NATURAL SELECTION 37 



and even if it were, as it is not, the only factor, why should 

 all characters of animals and plants be useful to their 

 possessors? Would not many chance variations be pre- 

 served whether they were useful or not? Hurtful char- 

 acters, of course, would be eliminated, but why should not 

 certain neutral characters persist without reference to natu- 

 ral selection ? It is truly a remarkable fact, and one hardly 

 to have been anticipated, that so large a proportion of the 

 habits and structures of organisms are useful to their pos- 

 sessors. On page 66 et seq. is shown one way in which 

 useless characters may be preserved. [Physiological seg- 

 regation.] 



A third objection urged against the importance of the 

 agency of natural selection in evolution is that certain 

 organs which are useful in their present condition could 

 hardly have been so when beginning to form in the past, 

 or, at least while as yet very slightly differentiated, could 

 hardly have been sufficiently useful to be of " selection 

 value," i.e. to secure the survival of the animals or plants 

 possessing them. This is really a modification of the 

 objection last mentioned. In reply we may say, as we did 

 in the last case, that it is difficult to say what might be 

 the usefulness of the lowly developed organs from which 

 the at present clearly useful organs have come by modifica- 

 tion. If it is difficult to determine the usefulness of an 

 organ in a living animal which we can study, how much 

 more difficult it must be to decide as to the usefulness of 

 an organ in an extinct animal, and the early stages in the 

 evolution of organs at present useful were generally passed 

 through in animals or plants of a kind no longer found 



