Notes on Recent Literature. 
380 
in a species of Aster. 1 In jfuglans , 2 no egg-apparatus is organised, 
the three micropylar nuclei remain free, and any one of them may 
be fertilised. These instances are of course exceptional, but they 
are indications that there is probably no serious difficulty in the 
way of regarding all the nuclei in the sac, other than the egg, as 
representing gametes which were once free and potentially sexual. 
These have now become specialised, until, with the exception of the 
polar nuclei, they are no longer free nuclei, but cells, and are either 
functionless and evanescent (the antipodals in most cases) or have 
acquired new functions (the synergids usually, and the antipodals 
where they assist in the nourishment of the sac). The polarity of 
the Angiosperm sac and the occurrence of double fertilisation 
(triple fusion) are the chief difficulties in the way of adopting this 
scheme. As regards the former, the difficulty is perhaps less than 
it appears at first sight. In Wehvitschia the sac at the two-nucleate 
stage is bipolar, as in the Angiosperm, but as nuclear division 
proceeds with great rapidity, while the increase in size of the sac is 
slow and no central vacuole is formed, this polarity is lost, the nuclei 
becoming evenly distributed. As the nuclei became fewer, with the 
organisation oftheeggin successively earlier generations, the probable 
slowing-down of division in relation to the enlargement of the 
sac would result in the formation of a central vacuole at a relatively 
early stage, and a consequent grouping of the first-formed nuclei. 
Triple fusion also may be explained as a consequence of this decrease 
in the number of nuclei formed; no process that can be regarded 
as comparable with it occurs in Wehvitschia, where the unused male 
nucleus merely degenerates, and it is probable that the same was 
the case in the primitive Angiosperm. Pearson suggests that triple 
fusion was adopted when the decrease in the number of nuclei 
available for fusion, with the consequent reduction in the number 
of primary endosperm nuclei, and therefore of the amount of endo¬ 
sperm formed, had decreased the efficiency of the endosperm as a 
feeding tissue. The amount of such tissue formed by each fusion- 
nucleus in Wehvitschia is small, and in a sac containing only one of 
these nuclei the developing embryo would have but a poor food- 
supply, unless the efficiency of this endosperm could be increased. 
The introduction of a male nucleus into the fusion would, as in the 
case of the fertilised egg, increase the stimulus to growth, and 
“ direct advantage would seem to accrue to the embryo from the 
supply of a food substance of the same double origin as itself.” 3 
The steps by which triple fusion was initiated must remain con¬ 
jectural, but it may be pointed out that the construction of the 
Angiosperm sac, with its central strand of protoplasm along which 
the polar nuclei move, is very favourable to the possibility of the 
unused second male nucleus being drawn into their fusion. It 
should also be noted that though endosperm formation consequent 
on a triple fusion has thus probably become the rule among 
Angiosperms, cases are occasionally reported 4 where the primary 
endosperm nucleus has, as in Wehvitschia, given rise to a certain 
amount of endosperm before fertilisation. 
1 Opperman, 1904. 
2 Nawaschin and Karsten, ex Coulter and Chamberlain, 1903, 
pp. 90, 91. 
3 Thomas, 1907, p. 426. 
4 Coulter, 1898, p. 83; Du Sablon, 1908 ; Stephens, 1909, p. 368; 
Cook, 1909, p. 57. 
