eee eS SR SR ea oh pie a eg See Pd ets oe, SaaS ee Te a ay a Le ee ee ae ee ra ee a a ae ores 
Tgro] BROW N—ASCOCARP OF LEOTIA 455 
and are derived from a single nucleus with the haploid number 
of chromosomes. So far as the writer knows, there is no reason 
for thinking that there is. It is.well known in agriculture that 
self-fertilization causes a diminution in the size and vigor of maize. 
SHULL (26) has shown that ordinary maize is a complex hybrid. 
If the elementary species are segregated out of the hybrid, self- 
fertilization has no bad effect, and it would seem that the deteriora- 
tion of agricultural maize after self-fertilization is due to the 
segregating out of the smaller elementary species, which when 
recombined will again produce the more vigorous agricultural 
variety. According to the Mendelian interpretation, the harmful 
effect of self-fertilization in other genera is due to the same cause. 
The value of cross-fertilization for evolution is evident when we 
remember that it affords a means for trying a new character in a 
great number of combinations, one of which may finally be espe- 
cially adapted to its environment. It may be that the persistence 
of cross-fertilization is due to its value in evolution, and that the 
presence of self-fertilization in such a relatively small number of 
the higher plants is due to the fact that such plants had less chance 
to interchange characters by crossing and so to develop along 
favorable lines. 
It is evident that the above beneficial effects could not be 
produced by the fusion of distantly related nuclei if these were 
all in the same plant and were derived from a single nucleus with 
the haploid number of chromosomes, for even if the potentialities 
of one of the fusing nuclei were to be changed in some way, the 
change would probably be inherited by only one-half of the result- 
ing progeny, and the nuclei of these would have the same poten- 
tialities as the changed nucleus and there would be no chance for 
further mixing. “The end result would be the same if the nucleus 
whose potentialities had been changed had divided, and the two 
daughter nuclei had fused and given rise to the progeny. But 
even if a change in the fusing nuclei could be thought of as bene- 
ficial, there is no reason for thinking that such a change would 
occur. The work of JENNINGS (18) and JOHANNSEN (19) seems 
to show that the germ cells of organisms derived from a single 
pure individual are very rarely affected, even if the organisms 
