522 PROFESSOR GREGORY ON 
teral segments arcuate, ends acute, inner curve-lines strongly curved, nodule 
elongated into a transverse bar. The converging bars in the middle are numerous, 
in apposition, and the whole frustule very hyaline. It is marked by fine trans- 
verse strize, easily seen on the lateral segments, and which may also be traced, by 
careful manipulation, across the whole frustule. Strize about 40 in 0-001’. 
I first observed this fine and conspicuous form in the Glenshira Sand, but 
postponed the description of it that I might examine it farther. I have found it 
more frequently in Lamlash Bay and in Loch Fine. 
88. Amphora Arcus, Greg. PI. XIII., fig. 88. Form of entire frustule barrel- 
shaped, ends truncate. Segments arcuate, sub-apiculate, marked by coarse moni- 
liform strize, which are also seen over the entire form. Length from 0-0035" to 
00045"; breadth about 0:002." Strize 16 or 18 in 0:001". The longitudinal bars 
in the middle between the lateral segments are about 16 or 17, closely set, and we 
can see that they are segments seen edgewise. 
This fine form was rather frequent in the Glenshira Sand, but was only then 
known in the shape of detached segments, two of which I figured in my first paper 
on that deposit (J/ic. Jour., vol. iii., pl. iv., fig. 4), without at that time under- 
standing its real structure. Subsequently I recognised the entire frustule in the 
Glenshira Sand, and also in the dredgings from Lamlash Bay. In my third paper 
I figured an imperfect specimen (7Z'rans. Mic. Soc., vol. v., pl.i., fig. 37), and [now 
give a more perfect one, which shows that it really belongs to the complex group. 
See fig. 88. 
89. Amphora Grevilliana, Greg. Pl. XIIL, fig. 89. Form of entire frustule 
nearly oval, broad, slightly truncate at the extremities, which are broad, the line 
joining the apices of the lateral segments being concave, so that a short process 
stands out at each side. Length about 0-006’; breadth about 0-002”. In the spe- 
cimen figured (fig. 89) the form of the lateral segments is perfectly seen, probably 
because the frustule is viewed from the flat side, as we view an orange cut in 
half from the cut side. These segments are precisely like that figured in my last 
plate from the Glenshira Sand (7’rans. Mic. Soc., vol. v., pl. i., fig. 36*). They 
are broad, arcuate, with somewhat acute rostra, and three curve-lines on their 
surface, which are peculiar to this species. They are marked with strong trans- 
verse moniliform strize. Strize about 28 in 0-001". Between the lateral seements 
are five or six convergent bars, which are the backs of other segments, and which 
are in apposition, except at the ends, where they diverge a little, from their 
apices becoming suddenly narrower. 
This fine species occurs both in the Glenshira Sand and Lamlash Bay. But 
the entire frustule given under this name in the plate just referred to (at fig. 36) 
does not belong to this species, as may be seen at once by comparing it with fig. 
89, and by the form of its segments, as seen in that figure. The segment in that 
plate, fig. 36*, is correct, as above stated. 
