498 PROFESSOR GREGORY ON 
lations. S.V. linear, narrow, broadest at a point above the middle, from which it 
becomes narrower both ways; the shorter half being rather broader than the other, 
and rounded at the apex. The longer and narrower half is also rounded, and 
very slightly expanded at the end. Strive coarse, not reaching the median line, 
but leaving a somewhat broad raphe in the middle. Strize about 16 in 0-001’; 
Length about 0:0015"; breadth of F.V., at larger end, 0:0004’, at smaller end. 
0-0003". Breadth of 8.V. 0-0002” at the broadest point. 
This form resembles both a Meridion and a Gomphonema. The absence of a 
central nodule prevents me from referring it to the latter genus; and with reference 
to the former, the mode of attachment, as well as the form of the frustules, agree 
pretty well with it. But I have not seen more than four attached; so that it is 
still doubtful whether it forms a spiral filament, as the slightly cuneate frustules 
must tend to do. It occurs, by no means sparingly, both in Lamlash Bay and in 
Loch Fine. Future observations on the living form will decide the question of 
its generic position, but in the meantime it is a well-marked species. 
42. Pyxidicula cruciata, Ehr. Pl. X., fig. 42. Form of V. cup-shaped, hemi- 
spherical, with hexagonal cells over the whole surface, and in one direction, a 
crest composed of square or irregular cells running round the hemisphere in a 
plane at right angles to that of the junction of the two valves. Cells large, and 
of uniform size. Diameter 0:0019’. 
The form here figured is only a detached valve, the entire frustule not having 
yet occurred in these dredgings. But it agrees precisely with EnrenBERG’s figure 
of the valve. The specific name is given on account of the arrangement of some 
of the cells, as seen in a view obtained by looking down on the frustule, at right 
angles to the junction, in the form of a broad rectangular cross. This cannot be 
seen in the view here figured. The species is very scarce as yet, having occurred 
very sparingly in Lamlash Bay. But it is interesting, as being one of the forms 
which Enrensere has figured from the Aigina Clay Marl already mentioned at 
p- 482, and from the deposit of Richmond, Virginia. I have placed it in this 
group, because I have reason to think that Pyxidicula is a catenate form, 
though we cannot expect to see this in fossil deposits. I would here refer to 
the beautiful new form, detected by Professor WaLker Arnott, and figured in 
the Appendix to this paper, which, in the frustules, is so closely allied to the 
present species, that it may prove to be actually P. appendiculate of EHRENBERG, 
a form which occurs along with P. cruciata. Professor W. ArNnotr’s form is 
decidedly catenate. 
43. Orthosira angulata, n. sp. Pl. X., figs. 43 and 43 6. Form of F.V., which 
occurs solitary, and in chains of from two to six, rectangular in the middle, acu- 
minate and truncate at the ends. On the margin, which is often slightly incurved, 
are seen denticulations arising from the cells of the disc or S.V. Length from 
0:0005’ to 00015"; breadth of F.V. from 0-0003” to 0:00045’. Diameter of disc, 
