To4 BOTANICAL GAZETTE [FEBRUARY 
spermatogenesis, and consequently the cause of sterility in hybrid 
pigeons. But in the case of O. /ata such an explanation can scarcely 
suffice, for we cannot imagine that any incompatibility or failure of 
fusion of the maternal and paternal plasms or chromosomes could 
almost simultaneously bring about the disintegration of the tapetal 
cells. GREGORY (9) suggests that ‘‘it seems more likely that sterility 
of the male organs is the expression of some deeper-lying phenomenon 
affecting the whole plant,” and a similar view is held by ‘TISCHLER 
(31). Thus it appears that the cause must be sought in some more 
wide-spread phenomenon in the hybrid, causing general lack of nutri- 
tion of the parts which degenerate; hence the tapetum and pollen 
mother cells degenerate simultaneously and for the same reason. 
Recent papers, which need not beJreferred to in this connection, 
have also shown that hybridism may cause abnormalities or failure 
of development in the female gametophyte, though this is apparently 
less common. These facts, together with the usually normal and in 
some cases increased luxuriance of the sporophyte, point to the 
gametophyte as that part of the life history which suffers chiefly from 
“lack of nutrition.” The sporophyte generation can evidently be 
carried through successfully if fertilization is once accomplished; 
but something which first displays itself as lack of nutrition (as 
shown by the failure of the tapetum to function, as well as perhaps 
by the pale staining of the chromatin of the pollen mother cells in 
some cases) prevents the normal development of the gametophyte. 
That this is not the whole case, however, is shown by the fact 
that the reduction divisions are abnormal, even when the tapetum 
appears perfectly developed and functional. Hence something other 
than “‘lack of nutrition” must be the cause of these abnormalities in 
development. The ingrowth of the median wall layers to fill the 
loculus after the disappearance of the pollen, points rather in the 
opposite direction. But perhaps a harmony of these two points of 
_ view may be reached. If the view, recently emphasized by Cook and 
SWINGLE (6a), ALLEN (1), CARDIFF (6), and others among botanists, 
may be accepted, namely, that the maternal and paternal chromatin, 
which contribute to the fusion nucleus at fertilization, maintain their 
separate identity throughout the sporophyte generation, so that the 
act of fertilization is really only completed with the fusion of the 
