^889.] On Inheritance in Evolution. 1063 
soma, and the effect on the germ-plasma. Those who sustain the 
view that acquired characters are inherited, must, I believe, under- 
stand it as thus stated. The character must be potentially ac- 
quired by the germ-plasma as well as actually by the soma. Those 
who insist that acquired characters are not inherited, forget that the 
character acquired by the soma is identical with that acquired by 
the germ-plasma, so that the character acquired by the former is 
mherited, but not directly. It is acquired contemporaneously by 
the germ-plasma, and inherited from it. There then is truth in 
the two apparently opposed positions, and they appear to me to 
be harmonized by the doctrine above laid down, which I call the 
Theory of Diplogenesis, in allusion to the double destination of 
the effects of use and disuse in inheritance. 
From the preceding considerations we learn that a new char- 
acter is not inherited unless it is acquired by the germ-plasma, as 
well as by the soma. Should it fail of the former it will not be 
inherited, although it may appear in the soma. It is also evident that 
the same character appears in the soma of later generations by 
virtue of its inheritance by their germ-plasma. Hence should it 
fail to appear in the adult soma of one generation, it might arise 
in a later one ; and hence the possibility of atavism, and an altern- 
ation of generations. Intermittent stimulus might be followed by 
intermittent activity of growth energy. This would be especially 
apt to occur during the assumption of sexuality by animals and 
plants whose reproduction had been performed by cell-division or 
budding only. And such is the character of most types of al- 
ternate generations ; a sexual type alternates with a non-sexual 
type. The advantages being on the side of sexual reproduction 
on account of its increased opportunity of variation, it has re- 
placed the more primitive method by the process of natural selec- 
The chief source from which acquired characters are introduced 
into the germ-plasma, and hence into the soma o'i the next gen- 
eration, is probably the spermatozo5id ; since it is endowed with 
a greater potential energy than the ovum. The latter furnishes 
nutritive material for the supply of the needs of growth. That 
the male is the chief source of variation is also indicated in the 
