02 
GrevillEj on New Diatoms, 
Hah. Barbadoes deposit^ Cambridge estate ; in slides com- 
municated by C. Johnson,, Esq. 
A very neat-looking diatom^ closely allied to T. venulosum, 
of which it may possibly prove to be a variety. But in all 
the specimens I have seen there is only a single pair of short 
lines to each angle (a line on each side). 
Triceratium implicitumj n. sp., Grev. — Minute; valve 
with convex sides and very rounded angles ; surface nearly 
filled with a closely sinuate network of minutely branching 
linesj not reaching to the margin. Distance between the 
angles '0022'^ (Fig. 25.) 
Hab, Barbadoes deposit^ Cambridge estate ; in slides com- 
municated by C. Johnson, Esq.; exceedingly rare. 
This very peculiar little species may be regarded as re- 
motely allied to T. labyrinthceum, being similar in outline 
and size, and having the centre filled with a sinuous network. 
In the last-named diatom the cells are very much larger, and 
are, moreover, distinctly punctate. In our new species the 
network is minute and delicate, without puncta, and having 
a sort of prickly appearance, in consequence of very short 
lines being given off from the walls of the cellules. 
Triceratium zonatulatum, n. sp., Grev. — Minute; valve 
cruciform, the four angles with nearly parallel sides and 
broadly rounded ends ; surface with a faint, circular nucleus, 
surrounded by faint, scattered puncta ; angles nearly filled up 
with minute crowded puncta, leaving a blank, transverse space 
between them and the central puncta. Distance between 
the angles -0012^^ (Fig. 17.) 
Hab, Singapore; obtained from shell-cleanings, by Lau- 
rence Hardman, Esq. 
This, and a number of other interesting species, are pre- 
served in the cabinet of Mr. Laurence Hardman, who has 
kindly placed in my hands a large number of exquisitely 
mounted slides for examination. It is difficult to say whether 
the four-angled form of many of these Triceratia be the 
normal one. In some cases I believe that it is, and that in 
others five or six angles may also be a constant character. 
On the other hand, we know that additional angles are some- 
times mere variations, as in T. Favus, T, scitulum, &c., which 
are occasionally four-angled, and in 7\ striolatum^ as figured by 
Brightwell Mic. Journ.,^ Vol. I, PL IV), where the angles 
vary from three to five. In the little species now before us 
the prominent characters are the large portion of the angles 
filled with crowded puncta, and the transverse blank spaces 
which separate the mass of puncta referred to from the appa- 
rently somewhat depressed centre, 
