94 THE INFLUENCE OF INANIMATE SURROUNDINGS. 



cells are all supplied with pigment, as in many reptiles, but 

 these certainly are not true chromatophores, and so are in- 

 ci.pable of occasioning any change of colour in the akin ; but of 

 course their constant hue must affect that of the skin gene- 

 rally, as well as the marking produced by the more deeply 

 seated chromatophores. The true chromatophores lie in dif- 

 ferent layers in the cutis ; close to the epidermis, light-coloured 

 yellow cells occur, beneath them the red or brown, and, in the 

 deepest layer, the black. In some spots the pigment-cells of 

 one kind or the other may be wholly wanting ; sometimes the 

 black ones form a close mass in one spot, w-hile in others the 

 red or yellow predominate, but very few spots ai-e devoid of 

 pigment altogether. It is on this distribution and stratification 

 of the chromatophores and their alternate expansion and con- 

 traction that the pattern (so to speak) depends which the frog's 

 skin displays at any given moment. If all the chi'omatophores 

 are relaxed, brown or black will predominate, and in the spots 

 where light-coloured chromatophores lie in patches their hue 

 will be dulled ; if they contract while the light ones are still 

 extended, these latter will be more conspicuous. Heincke 

 detected in Gobius Ruiltens'parri yet another kind of chromato- 

 phores, which were filled wdth iridescent crystals of marvellous 

 delicacy ; they are visible, according to the degree of contrac- 

 tion, as spots of metallic sheen, or are altogether invisible. 



The property on which the contractility of the chromato- 

 phores depends is at present unknown, although various 

 hypotheses have been suggested in explanation of it. It is of 

 little importance for our purpose to leaxn which of all these 

 antagonistic hypotheses will ultimately be proved to be the 

 right one, since we know that all living protoplasm is essen- 

 tially contractile, and moreover that all cells devoid of mem- 

 brane — like the young ovum-cell, the white corpuscles of the 

 blood, and others — sometimes possess this contractility in a very 

 high degree. And the chromatophores belong to the class of 

 cells without membrane ; hence we need not be surprised to 

 find that they contract like other similar cells. 



It was formerly supposed that the exciting agent which gave 

 rise to the contraction of the chroma' ophores must act upon 



