90 EVOLUTION OF LIVING ORGANISMS. 



remembered that organs rarely if ever can be said to 

 "begin." Entirely new functions and entirely new 

 organs are not suddenly developed. All are evolved by 

 the gradual transformation of, addition to, or subtrac- 

 tion from something already there. The wing of a bird, 

 unique as it is, has had no sudden beginning it has been 

 gradually transformed from the fore-limb of the reptile- 

 The extinct Archaeopteryx from the Jurassic strata has a 

 modified limb in a beautifully intermediate state. The 

 one-toed hoof of the horse is not a new organ ; it is derived 

 from an ordinary five-toed foot by the gradual loss of the 

 lateral digits. Almost every stage of its history can be 

 traced in the fossils. What more complex or useful 

 organ than the human eye ? Yet it is merely an instance 

 of the extreme specialisation of the property of response 

 to light generally distributed over the surface of the body 

 in the lowest forms. In the evolution of an organ by 

 selection every stage must be useful, and it is often diffi- 

 cult to picture the intermediate conditions ; but we 

 must not jump to the conclusion that they could not have 

 existed. The heart of the amphibian has one ventricle, 

 in which the venous arid arterial blood become more or 

 less mixed ; that of the bird has the ventricle completely 

 subdivided into two chambers, so that the venous is kept 

 quite separate from the pure arterial blood. Now if 

 reptiles were unknown, we could well imagine an op- 

 ponent of the theory of natural selection stating dog- 

 matically that the intermediate steps in the formation 

 of the dividing septum could not have been useful, and 

 therefore could not have been selected. To be effective 

 at all, he would say, the septum must be complete from 

 the beginning ; if the venous is to be separated from the 

 arterial blood an intermediate stage would be of no use. 

 Such arguments are constantly heard. Fortunately, 

 in this case, we can point to the reptilia, where these 

 very intermediate steps occur, and the incomplete 

 septum can be shown actually at work. 



