On the Classification of Diatomacese. By Paul Petit. 1 3 
“ First sub-family — Endochrome lamillate : Placochromaticeae. 
“ Second sub - family — Endochrome granulate : Coccochro- 
maticeae. 
“ In spite of the labour bestowed by Dr. Pfitzer on the study of 
the endochrome, he has allowed himself to be led away by previous 
German classifications. I have, however, followed him in the sub- 
division into symmetrical and non-symmetrical forms, and valves 
with or without a central nodule. This method of division has the 
grave inconvenience of separating groups between which there is 
an affinity, as, for example, the Meridiese and the Licmophoreae, and 
of placing the Tabellariese before the Licmophoreae, whilst, as we 
shall see farther on, that the latter forms the connecting fink be- 
tween the Fragilariese and Tabellariese (tandis que comme on le 
verra plus loin ce sont ces dernieres qui etablissent le passage des 
Fragilariees aux Tabellariees). 
“ I am far from admitting all the groups of Dr. Pfitzer ; above 
all, those that he has established sometimes for a single genus, and 
even for a single species, for example, Amphitropidese.* It is the 
same with the new genera created for those species that are found 
at the extremity of a genus, and presenting only slight variations 
from the type form in their plasma. These species, on the contrary, 
by the modifications of their endochrome, establish the passage of 
the groups into each other. 
“ The Diatomacese form a very natural family, having on the 
one hand a connection with the Monads through the genus 
Cocconeis, and on the other with the Confervaceae through the 
genus Melosira. 
“ It is, in fact, the genus Cocconeis that offers physiologically 
the simplest organization ; a layer of endochrome rests on one 
valve only, leaving the other completely independent. On the 
other hand, the genus Melosira, includes those species which have 
cylindrical frustules connected under the form of filaments, and 
containing an endochrome the granules of which indicate a great 
affinity with those of the Chlorophyllacese as well in their form as 
disposition. We know that the Melosira becomes green when 
desiccated.! 
“ Much remains to be done to firmly establish the groups. It 
is only with time and patient research that we can arrive at this 
result. There still remain a great many genera in which the 
disposition of the plasma has to be studied, but one is able to foresee 
that according to the immutable laws of nature all of them can 
be placed in the proposed groups. Some additions will possibly be 
necessary for future discoverers, but I dare to hope that these 
alterations will not affect the bases of my method. The number of 
genera is too large to allow me to indicate all of them. I therefore 
* Pfitzer, loc. cit., pp. 94, 95. f Kiitzing’s 1 Baeillarien,’ 1844, p. 23. 
