Vol. X] VON GELDERN—SKIN OF ANOLIS CAROLINENSIS 107 



the epidermis sufficiently to block the action of both the yellow- 

 oil droplets and the leucophore layer. The pigment granules now 

 act entirely alone and produce the brown state (Figs. 12 and 13). 



Whether the yellow oil droplet layer can be dispersed under 

 varying stimuli in those scales where it normally exists, that is, 

 whether it is possible for a lateral green scale to assume the white 

 or bluish state in a short space of time, is difficult to say. If, as is 

 held by Keller for the chameleon, the zanthophores possess the 

 power of dilating and can take up all the space held previously 

 by the oil droplet layer and thus displace the latter to the edges of 

 the scale, it would at least be possible for a fairly rapid change to 

 occtir from green to white. Pouchet, however, believes the zan- 

 thophores of the chameleon to be of yellow color and if this be so, 

 Keller's explanation would not be valid. The zanthophores of 

 Anolis appear transparent in all the conditions under which I could 

 observe them. They seem to be absent in the skin of the ventral 

 surface of the body. However, my observations are by far too 

 meager for me to draw any definite conclusions. It is true that 

 the zanthophores seem greatly increased in the white scales of the 

 mid-dorsal stripe and that they form a transparent fairly thick 

 layer between the epidermis and the leucophore layer. 



The objections to Keller's views are that if the zanthophores 

 are able to disperse the oil droplet layer by dilating, they would 

 probably also disturb the arrangement of the terminal branches 

 of the melanophores, which we know'Soes not take place. One 

 would also conclude that these cells must be under the control of 

 the nervous system. No conclusions can be reached without 

 observing the living animals for long periods of time and noting 

 the changes of these white areas. Possibly the white areas are 

 merely variations and are more or less permanent. Black or brick- 

 red stripes may be also variations in the distally migrated pigment. 



The following table presents in a concise manner the mechanism 

 involved in the various color states. Keller's theory in regard to 

 the behavior of the zanthophores is included as a possibility, 

 since their attributed function becomes necessary to explain 

 white and blue changes in certain cutaneous areas, if these occur 

 with any degree of rapidity. 



