Cytology of a new variety of A chly a americana. 173 
still to some extent the case, as the study of apogamy teaches. 
Gametes were developed along two lines : on the one hand 
in close relation to a reducing-division, on the other without 
any such close relation to a reducing-division. The reducing - 
divisions in plants must then be regarded as having had two 
distinct origins at least, one of these possibly having an origin 
common with that which occurs in animals. 
As sex, including conjugation, had apparently several 
independent origins in the Thallophytes, it is reasonable to 
expect such differences. Our wonder is excited more by the 
uniformity than the diversity of the results as they lie before us. 
To guard against misconception, it may be expressly stated 
that, in forms like Fmcus , Dictyota , and Achlya , I see no 
reason why the number of chromosomes should not have 
been a very variable one at the outset ; whilst gametes with 
accurately reduced nuclei represented the most complete 
gametal development, and the law of the survival of the 
fittest led to the relative constancy in the number of the 
chromosomes which we now see. Certain apparent incon- 
stancies in the number of chromosomes receive in this way 
a provisional explanation. 
If these considerations have weight, there is no longer any 
necessity for a laboured reconstruction of well-grounded 
views as to the nature of Thallophytes to make them fit in 
with a theory of karyology. Moreover it becomes possible 
to explain the curious cases of apogamy so common amongst 
the lower plants. 
Bearing in mind the variations met with in the number 
of chromosomes, it is not unlikely that the number of chromo- 
somes in apogamous gametes may be doubled by growth, 
under the influence of definite external conditions. In plants 
like Fucus simple atavism would explain this. There is, how- 
ever, no need for a resort to atavism. The difference in 
number observed in the chromosomes of the cells of the 
prothallus of Finns silvestris must arise somewhere and 
somehow. The nucleus of the macrospore could not have 
more than one number. All the variations could scarcely 
