ARE THE DESMID3 AND DIATOMS SIMPLE CELLS? 
137 
families more or less brilliantly tinted, green in the one, yellow 
or greenish-yellow in the other, and at times minutely and 
sparsely granular ; the colourless formative protoplasm existing 
in a free state within the outer wail ; a centrally situated 
nucleus ; terminal vesicles ; chlorophyll granules, minute gra- 
nular masses, the office of which is as yet unknown ; oil glo- 
bules ; an outer perforate protecting wall ; and lastly, in a 
large number of species, an external gelatinous matrix. 
But although there is this close similarity both in structural 
and physiological characters, when we come to details some 
conspicuous peculiarities present themselves in the Diatomaceae, 
for which there would appear to be no parallel elsewhere. First 
and foremost amongst these is the substitution of an inorganic 
for an organic outer covering ; that is to say, one composed of 
silex, instead of cellulose, which, in some form or other, con- 
stitutes the protective and sustaining wall of every other vege- 
table cell ; and the composition of the siliceous covering, not 
of a single continuous piece like the cellulose covering of the 
Desmidiacese, but of several, and in certain genera an indefinite 
number of, distinct pieces.* 
The twin portions of the Diatom frustule, called the “ valves,” 
are, I presume, far too well known to require any description 
at my hands. But I must mention that, although the valves 
constitute twin members of the protective covering of the 
organism, so far as their development is concerned, they are 
entirely distinct from the “intermediate piece” or “connect- 
iug zone,” which is developed subsequently to the completion 
and consolidation of the valves, and is sometimes a permanent, 
sometimes a supplementary and deciduous portion of the struc- 
ture.! Hence although the general siliceous covering of the 
Diatom is undoubtedly the homologue of the cellulose covering 
of the Desmid, the dual composition of the valves, coupled with 
the also dual composition of the connecting zone, the over- 
lapping pieces of which slide freely and independently one 
within the other, renders quite untenable the opinion expressed 
by some writers, that the connecting zone is the silicified 
portion of the primordial utricle,! “ left exposed when the 
valves recede from each other in order to make room for the 
increase” just referred to. For, admitting for the sake of 
argument, that the first developed or innermost of the two 
overlapping pieces of the connecting zone were thus formed, it 
is obvious that when this is once consolidated the perfect 
* In u Khabdonema.” 
f See paper already referred to. 
t Of course the term, is here used as applying to the superficial layer of 
protoplasm in immediate contact with the siliceous wall. 
NEW SERIES, YOL. I. NO. II. L 
