FISHES OF LAKE SUPERIOR. 249 



Petromyzontid^ (Lampre^'-eels.) 



There are families in all departments of nature, whose peculiarities 

 call for an investigation of their more general relations rather than of 

 their structural details. The Petromyzons are in this case. Closely 

 aUied together and circumscribed in a most natural family, it is a ques- 

 tion whether they should be entirely separated from all other fishes 

 to form a great group by themselves, or whether they belong to one 

 of those great divisions in which the individual members diifer widely 

 from each other. In other words, should the Petromyzons stand by 

 themselves in a natural classification of fishes, as Prince Canino and 

 Joh. Miiller have placed them, or shall we combine them with skates 

 and sharks, as Cuvier has done ? To answer such a question, it is 

 necessary to discuss beforehand principles of the utmost importance 

 in the study of natural history, and above all to settle the follow- 

 ing difficulty : — Is the study of anatomical structure an absolutely 

 safe guide in the estimation of the relations of animals to each other ? 

 Cuvier, who made the study of comparative anatomy the foundation 

 of classification, carried out this principle in a most remarkable man- 

 ner, and improved the natural arrangement of animals most sur- 

 prisingly ; indeed, he made zoology truly a science by it ; but with 

 a tact that characterizes genius, he limited the absolute consequences 

 of this law by a true appreciation of the relative value of characters ; 

 introducing at the same time with the principle of classification ac- 

 cording to the structure of animals, that of subordination of charac- 

 ters, without which the first great principle might mislead us, instead 

 of helping to ascertain the true relations of organized beings. Now 

 it seems to me as if zoologists and anatomists had of late insisted too 

 strictly upon the absolute differences which exist between animals, 

 instead of attempting to appreciate the relative value of the differ- 

 ences noticed. Of course, as this latter point rests almost within the 

 limits of individual appreciation, it is more difficult to find the right 

 path here, than in almost any other department of zoological investi- 

 gations ; but I hope to be able to introduce another great principle 

 of zoological classification, which shall afford a safe guide to settle 

 such doubts ; I mean the study of embryonic development. 



Let me now show, in the present instance, how I consider it possible 

 17 



