946 THE AMERICAN NATURALIST.  [VOL. XXXII. 
gastralia. In crocodiles the gastralia are only equal in number 
to the pairs of ribs in the same region of the body. The cross 
rows of epidermal scutes also equal the ribs in number, but they 
do not follow the direction taken by the gastralia, as they do 
approximately in Sphenodon. In snakes the rows of epidermal 
scales equal the ribs. In general, a study of the scales of rep- 
tiles would, I think, show that originally, at least, there has 
been some simple numerical ratio between the segments of the 
body and the number of gastralia and epidermal scales. 
Gastralia of the form described have occurred not rarely in 
the vertebrates of past ages. They occurred in Archegosaurus 
and other genera among the Stegocephali, and in the Ichthy- 
osauria, the Sauropterygia, and the Pterosauria among the 
extinct reptiles. Being so widely distributed among the early 
reptiles, some of the latter showing relationships in some 
respects with turtles, it is very probable that the ancestors of 
the latter possessed similar gastralia. Whether there were 
several of these for each pair of ribs, as was the case with the 
Stegocephali, or two, as in Sphenodon, or one, as in Ichthyo- 
saurus, we cannot tell. y 
Assuming that the plastron of turtles has had its origin in 
such gastralia, it would be interesting to know how many of 
these have been concerned in its construction. From the great 
length of the plastron in most turtles, and the slender form of 
the gastralia, we might at first suppose that many were in- 
volved ; but this, I think, would be an erroneous conclusion. 
We must recollect that turtles possess only ten dorsal and no 
lumbar vertebræ. Hence not more gastralia can be included 
than those corresponding to ten pairs of ribs ; indeed, not so 
many. 
The umbilicus in young turtles is placed where the suture 
between the hyoplastra and hypoplastra crosses the mid-line ; 
hence this suture is an approximately fixed line. It corre- 
sponds pretty closely to the suture between the fourth and the 
fifth ribs. Therefore, four pairs of ribs belong in front of it ; 
six pairs behind it. Of the four pairs in front of these sutures, 
we must, it seems likely, concede that at least two originally 
were connected with the sternum below, and hence would not 
