506 EMBRYOLOGY OF THE TURTLE. Part JIL. 
lines, thus giving the field of view a striated appearance. As seen with a mag- 
nifying power of five hundred diameters, it is impracticable to represent these 
minute strix2 with the pencil, except as the faintest lines possible. It will be 
noticed that the granules of the nodular fibres are not so large as those of the 
more interior layers; but this dimimution is not constant, as will soon be seen. 
In some species the different layers are very distinct from each other, (PI. 
9b, fig. 6e,) and keep their components so closely within themselves, that the 
passage from the one to the other seems almost an interval. This is particularly 
noticeable in Platypeltis ferox. Again, in others the irregularities of each layer fit 
into those of its neighbor; so that it is with difficulty that the respective boun- 
daries of one or the other layer can be recognized. 
As we advance outwardly, we do not find that the fibrous arrangement 
becomes regularly more and more apparent; but, on the contrary, here and there 
may be seen a layer, or rather two or three successive layers, composed of sep- 
arate granular bodies, oftentimes much broader and coarser than the delicate fibres 
(Pl. 9b, fig. 6f 4) of the strata which lie on both sides of them, and still display- 
ing their tendency to trend in particular directions (Pl. 9b, fig. 6f, @) in their 
respective layers. It is a very easy matter to peel off these coatings one from 
the other, and view them separately; yet, where three or four are superposed, there 
may be sufficient light transmitted to study them as they are naturally related. 
However, with these breaks in the continuity there occurs a pretty regular oblit- 
eration of the nodular appearance of the fibres, their components becoming grad- 
ually more and more intimately united to each other, as they are situated suc- 
cessively nearer and nearer to the outer surface of the shell membrane, till finally 
each fibre has become uniform and apparently structureless throughout its length. 
The outermost of these layers, next to the hard calcareous deposit, are composed 
of the smoothest and most uniform fibres, (Pl. 9a, fig. 45, and Pl. 9b, fig. 6g,) 
resembling at times excessively elongated tabular crystals. Before the shell is 
deposited, these layers may be recognized by the peculiarly brilliant nacreous appear- 
ance which strikes the eye. In Glyptemys insculpta, where this has been noticed 
most frequently, the component fibres are of excessive tenuity and compactness 
among each other, the latter feature tending, no doubt, to heighten the polished 
aspect of the surface of the layer. 
The edge of a section made through the whole thickness of the shell mem- 
brane (Pl. 9a, fig. 43, ¢, d) appears more or less rough and dotted at intervals, 
where the ends of the fibres have been cut across at various obliquities; but 
between these the length of the threads may be recognized, and the layers dis- 
tinguished, with more or less certainty, according to the species. 
A few words are necessary in regard to the nature and origin of the gran- 
