6 
RoPERj on Biddulphia. 
the markings are by no means constant,, and^ in fact^ no 
dependence whatever can be placed on its size or structure 
alone as a ground of specific character. 
On what, then, are the species to be founded ? I believe 
that the only safe and constant characters which can be 
depended upon are as follows : the structure and areolation 
of the separated valves ; the position^ not the number^ of the 
spines ; the size and form of the processes^ and their position 
on the surface of the valve. 
Considering the known species with these points in view, 
we find them fall into two very distinct groups. The first, 
having B. pulchella as the type, with more or less elliptical 
lobed valves, having undulating margins, distinct and rounded 
areolations,"^ with the spines rising from the summits of the 
lobes, and processes distinctly areolate, placed at the extremi- 
ties of the valves ; including three species. 
The other, having B. aurita as the type, with valves more 
or less lanceolate, elliptical, or orbicular, without undulations 
at the margin. The areolation generally indistinct, but 
when otherwise hexagonal. The spines sometimes central, 
sometimes submarginal; and the processes only partially 
areolate, varying considerably in size and length, and some- 
times placed at some little distance from the margin of the 
valve. This division includes ten species. 
I now proceed to describe the species, slightly altering the 
generic character as proposed by Professor Smith. The 
specific characters, where not inserted, are to be considered 
the same as in the ' Synopsis.' 
Genus Biddulphia, Gray. 
Frustules compressed or cylindrical, adhering more or 
less perfectly into a continuous or zigzag filament ; valves 
convex, elliptical, lanceolate, or orbicular, usually spinous and 
areolate; ar eolations rounded or hexagonal, with horn-like 
processes rising from the angles or margin of the valve. 
* Professor Smith applies the term " cellules " to these markings (see 
Introduction to 'Synopsis/ vol. i, p. xvii; and again, vol. ii, p. xix). He 
has evidently adopted this designation after much thought, and a careful 
comparison with other forms of cell-membrane, but the distinctive physio- 
logical character of a cell, that of being a completely closed vesicle, appears 
to me to be entirely wanting, or at least not at present shown to exist, even 
in those species where the markings are most obvious, such as Triceratium 
favus and Biddulphia pulchella, &c. I consider that the reticulated struc- 
ture of the siliceous epiderm is merely adapted to strengthen the cell-wall, 
and therefore prefer the use of the terms, areolation or reticulation, to dis- 
tinguish those markings on the valves which are usually described as cells 
or cellules. 
