368 THE EVOLUTION OF LIFE. 



cases to modes neither higher nor lower. Of two connate races 

 which diverged in the remote past, the one may have had 

 descendants that have remained tolerably constant in theii 

 habits, while the other may have had descendants that have 

 passed through widely- aberrant modes of life ; and yet some 

 of these last may have eventually taken to modes of life like 

 those of the divergent races derived from the same stock. 

 And if the metamorphoses of embryos, indicate, in a general 

 way, the changes of structure undergone by ancestors ; then, 

 the later embryologic changes of such two allied races, will 

 be somewhat different, though they may end in very similar 

 forms. An illustration will make this clear. Mr Darwin 

 says : " Petrels are the most aerial and oceanic of birds, but 

 in the quiet sounds of Tierra del Fuego, the Puffinuria 

 berardi, in its general habits, in its astonishing power of 

 diving, its manner of swimming, and of flying when un- 

 willingly it takes flight, would be mistaken by any one for 

 an auk or grebe ; nevertheless, it is essentially a petrel, but 

 with many parts of its organization profoundly modified." 

 Now if we suppose these grebe-like habits to be continued 

 through a long epoch, the petrel-form, to be still more ob- 

 scured, and the approximation to the grebe- form still closer ; 

 it is manifest that while the chicks of the grebe and the 

 Puffinuria will, during their early stages of development, 

 display that likeness involved by their common derivation 

 from some early type of bird, the chick of the Puffinuria 

 will eventually begin to show deviations, representative of 

 the ancestral petrel-structure, and will afterwards begin to 

 lose these distinctions, and assume the grebe- structure. 



Hence, remembering the perpetual intrusions of organisms 

 on one another's modes of life, often widely different ; and 

 remembering that these intrusions have been going on from 

 the beginning ; we shall be prepared to find that the general 

 law of embryologic parallelism, is qualified by irregularities 

 that are mostly small, in many cases considerable, and 



