212 THE ROYAL SOCIETY OF CANADA 



with each other arise quite independently of each other, and widely 

 separated in space and time. Not only so, but these substances 

 which are indistinguishable from each other may originate in widely 

 dissimilar fashion. The same has been shown to hold true for organic 

 substances, as witness the numerous ways in which urea can be 

 synthesized from inorganic sources as well as in living organisms. 



Apparently the only good reason for believing that this is not 

 also true of organismes themselves is that they are comparatively so 

 complex and so rare. The more complex a kind of thing is, and the 

 less frequently it occurs, the more reasonable it is to suppose that the 

 conditions necessary to produce it have obtained only once, and that 

 consequently all of that kind are connected in origin. 



Are we altogether justified in making this supposition? As to 

 the actual origin of entirely new forms v/e have not yet the necessary 

 knowledge to state that a certain new form has arisen repeatedly at 

 different times and places. In the origin of new forms by the crossing 

 of common species, we are able to affirm that the same kind has been 

 produced on many occasions, and the study of genetics has shown us 

 how these new kinds can be produced at will. 



In systematic study there has arisen a large body of facts that 

 supports strongly the view that the same species has arisen repeatedly 

 and even by different paths. When almost identically the same form 

 occurs at places remote from each other, those who are committed 

 to the view that a species has been produced but once, predicate as a 

 matter of course some transference of the species from one place to 

 the other, and in this they are justified, for the spreading of a species 

 is a phenomenon readily demonstrable, and frequently observed, 

 whereas the production of a new species has so far been a doubtful 

 thing. 



However, the phenomena of parallelism afford cases that admit 

 of no other interpretation than that frequently tlie same evolutionary 

 process has been repeated. Examples of intergeneric parallelism are 

 not infrequent among Ascidians as well as in other groups of animals, 

 and a hypothetical case may be used as an illustration. In two 

 closely allied genera, X and Y, there are parallel series of species, Xa, 

 Xb and Xc, and Ya, Yb and Yc, where Xa and Ya are similar except 

 in generic characters, and the same for the other pairs Xb and Yb, 

 Xc and Yc. If the classification correctly interprets the phylogeny 

 or origin of these forms, the following arrangement would approximate 

 their ancestral history. 



