72 
Dr. Gregory_, on the Post-tertiary 
34. Amphora obtusa, n. sp. This is one of the most 
striking forms in the deposit, in which the detached valve, as 
seen in fig. 34"^, is by no means rare ; while the entire frus- 
tule, fig. 34, is very scarce. The detached valve has also oc- 
curred to Dr. Greville in the Trinidad sand above alluded to. 
Form of the entire frustule nearly rectangular, broad, the 
ends very broadly rounded. The outline of the detached valve 
is generally a little convex on the outer margin, occasionally 
it is very slightly incurved there, as in fig. 34"^. The ends of 
the valve are generally very obtuse, but, as in fig. 34"^, some- 
times produced to a sort of beak. The inner margin of the 
valve is gently undulated, so that it was not easy to suppose 
two of them in opposition till the entire frustule, fig. 34, oc- 
curred, in which the margin appears more nearly straight. 
The two valves are in close opposition, except at the centre 
and just at the ends ; but I have also seen them separated by 
a narrow rectangular space. The curve lines are strongly 
marked and strongly curved, projecting in the middle nearly 
to the centre, and near the ends almost coalescing with the 
outer margin. There is a nodule at the middle point of each 
curve line, and also near each end of the valve ; but here the 
outer margin seems to bend inward in a curl and form the 
nodule, which is apparently on a diff'erent plane from the rest 
of the inner curve line. Length about 0-0037'' to 0'004". 
The valve is transversely striated. Strise very fine, about 70 
in O'OOl". They cannot be seen on all parts of the valve at 
the same time, because the valve is very thick, and the parts 
on opposite sides of the curve line are in different planes. 
(287.) 
35. Amphora crassa, n. sp. This and the two following 
species possess a most remarkable and peculiar structure, 
which makes me very doubtful whether they belong to this 
genus. If they do, they will form a well-marked sub-genus. 
It is, however, possible that it may be found necessary to 
form a new genus for their reception. The first of them, 
A. crassa, is represented in fig. 35. 
Form nearly rectangular, rather narrow ; corners rounded, 
ends truncate. Length from 0'002" to 0-0032". The whole 
visible surface of the valve is divided into about eight vertical 
bars, which seem to converge on the ends. The curve lines 
project half way across each valve in the middle, but retreat 
rather suddenly to very near the outer margin, and run along 
it, almost coalescing with it. The vertical bars are trans- 
versely striated, and the striae are coarse and broad, about the 
same width as those of Pinnularia borealis, Ehr. (P. latestriata, 
W. G.) There is no appearance of separation between the 
