iS94] 385 LPoulton. 
a chavactei- of the species witliout selection, but by recession will 
tend to be lost in the subsequent generations. 
I have briefly touched on some of the chief difficulties which 
are advanced against natural selection. I now propose to devote 
the remaining part of my time to the difficulties which seem to 
me to ap])ly to the Lamarckian theory. 
Lamarckian evolution, as I have mentioned before, depends 
upon acquired characters. A good deal of misconception has 
arisen from this use of the word "acquired." An acquired char- 
acter has sometimes been interpreted to mean any character that 
an animal has come to possess ; hence, inherent and acquired 
characters have been confused. The word "acquired," as used 
by biologists, must be understood to have a limited and special 
application, meaning only those characters which have been pro- 
duced in the organism by the incidence of external forces, or by 
the action of its own forces, use and disuse of parts, and so on, 
during its life. Weismann has suggested the term "blastogenic" 
for chai-acters potentially present in the germ at the very beginning 
of life, and "somatogenic" for those which appear afterwards and 
are not potentially present in the germ. Here blastogenic is tlie 
equivalent of inherent, and somatogenic of acquired. 
Some years ago I suggested that the terms "centripetal" and 
"centrifugal" might be employed to express this acquired differ- 
ence, acquired characters being centripetal, because they are 
impressed upon the body or one of its parts from without ; inherent 
characters being centrifugal, because, arising from within, due to 
the essential nature of the organism itself, in the course of devel- 
opment tliey come to appear, as it were, on the surface as visible 
features. 
Wiien we now consider the transmission of acquired characters, 
upon which the Lamarckian theory certainly depends, we are led 
first of all to inquire whether it is possible to frame a theory of 
heredity within which such transmission can be included. If, for 
instance, there is a change in the brain of an animal, owing to 
the exercise of some part of it, how can such a change in the 
brain-cell be transferred to the germ-cells of the animal, so as to 
be transmitted to its offspring? It may be objected, if you can 
prove that such transmission does take place it is no matter 
how it takes place. Quite true, if the evidence is sufficient and 
indisputable. But we must remember that the amount of evi- 
PROCEEDINGS B. S. N. H. VOL. XXVI. 25 AIG. 1894. 
