24 THE AMERICAN NATURALIST. 
on the comparative anatomy of vertebrates, repeatedly empha- 
sizes the idea of the independence of scutes and bony plates. 
The conclusions to which the present studies lead do not favor 
this view, but lend support to the opinion expressed in cautious 
terms by Goette.(99, p. 430), and more radically by Gadow 
(99), to the effect that in primitive turtles each bony plate was 
associated with a single scute. Supposing such relations to 
have existed, it is easy to conceive how the present conditions 
could have been brought about ; for, if the migration of the 
ectoderm were to be so retarded as to take place after its divi- 
sion into scutes, the posterior scutes would be carried away 
from the bony plates te which they belonged, and in con- 
sequence of crowding some of them might be suppressed, with 
the result that a carapace with a given number of bony plates 
would be covered by a smaller number of scutes. Although 
such an explanation of the present condition of the chelonian 
carapace must be tested by experiment, the evidence derived 
from the study of the abnormal specimens described above 
shows that there is a closer relation between bony plates and 
scutes than has been generally admitted heretofore. 
REFERENCES. 
Gapow, H. : 
'99. Orthogenetic Variations in the Shells of Chelonia.. Zoölogical 
Results based on Material collected by A. Willey. Pt. iii, 
Pp. 207-222, Pls. XXIV-XXV. : 
GEGENBAUR, C. ; 
98. Vergleichende Anatomie der Wirbelthiere. Erster Band. xiv + 
978 pp. Leipzig, Engelmann. 
GOETTE, A. 
'99. Ueber die Entwicklung des knéchernen Rückenschildes (Carapax) 
der Schildkröten. Zzi;. J. "iss. Zool. Bd. Ixvi, pp. 407-434, 
Taf. XXVII-XXIX. | 
HARRISON, R. G. 
'98. The Growth and Regeneration of the Tail of the Frog Larva. 
dick. F Entwickelungsmechanik d. Organismen. Bd. vii, 
PP. 431-485. i 
