''268 
Transactions of the Eoyal Society of South Africa. 
poles of the sac, and they usually show a lower degree of specialisation than 
in a sac in which polarity is strongly developed. 
The fact that most of the sac-nuclei (sometimes only a small minority) 
are incorporated in units of protoplasm is a character of the Angiosperm as 
compared with Grnetum. The transition from the state of a free nucleus to 
that of a uninucleate-cell is very frequently effected in the ontogeny of the 
individual. In Gnetum and Welwitschia we have examples in the constitu- 
tion of the generative-cell in the pollen-grain or pollen-tube, and, if Karsten 
and Lotsy are correct in stating that the male gamete functions as a free 
nucleus, in the cell-formation which must immediately precede or follow 
fertilisation. The nuclei of the Angiosperm-sac are free in their earlier 
stages ; the later incorporation of some of them in uninucleate-cells does not 
invalidate a comparison with the free nuclei of G-netum. 
We have already seen that the fusion v/hich produces the primary 
endosperm-nucleus does not depend for its success upon any particular 
morphological or physiological characters inherent in the polar-nuclei as 
such. For the further consideration of this question reference may be 
made to Table II wherein are diagrammatically indicated certain characters 
of ten selected embryo-sacs.''^ In each case the descent of the sac from the 
mother-cell is traced, the nuclei present in each generation being indicated 
by dots. The nuclei which fuse to form the primary endosperm-nucleus 
are enclosed in an inner figure. t The participation of the second sperm- 
nucleus in the fusion is disregarded (see p. 264). The embryo-sac in 
the forms represented may reach the condition in which it is ready for 
fertilisation immediately after the 6th, 5th, 4th, or 3rd simultaneous nuclear 
division, commencing with and including the first division of the mother- 
cell. For the sake of convenience these are called the 7th, 6th, 5th, or 
4th nuclear generation, respectively. In Euphorbia virgata (Col. i) the 
embryo-sac is matured in the 7th nuclear generation, in most Angiosperms 
(Col. ii) in the 6th, in six other cases (Col. iii-viii) in the 5th ; in Lilium 
and Cypripedium (Cols, ix, x) in the 4th. 
* Little of the very extensive literature on this subject is available here. The 
information contained in the table is derived from — 
Dessiatoff, 1911 ex Samuels, 1912 {Euphorbia virgata). 
Johnson, 1914 {Peperomia hispidula) . 
Campbell, 1908; Brown, 1908 {Peperomia pellticida). 
Stephens, 1909 (Penaeaceae). 
Samuels, 1912 (Gunnera). 
Modilewsld, 1909 ex Samuels, 1912 (Onagraceae). 
Went, 1908 (Podostemaceae). 
Pace, 1907 (Cypripedium). 
The macrospore mother-cell is indicated in the table by the symbol © . 
t Except in Column vii. We have not access to Modilewski's original paper, and 
the details of the endosperm-fusion are unknown to us. 
