40 ORGANIC El'OLUTION 



recent species there is a progressive series of modifications 

 of one or more organs, showing that there has been a grad- 

 ual, steady change in a particular direction, the several steps 

 in this change being very slight. In the fossil remains 

 which give us the history of the evolution of the horse 

 (Plates 46 and 47) we see the gradual loss of the outer toes, 

 and a corresponding increase in size of the middle toe, a 

 gradual increase in length of the molar teeth, and a gradually 

 increasing complexity of the ridges on their grinding sur- 

 faces. It has been claimed that the several steps in these 

 modifications are not of enough importance to have given 

 their possessors decided advantage in the struggle for exist- 

 ence, and that their progressive development in these par- 

 ticular directions must indicate an inherent tendency to 

 become modified in these directions. If this progressive 

 modification in the ancestors of the horse be due to some 

 inherent tendency rather than to natural selection acting 

 on a great number of all sorts of variations, selecting only 

 the useful ones, then this casts doubt on the importance of 

 the role of natural selection in other cases. May not much 

 of the evolution of which we have evidence be due to similar, 

 not understood, inherent tendencies ? 



The last and by far the most important objection, which 

 we will mention, to the idea of evolution by means of natural 

 selection is this : It is well known, of course, that, in general, 

 the offspring of any pair of parents tend to be somewhat 

 intermediate in character between the two parents. 1 Now if 



1 This statement needs slight modification, as will appear later when we come 

 to the mention of Mendel's laws in their relation to the persistence of variations 

 (page 44). 



