262 



AMERICAN TESTUDINATA. 



Part II. 



and by their separation through cells of the Malpighian layer. Generally tliis Huid 

 is again combined with the pigment cells described above, the forms of which, more 

 or less radiated, more massive or more slender, j^roduce again difterent effects. 

 Under the microscope, the free fluid coloring matter looks generally yellowish, if 

 the effect is yellow; reddish, if the effect is red. Water added to this fluid when 

 taken from the living specimen, causes it to collect in larger and smaller drops, 

 and then their oily character^ becomes evident by the characteristic blackish margin 

 of the drops. We have still to mention another kind of color, which we see 

 only in one genus of North American Turtles, namely, the white on the head of 

 some specimens of the genus Cistudo. This appears under the microscope to be com- 

 posed of grayish black heaps, and if these are farther isolated, we find them com- 

 posed of thin transparent plates, breaking like glass. All these pieces together 

 jiroduce the impression of a white tint upon the eyes, by interference of the rays 

 of light, just as the powder of glass, the smallest pieces of which are also trans- 

 parent under the microscope. 



The range of variations which the colors exhibit in one and the same species, 

 in many genera of our Testudinata, is almost incredible ; and unless these variations 

 are carefully studied, and their transitions watched for a long time, in every stage 

 of growth, it is impossible to know how far they agree with the natural limita- 

 tion of species. For this reason most descriptions of the colors of our Turtles 

 are incomplete and unsatisfactory, being generally drawn from a few specimens. 

 In several instances, nominal species have been distinguished merely upon differ- 

 ences in the coloration. This has been done to the greatest extent in the genus 

 Ptychemj's, as we shall see hereafter. Generally speaking, there are, however, cer- 

 tain tints which prevail in some species, while other tints are more common in 

 other species, and in these cases the colors afford, to some extent, good specific 

 characters. But it sometimes happens that not only the patterns of coloration, 

 but even the colors themselves, are the same in every species of the same genus, 

 so that coloration requires a special preliminary and extensive study for every genus, 

 before it can be applied to the systematic characteristics of these animals. 



' In relation to the nature of this oil, see D. 

 Weinland on Birds' Feathers, in Cabanis, Journal 

 fur Ornithologie for 1854. He supposes that the 

 yellow oil turns reddish by a kind of oxydating 

 process, and thus, perhaps, also the reddish to 

 brown, and this to black. Such an oxydation takes 

 place, as we know, for instance, with extravasated 

 blood, which turns black very likely by a pro- 

 cess of burning. It is true, this is a pathological ex- 



perience, and it may not seem proper here to refer to 

 it ; but pathology rests upon the same laws of organic 

 chemistry as physiology. For studying these colors 

 in Turtles, we recommend as fine objects the red and 

 yellow rings on the marginal plates of Chrysemys 

 picta and marginata. The beautiful brown-green 

 color of the dorsal shield of the latter is produced 

 by a network of black lacuna; lying on a homoge- 

 neous layer of yellow oil. 



