Wallich^ on the Diatom-valve. 
137 
degree, by the form of the frustule, and by the direction in 
which that form exercises a constrictive force, whilst the 
siliceous material of its valves is still in a plastic condition. 
If we admit this proposition, we cannot fail to comprehend 
how materially the nature of valvular markings may be modi- 
fied by any variation in the condition to which the parent 
frustule may be exposed during the period of division ; and 
we at once recognise the futility of drawing specific characters 
from the mere numerical estimate of striae within certain 
limits, or, indeed, from any structural peculiarities apart 
from such as are constant. 
Much of the confusion that exists with regards to the 
striation^' or lineation^^ of the Diatomacese arises, I con- 
ceive, from the vague manner in which these terms are 
employed to indicate different portions of the valvular struc- 
ture. Thus, in Pleurosigma, the terms are used, by some 
writers, to indicate the lines presented to the eye by the 
coalescence of the several series of intra-linear spaces ; whilst, 
by others, they are intended to denote the lines formed by 
the boundaries of those spaces. This last is certainly the 
correct view ; and it is borne out by the circumstance that 
the outline and peculiarities of the intra-linear spaces are 
determined, as shall presently be shown, by the inherent 
order of the lineation, and not the lineation by the inherent 
development of the spaces. 
In the ^ Synopsis of British Diatomacese,^ the valve of 
Pleurosigma is stated as being striated ; strise resolvable 
into dots, which are frequently hexagonal.^^ Other writers 
are also in the habit of alluding to ^'striae composed of 
dots.-*' From this definition it is impossible to gather 
whether the lines we observe crossing each other at certain 
angles are indicated, or the spaces bounded by the intersec- 
tions of those lines. In reckoning the number of lines in 
the one-thousandth part of an inch, the measurements are 
evidently derived from the first of these points ; whereas it 
is undeniable that the so- called " dots occupy the intra- 
linear spaces. The number of dots,^^ and the number of 
" strise,^^ therefore, can never tally ; and, for a similar 
reason, it involves a contradiction to say that the striae are 
resolvable into dots.^' It remains still a point of dispute 
whether the intra-linear spaces are depressions or eleva- 
tions. It is not improbable that they are elevations in some 
species, and depressions in others. Fortunately, however, 
the solution of the question is a matter of no great moment 
for purposes of classification ; and it becomes of still less 
importance when we bear in mind, that as a valve happens 
