Identity of Navicula crassinervis, &e. By W.H. Dallinger. 175 
everywhere manifest, where from their extreme minuteness and 
multitude, large numbers of microscopic organisms can be con- 
stantly compared. 
It is pre-eminently true, for example, of the desmids and the 
diatoms. Of course I need not remark that the mere study of 
mounted specimens is utterly incompetent for proof in such an 
investigation. One might as well attempt to generalize upon the 
osteological characteristics of a given skeleton by the observation 
of a limited number of specimens seen through the glass fronts of 
museum cases. It can only be done, as the Fellows of this Society 
well know, by the careful examination of unmounted material in 
every variety of condition. 
I have already said that my knowledge of the Diatomaceze does 
not extend much beyond the silicious frustules ; whilst I have spent 
comparatively little time in endeavouring to accomplish difficult 
feats with test-objects. But a working microscopist must have a 
set of crucial standards, thoroughly known to himself, to serve for 
comparison ; and N. rhomboides in its several varieties was very 
early adopted as mine: hence I have had an interest in this frus- 
tule which has extended over years ; and especially since the pro- 
minence given to “F', Saxonica” and “N. crassinervis” I have 
examined very large quantities of material derived from both 
England and America. It is my habit to make a drawing of any 
fact of moment in any of my observations; and I invariably 
do this in relation to variations of form in any well-known object 
or objects. Since reading the rule given by Mr. Ingpen, on which 
the determination of genera and species is said to depend, I have 
turned over the leaves of my folio, and have found some singularly 
suitable cases of variation in the “shape” of N. rhomboides (of 
course including F. Saw. and N. crass.), which are of considerable 
service; all the more indeed that some of them have been drawn 
for years, and all were drawn in entire independence of the question 
in hand. 
In Fig. 1, Plate CLXXVI., are the outlines of four frustules 
taken from a gathering of “ Cherryfield” rhomboides. They were 
of course drawn at different times, and from different “dips” of the 
material, and are doubtless not commonly to be found ; but whoever 
is supplied with sufficient material and patience might, I am con- 
vinced, select and arrange frustules precisely in the order in which 
they stand in the Plate. Apart then from the general symmetry, 
if the ends of the frustules drawn be examined, it will be seen that 
between A and D there is a remarkable difference; but if B and C 
be inserted in order between them, we have comparatively easy 
steps from the one to the other. 
Now the only diatom that I know of that in all respects comes 
near to A and B is Navicula cuspidata; but (1) as a rule the 
02 
