21 8 EVOLUTION OF TO-DAY. 



will always be a comparatively narrow land, and can 

 at best only support a few individuals, while the 

 larger areas can support many. We know that 

 under these circumstances, the numerous species 

 would replace the less numerous variety. For in- 

 stance, if one species inhabited the low lands, and 

 another the mountains, the intermediate land will 

 be only a narrow strip at the base of the mountains. 

 It is evident that this narrow strip would not sup- 

 port a separate species, but would rather be in- 

 habited by stragglers from the two species living on 

 either side of it, and the two species would thus 

 overlap. Thus, it is easily understood that inter- 

 mediate forms will tend to be exterminated. 



But this does not alter the fact that these transi- 

 tional forms must have at one time existed, and 

 should, therefore, be found in the remains of animals 

 living in the past. We have already discussed this 

 question, and have seen that no such numerous 

 transitional links exist in fossils. The reason for 

 this has been found in the extreme imperfection of 

 the geological record. It is, however, very ques- 

 tionable whether we are justified in considering this 

 imperfection sufficient to account for the almost 

 utter absence of these intermediate links. In one 

 or two very exceptional cases the requisite amount 

 of intermediate forms is found, but, as an almost 

 universal rule, this is not true. The history of the 

 horse family is about as complete as any history yet 

 discovered, but even here it is impossible to discover 

 the numerous links which we would expect. For, 

 while we do find a number of intermediate stages in 



