4 
RoPER^ on Biddulphia. 
genus Denticeiia, but wliich he considered nonconcatenate, in 
tlie same way tliat lie had already separated Zygoceros from 
Biddulphia; but as hitherto both Biddulphia turgida and 
rhombus^ have generally been obtained from mud deposits, and 
after boiling in acid_, it is highly probable they will both be 
found, when met with in abundance in a living state, to be 
filamentous forms, and therefore not entitled to any other 
than specific distinction. 
Having thus sketched the origin of the various genera 
into which the genus Biddulphia has been divided by Ehren- 
berg, I shall now proceed to Kiitzing, the most systematic of 
the German writers. In the arrangement of the family he 
chiefly follows that author, but, as I have already mentioned, 
retaining the name of Odontella of Agardh, for those species 
included in the genus Denticella; but he ignores entirely 
the Biddulphia pulchella of former authors, and separates 
it into three distinct species, naming them tri-locidaris, 
qidnque-locularis , and septem-locularis, the difference being 
merely the number of lobes into which the valves are 
divided, a character utterly unworthy of notice as afibrding 
specific distinction, as no gathering can be made of 
B. pulchella, when it occurs in any abundance, without 
meeting with frustules divided by a diflPerent number of 
costae, varying from three to seven. 
From a careful examination of the various papers referring 
to this genus, I find it has at various times been described 
under the following generic names : 
1. Conferva, 
2. Diatoma, 
3. Biddulphia, 
4. Odontella, 
5. Denticella, 
6. Zygoceros, 
7. Cerataulus, 
8. IslLmia, 
9. Triceratium 
th 
species. 
58 
Rejecting altogether some of the species described by 
Ehrenberg from small fragments, and of which it is im- 
possible to form the least opinion, either from his descriptions 
or figures, I believe these may be reduced to about eleven good 
species, and together with two hitherto undescribed, as far as I 
know J include all the well-established forms that can be 
separated by good specific characters. 
I have already alluded to the great variation that exists in 
The Cerataulus turg'idus and Zygoceros rhombus of Ehrenberc 
