XI THE COLOURS OF AMPHIBIANS 231 



difference expressing itself especially in diminished 

 brightness of tint and in the absence of the cuticular 

 modification which gives rise to a blue colour. 



Relation of Larval and Adult Coloration 



It is unfortunate that the paucity of observations 

 makes us unable to say much on this interesting 

 subject, but the statements given above suggest one 

 or two points. 



First as to the position of the pigment and its 

 meaning. We have seen that the dark pigment, 

 both in adult and in larva, occurs to a slight extent 

 in the epidermis, and much more markedly in special 

 pigment cells in the dermis or true skin. The 

 epidermis, with its contained pigment, is periodically 

 cast and renewed, so that there must be a slow 

 elimination of pigment from the body in this way. 

 We have already seen that it is most probable that 

 the pigment originates in the dermis or deeper tissues, 

 so that there must be a continual migration of pig- 

 ment from the dermis to the epidermis. Such a 

 state of affairs occurs very frequently in vertebrates, 

 and is, as noted above, often regarded as a proof that 

 the coloration is the result of an excretory process. 

 The development of the coloration, as already 

 described, supports this theory in so far as it tends 

 to show that the dark pigment usually increases in 

 amount during growth, and that it tends to be more 

 abundant in males than in females, in normal indi- 

 viduals than in those which are precociously sexual. 

 The proof would, however, be more cogent were it 

 not for the presence of the lipochromes. These can 



