127 
Primitive A ngiosperms. 
sporophytic number of chromosomes appear in an arbitrary way in the 
chalazal nucleus of the second generation. But we do not know whether 
this remarkable anomaly is general or even common. 
After fertilization two distinct bodies, the embryo and the endosperm, 
are formed in the embryo-sac, and they shortly fill it to the exclusion of 
all other tissue. The embryo is derived from the fertilized ovum, and of 
course belongs to the succeeding generation — the young sporophyte. The 
endosperm is derived from the union of three nuclei : two belonging 
to the female gametophyte of the mother plant, and one to the male 
gametophyte of the plant whose male element has entered the fertilized 
ovum. The synergids and antipodals disappear sooner or later : there 
is then no trace left in the embryo-sac of the female gametophyte as 
it existed before fertilization, and no tissue whose nuclei show the gameto- 
phytic number of chromosomes. 
The only possible representative of the female gametophyte is the 
endosperm, and before 1 898 most botanists considered it to be a belated 
female prothallus. This view, though sometimes attributed to Hofmeister, 
was in fact first maintained by Strasburger ( 85 , pp. 137-139) in 1879. 
Le Monnier ( 53 ) published an alternative hypothesis in 1887. He 
regarded the union of the polar nuclei as a sort of fertilization. The upper 
polar nucleus is sister to that of the ovum : it may well be considered 
as a female nucleus. The lower polar nucleus must then represent the 
male element. The endosperm is developed as a result of their union : 
it is therefore a short-lived misshapen embryo. 
Botanists are still divided between these two camps, though Nawa- 
schin’s discovery of 1898 has certainly increased the number of Le Monnier’s 
followers. But it is no part of my plan to discuss their rival claims here. 
The pregnant fact in dealing with the ancestry of Angiosperms is that the 
question is still undecided. Either view can be held, and is held, by 
botanists of undoubted authority. In other words, we have so far no links, 
living or fossil, which enable us to trace the historical development of the 
endosperm in Angiosperms with any degree of certainty. 1 
Such evidence may be sought in two directions. We may examine 
the endosperm formation among aberrant Angiosperms, particularly among 
such forms as we are disposed on other grounds to consider primitive 
in structure. Or we may seek for some suggestion of the origin of the 
endosperm in the comparative structure of the Gymnospermous prothallus. 
Knowledge of the minute structure of living plants is as yet so far 
from complete that we may still hope to find an Angiospermous embryo-sac 
in which the endosperm shows some decided trace of its origin — whether it 
prove to be a belated prothallus, a specialized sporophyte, or some other 
1 Miss Berridge’s interpretation of certain facts recently described in Ephedra may perhaps 
afford a clue. See p. 133. 
