804 REPORTS ON THE STATE OF SCIENCE.—1913. 
where the thecal shell was incomplete (compare fig. 9). This lateral 
row soon reached a strong development, connecting the costals with the 
plastron, and thus completing the thecal shell by a row of epithecal 
elements, the marginals.*° The other epithecal elements disappeared 
in the living thecophorous Testudinata (perhaps a few have been 
retained in Graptemys). In the ancestors of the Leatherback, how- 
Fig. 9. Hypothetical Prochelonian, dorsal view, to show the primitive state of the testudi- 
nate shell. Thecal ossifications are dotted, epithecal elements black; only the 
larger epithecal ossifications have been indicated ; between them were probably 
present numerous very small, irregularly distributed epithecal elements, perhaps 
represented by horny scutes only; the lateral epithecal ossifications (of the 
marginal and supramarginal row) are strongly developed, as they cover the 
space between the dorsal thecal ossifications and the plastron. For the identifi- 
cation of the rows of epithecal elements compare fig. 8 ;_ the rows are represented 
as continuous on the neck (as in the just born Dermochelys) and on the tail (as 
in Chelydra). 
ever, these epithecal elements, instead of being more and more 
reduced, formed a new epithecal shell. We may assume that these 
ancestors reducing their heavy thecal shell and horny scutes in adapta- 
tion to a swimming life in the open sea, at the same time in some 
40 Sometimes the marginals are less firmly united with the costals, as these with 
tre lige The same interpretation of the marginals is brought forward by Jaekel 
A 
