THEORIES OF MENTAL EVOLUTION 267 



All three authors base their theories on the assumption that 

 acquired mental characters ^are capable of transmission, and 

 therefore of accumulation in subsequent generations. 



causing the tissue of B to contract and this even though no wave of 

 contraction has passed through the tissue from A to B. Such is a very 

 meagre epitome of Mr. Spencer's theory, the most vivid conception of 

 which may be conveyed in a few words by employing his own illustra- 

 tion, viz. that just as water continually widens and deepens the 

 channel through which it flows, so molecular waves of the kind we are 

 considering, by always flowing in the same tissue tracts, tend ever more 

 and more to excavate for themselves differentiated lines of passage. 

 When such a line of passage becomes fully developed, it is a nerve-fibre, 

 distinguishable as such by the histologist ; but before it arrives at this 

 its completed stage, i. e. before it is observable as a distinct structure, 

 Mr. Spencer calls it a line of discharge. 



" Such being the manner in which Mr. Spencer supposes nerve-fibres 

 to be evolved, he further supposes nerve-cells to arise in positions where 

 a crossing or confluence of fibres gives rise to a conflict of molecular 

 disturbances ; but it is unnecessary for present purposes to enter upon 

 this more elaborate and less satisfactory part of his theory. Less satis- 

 factory not only because more speculative, but because the whole weight 

 of embryological and histological evidence appears to me to be opposed 

 to the speculation. For the whole weight of this evidence goes to show 

 that nerve-cells are the result of the specialization of epithelial or 

 epidermal cells that is, that they arise, not out of undifferentiated 

 protoplasm, but by way of a further differentiation of a particular kind 

 of already differentiated tissue, where this is exposed to particular kinds 

 of stimulation." All I desire now to point out is the a priori probability 

 that nervous channels become developed where they are required simply 

 from the fact of their being required that is by use. 



u And this A priori probability derives so much confirmation from the 

 fact that it is scarcely possible to refrain from accepting it as an answer 

 to the question above propounded, namely, How are we to explain the 

 fact that the anatomical plan of a ganglion with its attached nerves 

 comes to be that which is needed to direct the nervous tremors into the 

 particular channels required 1 It is a matter of daily observation that 

 * practice makes perfect,' and this only means that the co-ordinations 

 of muscular movement which are presided over by this or that nerve- 

 centre admit of more ready performance the more frequently they have 

 been previously performed which, in turn, only means that the dis- 

 charges taking place in the nerve-centre travel more and more readily 

 through the channels or nerve-fibres which are being rendered more 

 and more permeable by use. So much, indeed, is this the case, that 

 when an associated muscular movement takes place with sufficient 

 frequency, it cannot by any effort of the will become again dissociated ; 

 as is the case, for instance, with the associated movement of the eyeballs, 

 which does not begin to obtain till some days after birth, but which 

 then becomes as closely organized as any of the associated movements 

 in the muscles of the limbs. 



" And if this is the case even in the lifetime of individuals, we can 

 scarcely wonder that in the lifetime of species heredity with natural 

 selection should still more completely adapt the anatomical plan of 

 ganglia, with their attached nerves, to the performance of the most 



