2 
Greville_, on New Diatoms. 
is that genus at present defined. In both sections, as it 
stands (including both Pz/<:*7i<iicM/« and Dictyopyxis of Ehren- 
berg), there are^, according to my view, species bearing- 
little^ if any, generic affinity ; and, .as the minute fossil 
diatoms I am about to describe, possess a striking character 
of their own, I prefer to keep them apart, rather than add to 
the existing uncertainty and confusion, which, after all, is 
mainly owing, as my friend Mr. Ralfs has remarked_, to vari- 
ous so-called species being still little known. 
Pyxilla Johnsoniana, n. sp., Grev. — Frustule cylindrical- 
oval, simple (no contraction at the suture). (Fig. 6.) 
Hab. Barbadoes deposit, Cambridge estate; in slides 
communicated by C. Johnson, Esq. ; very rare. 
This rare species has only occurred to me twice, but it is 
highly probable that in some other samples of the deposit, 
both it and the following may be more frequent. The 
cellulation is so minute as to be correctly defined as punc- 
tate; but the cellules, when sufficiently magnified, appear 
to be regularly hexagonal. The suture is situated at some- 
what more than one third of the total length from one 
extremity. Length of frustule "0025^^ 
Pyxilla Barbadensis, n. sp., Grev. — Frustule contracted 
at the suture, one valve cylindrical, the other globose. 
(Pig. 5.) 
Hab. Barbadoes deposit, Cambridge estate ; in slides 
communicated by C. Johnson, Esq. 
A most beautiful diatom. The globose valve with its 
contracted base and terminal apiculus, resembles the bulbous 
dome which crowns the minaret of an eastern mosque. 
Although there is a great contraction between the valves, 
the suture itself is acute and somewhat prominent, as in 
some of the Creswellice. The pu notation is as minute as in 
the preceding species_, and^ under a high power, comes out 
equally beautiful as hexagonal reticulation. Length '0030"'. 
Cresswellia. 
Cresswellia Palmeriana, n. sp., Grev. — Very large; frus- 
tules in front view short, cylindrical, with truncate ends; 
connecting processes numerous, truncate; cellules puncti- 
form at the suture, becoming larger and hexagonal towards 
the ends. (Fig. 9.) 
Hab. Hong Kong harbour_, John Linton Palmer, Esq. ; 
Shark^s Bay, Australia, in stomachs of Ascidians ; Dr. 
Macdonald. 
