THE OUTER AND INNER TISSUES OF ANIMALS. 307 



itself under its new conditions? It produces a superficial 

 substance which plays the part of the epidermis and grows 

 outwardly. For since the surface, subject to friction and 

 exfoliation, has to be continually renewed, there must be a 

 continual reproduction of an outermost layer from a layer 

 beneath. That is to say, the contact of this deep-seated tissue 

 with outer agencies, produces in it some approach towards 

 that character which we find universally characterizes outer- 

 tissue. But while we see under this exposure to the con- 

 ditions common to all integument, a tendency to assume 

 the structure common to all integument, we see no tendency 

 to assume any of the specialities of tegumentary structure: 

 no rudiments of glands or hair sacs make their appearance. 



Analogous conclusions may be drawn respecting the pro- 

 cesses of differentiation by which from the outer layer 

 nervous tissue and finally a nervous system are evolved. 

 Here, also, both direct and indirect equilibration appear to 

 have operated. Two reasons may be assigned for the belief 

 that the transformation of certain superficial cells into sensi- 

 tive cells was initiated by exposure to external stimuli. The 

 first is that, extremely unstable as protoplasm is, disturb- 

 ances received by the outer side of a specially-exposed cell 

 could scarcely fail to cause changes passing through it 

 towards the interior mass of the body, and that perpetual 

 repetition of such changes would tend to generate channels 

 of easy transmission through the protoplasm. The second 

 reason is that, if we do not assume this process of initiation 

 but assume that survival of the fittest was the sole agency, 

 then no reason can be assigned why the nervous system 

 should not have been at the outset formed internally instead 

 of being initiated externally and then transferred to *the in- 

 terior: the roundabout process would be inexplicable. At 

 the same time the production of a central nervous system by 

 introversion of superficial sensitive cells cannot be ascribed 

 to the differentiating effects of external stimuli, but must be 

 ascribed to natural selection. No perpetual repetition of 



