io66 The American Naturalist. [December, 
to its environment, internal and external. Not only does the con- 
clusion seem reasonable, but it is supported by the well-known 
effects of mental states on various functions of the adult organism; 
such as the formation of ptomaines in the digestive tract, the de- 
pression and acceleration of the heart's action, the control of erectile 
tissues, etc. ; to say nothing of ordinary muscular movements. 
It is for such reasons that I have maintained that there would 
have been no evolution of animal types at least, if acquired char- 
acters were not inherited. Promiscuous variations there would 
have been, but it is certain that the probabilities arc enormously 
against the persistence of any of them beyond one generation, 
should they not be inherited. That any succession of such vari- 
ations could be profitable is also highly improbable, to say nothing 
of the improbability of their displaying the direct relation to use 
which we find in them in point of fact. The explanation of the 
appearance of such continued series of direct adaptations, such as 
has been demonstrated for instance in the phylogeny of the horse, 
is only explicable to my mind on the supposition of the inherit- 
ance of the direct effects of use, which use has been primarily 
directed by sensation. Without these factors evolution would have 
been suppressed at its inception, just as we see that it has been 
retrogressive or downward so soon as one or another of these 
factors has been withdrawn. 
The time when the impressions of physical habits are conveyed 
to the reproductive elements has an important bearing on the 
question of inheritance. The life of an animal may be divided in- 
to three periods ; those of embryonic life, of adolescence,, and of 
maturity. During embryonic life impressions are exclusively 
somatic, and can be only obtained through or from parental stimu- 
lus and parental environment. Such will reach the embryo 
though nutrition, and through the direct mechanical contacts and 
strains of the environment. The environment of the oviparous 
forms is external to the parent ; that of viviparous forms is the walls 
of the oviduct, uterus, etc., within the parent. Ryder has shown 
with much reason that the nature of the contact of the chorion with 
the walls of the oviducts or uterus has determined the shape of 
the placenta ; and that the invagination of the embryo which re- 
