of Diaiomaceous Forms. 
13 
resembles a shortened P. ohlonga ; that which occurs in Dud- 
dingston Loch ; the subcapitate forms ; the capitate forms 
with straight sides ; that having the form of P. acuta ; and 
various others. 
If these two classes of forms, the known and the unde- 
scribed, really constitute one group, we find in that group 
nearly every shape which is seen in the genera Navicula and 
Pinnularia ; and also transition forms, connecting together the 
various types. 
The question naturally arises : Can all these varied forms 
belong to one species ? Now, at one time, certainly, each 
marked type of form would have been regarded as a distinct 
species. But the more extended observations of recent times 
have proved that form, shape, or outline, is not nearly so 
permanent a character as had been imagined. In a paper on the 
Mull Deposit (' Journal,' January 1854,) I pointed out, and 
illustrated by some figures, the remarkable tendency to variety 
of form in three species, namely, Eunotia bigibba, Kiitz : 
Pinnularia divergejis, W. Smith, and Himantidium hidens. 
I alluded also to the same tendency in Eunotia triodon ; and I 
again returned to the same point in this last species, in a 
short paper in the ' Journal' for July, 1854. Other examples 
are not wanting ; and the more the Diatomaceae are studied, 
the more do we perceive that, in many species at least, the 
shape or outline is subject to endless variations. It certainly 
appears at present as if, in many species, the form were constant : 
but we must be cautious in affirming this, for in two species 
which I adduced as examples of constancy in outline, namely, 
Navicula rhomhoides, and N. serians, we have now good reason 
to believe that important variations of shape occur. Just as 
N. peregrina seems to belong to the group of N. varians, 
so it appears that N. Crassinervia will prove to belong to N. 
rhomboides ; and that a form, apparently yet more widely 
differing from the latter, namely, that which I have lately 
described* under the name of N. interrupta, which is linear, 
narrow and obtuse, may be found to be another modification 
of N. rhomboides. The Revd. Professor Smith has also very 
recently detected a modification of N. serians^ most remarkably 
different in shape from the usual type. 
It will probably be found necessary, looking to the 
uniformity of markings and aspect in the forms here described, 
and to the existence of such numbers of transition forms con- 
necting the various types of outline, to form a species, 
Navicula varians, including these forms as sub-species ; or 
* In a paper read to the Microscopical Society, 25th of October last, 
which will appear in the next Number of the ' Journal.' 
