THE THEORY OF NATURAL SELECTION. 297 



forms must have greatly exceeded in total number the 

 combined number of individuals which are clearly 

 recognisable as distinct species, we ought to find more 

 abundant evidence of their existence than of the exist- 

 ence of the separate species. As a matter of fact, how- 

 ever, the study of extinct animals does not afford more 

 than very incomplete evidence as to the existence of 

 the numerous and closely graduated transitional forms 

 required by the theory of natural selection. It is true 

 that palaeontology has brought to light many forms of 

 animals which are distinctly intermediate in their char- 

 acters between groups which would otherwise stand far 

 apart. Thus, we have numerous extinct types which 

 bridge over the gap between the reptiles and the birds ; 

 and others which stand intermediate between the exist- 

 ing horses and their original five-toed ancestors. So far, 

 then, palaeontology unquestionably lends support to the 

 general theory of the evolution of species from pre-exist- 

 ing species. The theory of natural selection requires, 

 however, more than this. It requires that there should 

 be a series of intermediate types graduating into one 

 another by slight and hardly perceptible differences. In 

 some cases, as regards allied species of animals, such a 

 continuously graduated series can be shown to exist (in 

 some extinct Shell-fish, for example). In most cases, 

 however, it must be admitted that palaeontology has so 

 far failed to demonstrate the past existence of the 

 numerous and finely-graduated series of transitional forms 

 between different species absolutely demanded by the 

 theory of natural selection. Such transitional forms as 

 are known for the most part stand quite sharply distin- 



