.• ••» 
136 Wallich, on the Diatom-valve. 
in internal area^ of the two valves, arising from this source. 
In Isthmia, we also sometimes observe great variation as to 
length and breadth of frustules growing upon the same object. 
But it will be found that these marked differences do not 
occur in the same filament,, but on separate ones ; and that 
the primary frustule, or rather that valve of it by which 
epiphytic adhesion was first secured, frequently does not 
exceed in size the smallest of any of the other neighbouring 
terminal frustules. 
After similarly extended observation of the compressed 
filamentous genera, I have never found any ground to alter 
my views respecting the determinate period during which 
the siliceous value may be said to increase ; and, as an 
example in point, and one which, for several reasons, is 
amongst those best fitted to test its correctness, I would 
mention having carefully measured, by means of Ross's screw 
micrometer, frustules of the three species of Rhabdonema, 
at intervals, in filaments numbering as many as a hundred 
individuals, without the discovery of any difference in the 
length of the valves of sufficient magnitude to be referable to 
any other cause. When it is recollected that, in this genus, the 
frustules are annulate, and the entire structure would appear 
specially liable to variation in size, from the repetitive process 
alternating, as it were, with the extension of the frustulc 
through the deposition of the annulate portions, it will be 
admitted that a more satisfactory test genus could not have 
been selected. I would add, for the guidance of those who 
may desire to repeat these measurements, that a source of 
fallacy exists which must be carefully guarded against; 
namely, the change of apparent size depending on the struc- 
ture under examination not being placed perfectly flat upon 
the surface of the slide. A very little management will, 
however, suffice to ensure the proper position. 
As regards the cell- contents, and the gelatinous envelope 
by which the whole of the Diatomacese are, in a greater or 
lesser degree, surrounded, growth goes on, in all probability, 
indefinitely. The present observations must be understood 
to apply, exclusively, to the siliceous valve of these organisms; 
and are offered with a view to prove that the specific markings 
of any given form are definitely impressed upon it, either at 
the period when division is completed, or almost immediately 
afterwards ; and that whatever may be the normal shape of 
these markings — that is to say, their primary form in the 
young valve — being disposed in a determinate order with 
relation to each other, and to the boundaries of the frustule, 
their ultimate configuration is determined, in a principal 
