23 
RoPER^ on Biddulphia. 
T. favus in tlie centre^ we may diverge in lines ending 
amongst others in one like Zygoceros rhombus, especially in 
the front view_, and in another analogous to Amphitetras 
antediluviana and Meneghini^ in his remarks on Tricera- 
tium^ states that ^^the perfect resemblance of the primary 
surfaces_, and the large apertures at the three processes in 
the secondary^ render this genus precisely intermediate be- 
tween Amphitetras and Zygoceros/^"^ and in this opinion I 
fully concur. 
Professor Smithy in his divisions of the order^ has formed 
his first sub-tribe out of those genera^ with a deciduous con- 
necting membrane,, which he believed to be free or adherent^ 
and excluding Triceratium^ of which some species have been 
shown by Colonel Baddeley's gatherings to be filamentous. 
Eupodiscus radiatus, and probably E. sculptus, the remainder 
of this tribe^ might probably be well grouped together. The 
species of Triceratium^ however, he states had only been met 
with as isolated specimens^ and the points which now show 
clearly their affinity with Biddulphia were then unknown to 
him^ or I hardly think he would have included them in his 
first sub-tribe^ whilst he places the latter genus in the fourth. 
If we take the genera Amphitetras^ Triceratium^ and Bid- 
dulphia, we find them all formed on one common plan, com- 
posed of two valves, as in the annexed figures, 1 a and 2 a, of 
various outline, but all having two or more processes, and 
with a few exceptions two or more spines ; and these, as in 
all others of the class, united by a connecting membrane, but 
frequently of much greater breadth and persistance of 
character than in the other sub-tribes. In these characters 
they differ from all the other genera included by Smith in 
the ^Synopsis,^ and if the forms with five or six angles, probably 
included by Ehrenberg under the name of Amphipentas, be 
added, we find, as in the annexed diagram, that a complete 
series may be formed, of which Triceratium is the connecting 
link between the orbicular forms of Biddulphia and the many- 
angled valves of Amphipentas. 
The only other genera noticed by Ehrenberg or Kiitzing 
that appear to be allied to Biddulphia, are Terpsinoe and 
Hemiaulus. If to the former the Zygoceros bal(Ena and Zyg. 
australis, Ehr., were added, we should have a genus distin- 
guished from Biddulphia by the absence of the angular 
processes, but with frustules formed very much on the same 
plan, and separable into two groups of undulating and non- 
undulating valves, as in that genus, and it should probably 
be placed between Biddulphia and Isthmia, instead of being 
* Ray Soc, 1853, p. 487. 
