94 IMITATIVE COLOUmNG 



on, in order to remove the dark cells entirely, and thus leave the 

 true colour always visible. Our success in this direction has been 

 very small. The principal result has been to make us familiar with 

 the different forms assumed by the pigment cells, and the means 

 by which the changes can be most easily brought about. Our 

 observations having been made upon fish having scales, has been 

 rather a disadvantage, obliging us to depend more upon the fins 

 and tail than the general changes in the body. The skin under- 

 neath those scales having black pigment cells on their surface, is 

 usually found deeply dyed by pigment ; the eye and the interior 

 walls of the abdomen are also deeply set with dark cells. It is 

 difficult to say how far the violence done to a fish by taking it out 

 of water, or removing a scale, or cutting off part of a fin, may disturb 

 the nervous action through the whole body, but judging from a 

 large number of examinations made under different conditions, it 

 appears that the cells under ordinary circumstances are to be found 

 displaying at one time almost all the forms of which they are cap- 

 able; some having the form of a sphere, some flattened and showing 

 broken margins, some having tongue-like projections two or three 

 in number ; while others extend in radiating arms until the central 

 nucleus almost entirely disappears ; and in some the extension of 

 the cell has been carried so far that the colour is pale grey, hardly 

 distinguishable from the skin substance in which it is embedded. 

 To alter the form of these cells and cause them all to take on a 

 similar action, it is only necessary to expose the fish to different 

 kinds of reflected light. The light from a white porcelain dish 

 causes the contraction of all the cells into the least possible space, 

 and when continued, causes the cells to become small black globes ; 

 conversely, when a fish is placed in a black dish, the cells com- 

 mence to extend in every direction until they frequently touch and 

 interlace with each other. 



Ordinary cold water and the simplest means have been found 

 as efficient as the most elaborate precautions. We have tried 

 globes coloured inside with red, blue, yellow, and green ; we have 

 also used different colours of glass, so that the ligJit within the 

 globes should be tinted according to the colour employed, and 

 used the greatest care so as to carry on our experiments in warm 

 water and under conditions as nearly as possible similar to those 



