46 
Dr. Gregory, on the Post-tertiary 
equally well known as a marine form.* And I have also 
observed, that N. elliptica^ which varies remarkably in all 
obvious characters, sometimes acquires a nearly, if not a per- 
fectly costate striation, though usually strongly moniliform. 
As I propose soon to lay these observations before the 
Society, I shall not here go farther into the subject. 
17. Navicula clavata^ n. sp. This very fine form, represented 
in fig. 17, has at first sight some resemblance to N. Hennedii ; 
but on close inspection, it presents remarkable characters. 
Form elliptic, broad, witb broad rounded projecting masses 
at the apices, which are the extremities of the median line. 
Striation marginal, as in N. Hennedii, but the inner bounding 
line of the striated band, instead of being purely elliptic, as in 
that form, becomes towards the extremities, nearly straight, 
so as to form a kind of angle, giving to the included blank 
space between it and the median line, a very remarkable form. 
Median line complex. First there is in the middle, as in 
N. Hennedii, a narrow line proceeding from each end, and 
terminating on each side of the centre, and at a short distance 
from it, in long rounded expansions ; the other extremities 
are also rounded, but larger. Between the two central knobs 
lies a rectangular white space, extending in its length at right 
angles to the median line, and rather narrow. It reaches 
beyond the general width of the middle part, that is, the 
striated portion now to be mentioned, expands at the middle. 
On each side of the proper median line is a transversely 
striated band, which, near the ends, touches the median line, 
but near the middle, recedes a little from it on both sides. 
The striated band expands into large round heads, projecting 
beyond the true elliptical outline of the valve, and it also ex- 
pands a little in the middle. The white blank across the centre 
appears to have at each end a small striated patch placed trans- 
versely to it. The large swollen ends of the complex median 
line, not only project, forming short snouts, but stand out 
strongly from the surface of the valve. The striae appear rather 
coarser than those of N. Hennedii, about 20 in '001", and are 
very distinctly moniliform. Length of the valve, 0'0034". 
I may here mention that Dr. Greville has found in the 
same Trinidad sand which 1 have alluded to elsewhere in this 
* I observe that in Vol. 11. of the Synopsis, Professor Smith gives, as 
N. elUptica, Kiitz. var. P,the form which I found in Lochleven, and which 
resembles N. didyma. I admit that it seems to be a variety of N. elliptica^ 
Kiitz., but I cannot find any essential difference between it and certain 
forms of N. dydyma. Is it possible that N. elliptica, Kiitz. may take the 
form of N. dydyma in sea water, and that some other local cause may have 
produced the same modification in the fresh water of Lochleven ? 
