576 
Notes . 
Under all circumstances, whatever name the thing may bear, we 
must ask ourselves the same question as regards these phenomena 
as in the dimorphic Thallophytes ; we must inquire whether a more 
or less definite number of cell-generations must be passed through 
before the fruit-generation can follow. In the Bacillus of anthrax 
(. Bacillus anthracis ), Buchner (1890) has already proved that it can 
be propagated as long as you please by division, and that at any 
moment the formation of spores can be induced by direct influence 
of the outer world. Schreiber (1896), who has closely investigated 
the conditions of spore-formation in several Bacteria, has been able 
to prove still more definitely that the spore-formation always begins 
as a direct consequence only of external conditions. Starting from the 
germinating spore, it was possible, after the third division, to induce 
spore-formation again. In Chlamydomonas, I was able to prove with 
certainty that the cells, through innumerable generations, propagated 
in an exclusively vegetative manner, but that at any time the formation 
of sexual swarm-spores can be attained with ease and certainty. 
Most probably the same holds for the Desmidiaceae, in which certain 
species may be propagated for many weeks together by division, but 
from the first were capable of sexual propagation when exposed to the 
conditions characteristic for it. 
On the other hand, the Diatomeae seem to have a necessary alter- 
nation of generations, just as, according to the investigations of 
Maupas, the Ciliatae among the Infusoria. According to the theory 
founded by Pfitzer, the cells of the Diatomeae, whose silicified cell- 
wall consists of two parts fitted one within the other, do not grow 
in the direction which is usually styled longitudinal. The consequence 
of this is that on each division one of the daughter-cells maintains 
the length of the mother-cell, the other will necessarily be smaller 
by the thickness of the membrane. Thus by continued divisions 
the cells become smaller and smaller till, on reaching a certain 
minimum size, the process of auxospore-formation appears, by which 
the original maximum length is again attained. This generally 
acknowledged theory has been supported by the investigations of 
Miquel. He cultivated a number of Diatoms in artificial nutritive 
media, and noted in the successive generations a gradual lessening 
of the cells, till finally the formation of auxospores followed very 
freely. Thus, according to the statements of Pfitzer, Miquel, and 
others, the view might appear to be sufficiently established, that in 
