12 BULLETIN OK THE KUREALT OF FISHERIES. 



than that of the female, and posteriorly more triangular; the marginal plates are 

 apt to be revolute, the head is smaller, the nose sharper, and the tail longer. All 

 these characters, except the difference in size, are shown in the illustrations accom- 

 panying this paper. In the descriptions the female has in every case been placed 

 first, as the material available consists almost entirely of this sex. The laws of 

 several states where the industry is conducted forbid the sale, of terrapin less 

 than 5 or 6 inches in length, and therefore only females are caught and sent to 

 market, the males being thrown aside as worthless. Largely on this account the 

 collections of terrapin, in museums as well as in the terrapin pounds, contain very 

 tew males. Furthermore, in the pounds the males are thrown together regardless 

 (if the locality from which they came. If possible, this sex is more variable than the 

 other and is really not as satisfactory for the purposes of the present paper. 



Of the Georgia and the Louisiana terrapins, the writer has not seen perfectly 

 satisfactory male specimens, but males of all the other species have been examined 

 and will be dealt with in their proper places. 



DIFFERENCES DUE TO AGE. 



Aside from the natural increase in size and weight the turtles of this genus 

 exhibit some very interesting and important changes in form and sculpture as they 

 advance in age. The young of the northern species. J/, centrata, and its subspecies 

 resemble the female parent very closely, with the shell perhaps a little rounder and 

 the head proportionally a little larger. The color, however, in every case among 

 the large number I have seen, was a dark blackish-brown, the soft skin being 1 every- 

 where so thickly speckled with black as nearly to obliterate the ground color. The 

 top of the head and the lips were always dark. The carapace and plastron were more 

 or less flexible and the covering plates, while finely pitted, never showed concen- 

 tric ridges, and only occasionally the concentric markings so conspicuous in the 

 adult. As the young animal increases in size the plates of the shell are spread apart. 

 To fill the space thus left new shell is developed, underlying the original plates and 

 extendine- beyond their margins, where it appears in the first year as the first concentric 

 ridge. Presumably this growth takes place periodically, as Agassiz (1857, pp. 260, 290) 

 observed in several species of turtles, and the age of the turtle can be approximated 

 by counting the ridges. For the first six or seven years the growth appears to be 

 quite uniform, the ridges being w T ell separated and usually quite easily counted. 

 Later in life the growth is much slower and the ridges are so close together and so 

 narrow that it becomes impossible to distinguish them. The determination of the 

 age of a very large and old terrapin is rendered still more difficult by the fact that 

 the shell becomes worn off so as to obliterate all but the last-formed ridges. The 

 sexual differences in the young terrapin are apparent after the second or third year. 



The striking differences between the young and the adult of the Texas terrapin 

 are described in the paragraphs devoted to that species. Of the Florida species, I 

 have seen individuals ranging from 3A to 7 inches in length. The very young are 

 still unknown. There is, however, some indication of an increasing roughness of 

 the shell with advancing age up to the fifth or sixth year, when the inevitable wearing 

 away of the older ridges begins. 



