THE 
AMERICAN NATURALIST 
ee 
? DES: 
Vor XIX, July, 1896. 355 
THE CLASSIFICATION OF DIATOMS (BACIL- 
LARIACE). 
By CLARENCE J. ELMORE. 
There have been many systems of classification employed 
for the Bacillariacex, but very few of these have any valid 
claim, to be regarded as natural systems. They may be divided 
into three classes; (1) those based on the structure of the valves, 
of which Kuetzing’s, Prof. H. L. Smith’s, and that employed | 
by Kirchner are examples; (2) those based on the form of the 
frond, the connecting membrane, and the gelatinous envelope, 
represented by Rey. Wm. Smith’s; and (8) those based on the | 
_ structure of the endochrome and the manner of forming auxo- 
~ spores, represented by that of Paul Petit. The following i is me 
brief outline of the systems mentioned. 
‘Kirchner divided the Bacillariacex into two groups’ those os 
' Read before the Botanical Seminar of the University of Kapukai d March 21. - 
crypt ora von Eichleaien ; TES m eee 1878. 
ey iat Leipsic, Pu 
a whose markings are bilateral, that is, arranged on two sides of _ 
a longitudinal line or raphe, and? those with radial markings. oe 
