76 Dr. Gregory^ on the Post-tertiary 
elliptical median space. The lateral parts are marked with 
somewhat coarse transverse striae. Length about 0"0025". (290.) 
39. Amphiprora lepidoptera, n. sp. This beautiful form is 
also not unfrequent in the deposit. It is in some degree 
allied to A. alata, but not only is it much longer_, it has also 
a totally different aspect_, not being hyaline as A. alata is^ but 
conspicuous^ from the absence of this peculiarity. It has also 
a brown colour. It is frequent in some densities^ and very 
uniform in its characters. 
Form nearly rectangular^ with rounded ends^ rather narrow, 
strongly and sharply constricted in the middle. The median 
space is broader than the lateral parts or alse. There are 
well-marked nodules on each side where these parts meet and 
join the transverse terminal margins. The alse are finely, but 
distinctly marked with transverse parallel striae. Length 
from 0*004" to 0 006". Colour a clear brown, especially on 
the alae. This is a very elegant form, but being thin, it is 
often met with fractured. It is shown, of a good deal less 
than the average length, in fig. 39. (291.) 
(While this sheet was going through the press, I have 
ascertained that the form figured in the first part of this 
communication as Apr. vitrea /3 is the S. V. of Apr. 
lepidoptera. Professor Smith, in his second volume, refers it to 
Apr. elegans, from which, however, it is at once distinguished 
by the acute and apiculate extremities, as well as by its 
general aspect and finer striation.) 
40. Amphiprora recta, n. sp. This is a smaller species, 
still more nearly rectangular than the two preceding. It is 
frequent in the lighter densities. An individual, a good deal 
broader than the form usually is, is shown in fig. 40. 
Form nearly rectangular, very slightly constricted, or rather 
incurvate in the middle ; the corners rounded, the ends flat. 
Median space wider at the ends than in the middle. Lateral 
parts transversely striated ; striae fine, but distinct. Length 
from 0*002'^ to 0*0045"; breadth variable, the longest being 
always narrow. There is an appearance of a central nodule in 
each valve, besides the two terminal nodules ; and the valve 
on the S. V. must look like a Navicula or Pinnularia. But I 
have not been able, though the form is very frequent, to see 
any S. V. which I could refer to it. At first I took the frus- 
tule for the F. V. of a Navicula or Pinnularia ; but the halves 
are united as in Amphiprora ; the striation extends over the 
greater part of the F. V. ; and lastly, the valves, even when 
separated, do not lie on the S. V., as is almost invariably the 
case with those genera. I have therefore referred it to Amphi- 
prora. (292.) 
