35(i W. B. Scott — Variations and Mutations. 



Aryan tongues. Doubtless the analogy may be pushed still 

 farther and we may say with some confidence that just as 

 sound principles of etymology were deduced by tracing the 

 changes of words step by step back from their modern forms 

 to their ancient origins, so the existing animal forms must be 

 traced back, noting all the gradations, to their far distant 

 ancestors. 



The great problems of morphology have been attacked by 

 the aid of many distinct methods, each of which has its own 

 drawbacks and its own particular advantages. Most of them 

 suffer from the fact that they deal only with the present order 

 of things and thus somewhat resemble the attempt to work 

 out the derivations of languages which have no literature or 

 only one that is so falsified and vitiated with forgeries that the 

 order of succession is hopelessly lost. Final results can be 

 hoped for only by a combination of all the methods, but such 

 combination becomes more and more difficult because of the 

 continual accumulation of immense numbers of facts, which 

 requires strenuous labor to keep track of, even in a single 

 department, not to mention the theoretical taint now attaching 

 to nearly all morphological work and which causes hesitation 

 in fully accepting the results of any inquiry. While the 

 immensity of the task and its manifold variety necessitate 

 specialization of research, too narrow a specialization of knowl- 

 edge is a great evil and many an investigator might save him- 

 self from serious blunders, could he but learn the results 

 gathered by fellow workers in somewhat different fields, results 

 which have a most important bearing upon his own. 



The principal methods of morphological inquiry have been 

 comparative anatomy, embryology and palaeontology, each with 

 the limitations and advantages peculiar to itself. Not long 

 ago, and to some extent still, embryology was looked upon as the 

 final arbiter in all morphological questions, but the many diffi- 

 culties in the way of applying the method to particular cases 

 and the lack of any generally accepted canons of interpreta- 

 tion, have led to a reaction against embryological deductions, 

 which perhaps undervalues this method as much as it was over- 

 estimated before. 



Comparative anatomy suffers from the drawback that it can- 

 not with certainty distinguish between resemblances due to 

 genetic affinity, on the one hand, and those which are the results 

 of convergence or parallelism, on the other, and it possesses no 

 trustworthy criterion, by which it can test the taxonomic sig- 

 nificance of structural characters. Besides, this method deals 

 only with the recent animal world, a mere disjointed fragment 

 of what has existed in times past. It is a language without a 

 literature to register its changes. 



