544 
Reginald Ruggles Gates 
for all the differences between 0. gigas and 0. Lamarckiana, 
and the second factor may be one merely of readj ustment 
cousequent upon the first. It is probable that the number of 
cells is approximately the sarne in botli cases. 
IV. Origin of the 4 X number. 
There are several interesting possibilities as to tbe origin of the 
28 cliromosomes in 0. gigas , and it must be remembered that the 
occurrence we liave to account for is a comparatively rare one, due 
to a departure from the ordinary life cycle of 0. Lamarckiana. de 
Vries describes the appearance of a mutation as resulting from the 
union of a '’mutated” germ cell with an ordinary germ cell. However 
this view can scarcely apply in this case, since, although it is pos- 
sible that germ cells may occasionally be produced with the un- 
reduced number of cliromosomes, fertilization with such a germ 
cell would produce an organism with 21 instead of 28 cliromosomes. 
The possibilities of two such unreduced germ cells — an egg and a 
sperm — getting togetlier in fertilization are very remote. Moreover, 
no instances of this sort are known, and if this were the metliod 
of origin one would also expect to find mutants occurring with 21 
cliromosomes. 
It therefore seems much more probable that the double number 
of chromosomes originated in some otlier manner, eitlier at the time 
of fertilization or shortly afterwards. 1. Four nuclei may have fused 
iu the embryo sac at the time of fertilization, giving a fusion nu- 
cleus having 28 chromosomes, from whicli the embryo developed. Iso 
parallel for this is known at the present time, although Treub (1898) 
and Lotsy (1899) described a "pseudembryo” originating from the 
upper endosperm nucleus in the embryo sac of Balanophora. This 
embryo sac is probably forrned without reduction, the egg and syner- 
gids degenerating without fertilization. 2. It is possible that in the 
yonng embryo soon after fertilization, or in the fertilized egg itself, 
the nucleus failed to complete its division, forming something similar 
to a monaster, the daughter chromosomes afterwards forming a single 
nucleus. That the descendants of such a nucleus would have the 
double number of chromosomes is shown by the experimental work 
of Nemec (1906) and otliers in plants, and of Boveri in animals. 
Lillie (1902 and 1906) has also shown that by treatment with KCl 
and by otlier means the chromosomes in the egg of the annelid 
