AFFINITIES AND CLASSIFICATION OF ALGE. 53 
the thallus for the first time displays true branching, in the 
other direction to the Nostocacee, with unbranched thallus com- 
posed of moniliform rows of cells enclosed in a copious gelatin, 
and characterized by the formation of heterocysts which had 
already begun to be developed in the Rivulariacem. With these 
two families the second series, the PHYCOCHROMACEE or CYANO- 
PHYCEE, have attained their highest degree of development. 
Before passing on to the third series, the position must be 
discussed of one of the most sharply differentiated but most 
puzzling groups in the entire vegetable kingdom, the Diatomaces 
or Bacillariacew. By some writers they are regarded as exhibiting 
reproduction by conjugation, and are therefore included among 
the Conjugate ; and this view I have myself formerly held ; but 
am now convinced that the so-called formation of zygospores is 
but à modification of the formation of auxospores, and is not 
sexual. The view is certainly plausible that they have descended 
from the Conjugate, and especially from the Desmidieæ, which 
they resemble in some respects, by retrogressive metamorphosis. 
But, on the other hand, there are points which tell strongly 
against this theory. The enormous number and accurate differen- 
tiation of species, and their constancy from very remote geological 
periods, with other considerations, point to the family being a 
member of an ascending rather than of a descending series. To 
this must be added their remarkable powers of motion. It is a 
singular fact, of which one illustration has already been men- 
tioned, that the power of the whole organism to move from place 
to place (Ortsbewegung) is strictly limited to the very lowest 
members of each series, such as Protococcus, the Chroococcaces, 
and the Schizomycetes. As we ascend, this power becomes 
localized in special organs, such as zoospores and antherozoids ; 
until in the most highly developed plants it is almost altogether 
lost. In no family is this property more strongly and strikingly 
displayed than in the Diatoms. For these and other reasons I 
am disposed to look on the Diatomaces as very low down in the 
ascending series, and ranking among Protophyta rather than true 
Algse. If this view be correct, they must be regarded as a branch 
which came to an abrupt conclusion, never arriving at anything 
higher ; though, owing to the very sharp differentiation of the 
family, it is very difficult to conjecture where was the starting- 
point of the branch, possibly in the Cyanophycee rather than 
the Protococcacez. 
