33° BOTANICAL GAZETTE [APRIL 
This is indeed an ingenious interpretation. Without endosperm 
hybridization an embryo resulting from a cross would be forced to 
depend entirely upon the kind of food supplied by one parent in its 
early stages of development, as is the case in all plants where 
double fertilization does not occur. It is conceivable that a wide 
cross might so alter the developing zygote that it would be less — 
favorably nourished by food furnished by only one parent in the 
critical stages of itsdevelopment. Hybridization of the endosperm, 
no doubt, may help to adapt the food to the requirements of the 
hybrid embryo more or less intermediate between the two parents. 
It would be still more serviceable in the rare cases of supposed 
merogony (3, 4, 8,12). In these cases, however, nothing is known 
about the development of the endosperm, but what would be the 
nature of an embryo derived from such a wide cross that it would 
be retarded in its development because of an ill-adjusted food 
supply coming from one parent? Such an embryo would be so 
heterogeneous in its hereditary make-up that it would most likely 
not develop at all. In other words, the complexity permitted in 
the embryo would limit the diversity of hybridization before the dis- 
similarity in the composition of the food supplied by one parent could 
have any appreciable effect upon the development of the zygote. 
To postulate the origin of endosperm hybridization as an adap- 
tation having survival value, it is necessary to presuppose that it 
arose in plants which were naturally widely crossed. In such 
forms the effect of heterozygosis in increasing the amount of endo- 
sperm as shown in maize would, no doubt, have been operating. 
Hence, if it is feasible to account for the origin of double fertiliza- 
tion as an adaptation, it would seem more likely that such a process 
arose as a means of increasing the amount of food supplied to the 
embryo rather than as a method of adjusting its composition to the 
needs of the developing plant. In all probability both factors help 
in the early stages of a plant’s development. Whether or not it is 
an adaptation, or whether either of these factors was concerned in 
the initiation of this puzzling process, I do not attempt to decide. 
CouLTEeR and CHAMBERLAIN (6) do not distinguish between the 
fusions of like nuclei and the fusion resulting in double fertilization. 
They say: 
