METAMORPHOSIS OF AMPHIBIAN ORGANS 243 



The first color stage of the skin prevails during the largest 

 part of the larval period; it is characterized by a more or less 

 uniform pigmentation of the entire skin except the skin of the 

 belly, which however does not enter into our experiments. This 

 uniformity is produced during the first days of life by a continu- 

 ous layer of a yellowish pigment which is not contained in chro- 

 matophores, but exhibits a purely diffuse character. Soon after 

 the animals are hatched, melanophores appear scattered more 

 or less evenly throughout the skin of the head, back and tail. 

 With the gradual increase of the number of melanophores the 

 skin continually darkens without losing the uniformity of its 

 color. No definite patterns are worked out on the skin during 

 this period, which shall be referred to as 'even yellow brown' 

 in the following chapters. Of course, there are many variations 

 to be observed corresponding with individual differences as 

 well as with differences in the treatment of the animals and age 

 of the animals; but none are definite enough to play any role 

 here. Variations from a very light to a very dark shade usually 

 correspond with progressing age and also with similar differences 

 of the color of the background on which the animals are kept, 

 with the intensity of light and the differences in temperature. 

 Grayish, reddish or greenish tints may be intermingled but 

 without changing the uniformity of the coloration. Figure 1 

 indicates this stage; it shows a larva of Amblystoma punctatum 

 about six months old, kept in a dark room at 15°C. and fed on 

 worms. The same stage may be seen from figure 13. 

 * The next stage is the 'network.' It is characterized by the 

 appearance of a yellowish or greenish network on an even yellow 

 brown background. It also shows great variations: it may be 

 very dense or very loose; it may be produced by either fine and 

 numerous yellow streaks or by fewer and broader stripes and 

 blots. It may be of intense yellow or of a greenish color. Simi- 

 lar factors to those mentioned in the last paragraph seem to 

 produce these variations and we shall deal with some of them in 

 another paper. Here it is important to mention that the net- 

 work can be sharply defined as soon as it appears, though con- 

 siderable time is required to develop all of its characteristics 



THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 24, NO. 2 



