26 TRANSACTIONS LIVERPOOL BIOLOGICAL SOCIETY. 



the first. The picture that our mechanistic notions of 

 phylogeny afford is, on the other hand, that of a tree, of 

 one main stem branching into two, and of the repeated 

 branching of these until we reach the terminal twigs 

 represented by one species. Now palaeontology, which 

 alone might show which of these pictures is the true one, 

 will probably never do so ; and it cannot be urged that 

 morphology gives any unequivocal evidence. Every- 

 where in schemes of descent we have had to postulate 

 annectant forms, hypothetical " proto-vertebrata," 

 " proto-mollusca," etc. Does it not indicate that most 

 of our main groups of organisms — vertebrata, arthropoda, 

 mollusca, echinodermata, etc., are truly collateral to 

 each ? 



Yet we see along the course of these main lines the 

 evidence of branching, the actual transformism of 

 species, innumerable divergences, progressions, retarda- 

 tions and extinctions. Each of these is an adventure so 

 to speak, an attempt, mostly unsuccessful, to get the 

 better of inert matter. In its details the study of 

 evolution is the description of this process of trans- 

 formism, and we are, if anything, rather embarrassed by 

 the number of such descriptions, or hypotheses of 

 evolution. If we are to accept the notion of a vital 

 impetus meeting with opposition, inserting itself, as 

 Bergson says, into the interstices of inert matter, and 

 dissociating itself, then the process of transformism is 

 immanent in life. Terrestrial organisms are in their very 

 nature organisms undergoing transformism, and the 

 problem of evolution is the same thing as the problem of 

 the nature of life. 



But our hypotheses of evolution are mechanistic ones. 

 If living matter exists then evolution results from the 

 working of chemical and physical laws. In the process 

 of natural selection variabilitv is the datum. Certain 



