Igrt] COULTER—ENDOSPERM OF ANGIOSPERMS 381 
in the ontogeny of the gametophyte. When this differentiation 
occurs along with the first appearance of tissue after the free nuclear 
Stage (as in 7, orreya), it is clear that the great bulk of endosperm 
tissue is developed after fertilization. When the differentiation 
occurs during the free nuclear stage (as in Gnetum), it is clear that 
all the endosperm tissue is developed after fertilization. It was 
very easy, therefore, to see in the endosperm of angiosperms only 
belated gametophytic tissue. 
When the relation of the polar fusion to endosperm formation 
began to be appreciated, LE Monnier (1887) suggested that this 
fusion is a sexual act, and that therefore the endosperm is sporo- 
phytic. This would mean that the embryo and endosperm are 
twin sporophytes, the latter for some reason not developing the 
organization of an embryo. This explanation of the polar fusion 
does not seem to have met with much approval. It is important 
to note, however, that lack of approval was probably due to the 
fact that there had developed already a considerable knowledge 
of the great freedom of nuclear fusions within the embryo sac and 
within the endosperm. Clearly all such fusions could not be sexual. 
With the discovery of “double fertilization,” the endosperm 
Problem became conspicuous. One of the nuclei that enters into 
the triple fusion is plainly a male nucleus; one of the polar nuclei 
is sister to the egg nucleus, and this was taken to indicate its 
sexual character; the other polar nucleus has been regarded as 
vegetative in character. The fusion of an undoubted male cell 
and an assumed egg was regarded as an act of fertilization, and 
the product of such a fusion must be a sporophyte. This con- 
clusion as to the nature of the endosperm is inevitable if the triple 
fusion is to be regarded as involving a sexual fusion. 
If the endosperm is a sporophyte, it must be explained why it 
does not become organized as an embryo, but remains as formless 
tissue. Miss SARGANT (1900) offered a very ingenious explanation, 
effectively supported by what seemed to be confirmatory evidence. 
ccording to this explanation, the endosperm remains a formless 
mass of tissue (a ‘“‘monster’’) because a vegetative nucleus enters 
into the fusion and interferes with the legitimate result. This 
view is attractive, but hardly explains the increasing number of 
