GrREViLLEj 071 New Dlatoms. 
19 
occupy a space of about half the distance from the nodule to 
the margin. The only species which appears by description 
to approach this diatom is N. costata ; but on consulting the 
figure given by Kiitzing ('Bacill./ plate iii^ fig. 56)^ it is 
evident that there is no connection whatever between them. 
The latter has no extra-median lines, and although the valve 
is said to be " longitudinaliter punctato-costata/^ the punc- 
tate character arises, not from puncta in the direction of the 
transverse striation^ but from circular puncta arranged at in- 
tervals along either one or two longitudinal lines. The 
remarkable intra-marginal line of puncta so conspicuous in 
N, luxuriosa is wholly wanting. 
Navicula? Cistella, n. sp., Grev. — Valve quadrangular, 
about twice as long as broad, the angles rounded and some- 
what dilated ; surface marked with delicately punctate, longi- 
tudinal lines ; transverse stride very fine, parallel. Length_, 
•0015" to -0025". (Figs. 12—14.) 
Cocconeis? quadrata ; Roper, MS. 
Hab. Dredged off Lyme Regis, in eight fathoms^ water by 
the Rev. J. Guillemard_, 1855 ; F. C. S. Roper, Esq. Harvey 
Bay, Queensland, in a dredging communicated by Dr. Ro- 
berts, of Sydney ; not unfrequent. 
It is not a little interesting that after I had described this 
ambiguous, minute diatom from the antipodes, I should be 
informed by my most obliging friend, Mr. Roper, that it was 
actually discovered some years ago on our own shores. The 
very accurate drawings made by himself in 1855, and now 
lying before me, leave no doubt to be entertained on this 
point. But the real nature of the frustule seems to have been 
doubtful from the first. Mr. Roper regarded it as possibly a 
Cocconeis. The late Professor Smith, to whom specimens 
were submitted, could make nothing of it, and would not 
venture to refer it to any genus. In now publishing it as a 
doubtful Navicula J my object is to attract such attention to- 
wards it as may lead to its generic settlement. Looking at the 
side view, it is certainly as much like a Cocconeis as a Navicula; 
but the front view, which Mr. Roper had not seen, seems to 
point more towards the latter. After all, that acute diatomist 
may be correct in his conjecture {in litt.), that it may belong 
to an unknown filamentous form. And, indeed, before I had 
communicated with him, I had myself remarked in my MS., 
— " So little, indeed, does the frustule resemble a Navicula, 
that in hastily passing the slide across the field of view it 
might be taken to belong to one of the Melosirece^ It must 
be understood, therefore, that its present position is simply 
