38 THE EVOLUTION OF MAN. 



established ; others, on the contrary, are most dou]:>tful ; in 

 yet others, it depends upon the subjective proportion of the 

 knowledge of the naturalist and on his capability of draw- 

 ing conclusions, what degree of probability he will accord 

 to them. It is, at all events, necessary thoroughly to dis- 

 tinguish between the absolute certainty of the general 

 (inductive) theory of descent, and the relative certainty of 

 the special (deductive) hypothesis of descent. We can 

 never in any case prove the whole ancestral line of an- 

 cestors of an organism with the same certainty with which 

 we regard the theory of descent as the only scientific expla- 

 nation of the organic forms. On the contrary, the special 

 proof of all separate parent-forms must always remain 

 more or less incomplete and hypothetical. That is quite 

 natural. For all the records of creation upon which we 

 tely are in a great measure incomplete, and will always 

 lemain incomplete; just as in the case of Comparative 

 [^hUology. 



Above all, Palaeontology, the most ancient of all records 

 of creation, is in the highest degree incomplete. We know 

 that all the petrifications with which we are acquainted 

 form but an insignificantly small fragment of the whole 

 number of animal forms and plant forms which have ever 

 existed. For each extinct species obtained by us in a petrified 

 condition, there are at least a hundred, probably thousands 

 of extinct species which have left no trace of their existence. 

 This extreme and most deplorable defectiveness of the palseon- 

 tological record of creation, upon which it is impossible to 

 insist too strongly, is very easily accounted for. The very 

 conditions under which organic remains become petrified 

 neoeesitaie it. It is also partly explicable as the result of 



