1915] CURRENT LITERATURE 161 
showed almost exactly one-half of the ovules with complete embryo sacs 
“nearly filling the nucellus,” while the other ovules, “though full sized, had 
either quite aborted embryo sacs, the nucellus being a uniformly cellular mass, 
in which, however, the remains of the aborted megaspore could usually be dis- 
‘ingaished, or had only a minute cavity to represent the embryo sac.’’ The 
fact that somewhat less than half the seeds of ripe pods were found to have 
aborted is believed to be due to a selective elimination through dropping of 
young pods that chanced to have all or nearly all aborted ovules. In the F, 
generation about one-half of the plants had almost wholly normal pollen grains 
and in the other half approximately 50 per cent of their pollen grains aborted 
asin F,. The F, plants with normal pollen also had few aborted ovules just 
as the parent species had, while those with semi-sterile pollen had about 50 
per cent aborted ovules. The F, progenies of fertile F, plants were wholly 
fertile. The semi-sterile F, plants, like all F, plants, produced F; offspring 
a half of whose members were fertile and a half semi-sterile with regard to 
both pollen and ovules. Some of the fertile F; stocks were grown on a large 
scale and found to breed true to that characteristic in F, and 
The author points out clearly that the form of sterility observed in Stizo- 
lobium hybrids manifests itself only in the haploid, never in the diploid gener- 
ation. The F, ratio of 1:1 from selfed F, plants is not, therefore, a zygotic 
but rather a gametic ratio. It i is interpreted accordingly as simple Mendelian 
(simplex for genetic factors), it is obvious that donintics and recessiveness, 
either complete or partial, cannot in any way be concerned in this problem 
As a working hypothesis, the author assumes two genetic factors, K and z 
one present in S. deeringianum and the other in each of the other species. It 
is assumed that the presence of one or the other of these factors is essential for 
the development of normal pollen grains and embryo sacs, but that in the 
presence of both factors, just as in the absence of both, no development results. 
The hypothesis accounts for all the facts observed. The author promises the 
further crucial test of crossing together different fertile lines from the progeny 
of semi-sterile plants, in which, if the hypothesis holds, half of the crosses 
should yield only fertile and half only semi-sterile offspring. It would be 
instructive also to cross together the three forms other than S. deeringianum, 
all of which should produce wholly fertile hybrids. 
It remains only to be said that, as an alternative hypothesis, K and L 
might be regarded as inhibitors of complete gametic development when present 
singly, the one neutralizing the effect of the other when both are together. 
KL and ki would then result in normal and K/ and &L in aborted gametes, just 
the reverse of the author’s assumption. Both assumptions are equally in 
accord with the observed facts and the same further results are to be predicted 
from both. The author is to be congratulated upon the thoroughness of his 
