TESTUDO AREOLATA. 
any specimen which would justify the description given by Lacepede, “ les 
lames sont agreablement variees de noir, de blanc, de pourpre, de verdktre, et 
de jaune ” The prevailing colours however are brown, and greenish yellow; 
and in the specimen from which the present illustrations were taken, both the 
green and the bay were brighter and paler than in any other I have seen. 
The sternum is sometimes uniformly yellowish white; but in most there is a 
considerable portion of light brown or bay in each sternal scutum, which 
sometimes pervades the greater part of its surface. But it is not in the colours 
only that this species varies so remarkably; in the form and height of the shell, 
in the degree of revolution of the margin, and especially in the number of 
dorsal and marginal plates, it is also liable to some striking accidental 
deviations. In the latter respect, the variation has sometimes been assumed as 
a specific character, as in that of T. Cctfra of Daudin, “ margine loricm dorsalis 
scutellis 27This must have arisen from ignorance of the general uniformity 
of the law which assigns to all the Testudinata with plates, the same number 
of dorsal scuta, namely thirteen; a rule to which there is but a solitary excep¬ 
tion in the case of one species of marine turtle. The additional plates which 
thus accidentally occur, always appear as if severed from one of the normal 
ones. I have a specimen in which there are two additional costal scuta, one 
being introduced on each side between the fourth costal and the fifth vertebral. 
In another, a small square plate is placed between the fourth and fifth ver¬ 
tebral. In a third, there are ten marginal scuta on one side, and eleven on the 
other; a fourth has twenty-six marginal plates, and a fifth but twenty-two. 
All these varieties occur in seven specimens now before me, which have come 
into my possession promiscuously and without selection. 
I cannot agree with those authors who have referred Seba’s figure (Tom. 1 . 
tab. lxxx. fig. 6.) to this species. It appears to me much more to resemble 
the young state of T. tabulata , to which also the colour, a light yellow, more 
truly appertains. Nor does it answer to the character of T. pusilla of Linnseus, 
as given both by him and by Schneider. The only tolerable figure with which 
I am acquainted is that of Schospft, which however has the distinct marking' 
and the hard abrupt outline of colour which usually belong only to dried 
specimens. 
