobius flavescens. 
“In short, Gobius flavescens possesses in a high 
degree the power of adapting its own colour to that 
of the bottom with comparative rapidity. If the colour 
is such that the fish possesses chromatophores with the 
corresponding pigment, these are expanded. If there 
are no cells containing the corresponding pigment, all 
the chromatophores are contracted, and the consequent 
transparency of the body brings about the same result. 
The metallic lustre is due to extremely fine, crystalline 
disks, which are either spread on the exterior of a 
pigment-cell or completely fill it. These cells perform 
their function in a manner quite the reverse of that 
of the monochromatic cells; the stronger the contrac- 
tion, the brighter is their lustre, and the greater their 
expansion, the duller they appear. 
“The same variations of colour as we have now 
produced artificially, so to speak, also proceed in an 
aquarium, if its colour-relations correspond to those of 
the natural home of the fish. The only difference is 
