246 WALILICH, ON TRICURATIUM. 
Professor Bailey states that this form is allied to 7. spino- 
sum, which has been found in the fossil state in Virginia ; 
but the same reasons that s separates the St. Helena species 
from the Triceratia, of course apply equally to this. 
In the valuable paper on Triceraiium contributed to the 
‘Journal of Microscopical Science’ for July, 1856, by Mr. 
Brightwell, three forms demand notice as being at first sight 
allied to the one under discussion, namely, 7. sedigerum of 
Bailey, T. orbiculatum of Mr. Shadbolt, and T. formosum of 
Brightwell, the last being the 7. armatum of Mr. Roper. 
Before noticing Mr. Brightwell’s characters, 1 would draw 
your attention to a remark he makes on Professor Bailey 
having referred a four-sided form of Triceratium to the genus 
Amphitetr as, namely, that “the projection of a connecting 
membrane beyond the suture of the valve, which is one of 
the characters of Amphitetras, is not seen in these square 
forms.” A remark, if strictly accurate, at once fatal to any 
alliance between the St. Helena Diatom and those four-sided 
varieties, referred by Mr. Brightwell and others to the three- 
sided typical form. 
In a former paper on the same genus, in the ‘ Journal’ for 
July, 1853, Mr. Brightwell gives figures of three-, four-, and 
five-sided varieties of 7. striolatum, a name he alters in the 
recently published memoir into 7. formosum, already alluded 
to. Now in none of these does the connecting membrane 
project in the slightest degree. The colour of the frustule is 
moreover pale brown, indicative of very minute cellular struc- 
ture, whilst the horns are simple projections, and no spines 
exist on the surface of the valves; and lastly it is very much 
smaller than 7”. serratum. 
Amongst Mr. Brightwell’s species, 7. armatum comes 
nearer to the St. Helena form than any of the others. Mr. 
Roper, to whom we are indebted for a description of this 
species, thus characterises it in the ‘ Microscopical Journal’ 
for July, 1854: 
“ Frustules large, with straight or slightly convex sides. 
Angles produced into horn-like processes, with rounded ex- 
tremities ; cellular structure minute, partially radiating to- 
wards the sides and angles; six or more spurious processes 
projecting from the surface of the valve.” Mr. Roper’s 
specimens are described as approaching closely to 7. tri- 
daciylon, of Ehrenberg, a form also figured by Mr. Brightwell. 
But he states that 7. armatwn is deficient in the siliceous 
plate that is shown to exist around the sides of T. tridactylon. 
Mr. Roper thinks his specimens are not identical with Pro- 
fessor Bailey’s T. spinosum, whereas the spines are shown, 
