Translated hy Philip H. Nicklin. 113 



one exclusively, is entirely wanting, and will perhaps always be 

 so, if I am right in classing among impossibilities, the only means 

 of bringing about such assent. 



What is to be done, then, in the midst of all these difficulties ? 

 How shall we escape the horns of this dilemma; reform is want- 

 ed, hut permanent reform is at this time impossible ; nevertheless, 

 on the other hand, it is absolutely necessary to classify the objects 

 already knoivn, and those which are discovered every day. There 

 is nothing but provisional arrangement, nothing but system, (and 

 system is artificial method,) which can rid us of this embarrass- 

 ment. Let us, then, consult theory, experience, analogy, that 

 we may labor to come at the truth. 



Theory says, that there are essential diflferences between genus 

 and genus, between species and species. 



Experience tells us, that dissimilar animals are sometimes found 

 in similar shells, and animals almost identical in shells, apparently 

 very different. 



Analogy proposes, under these premises, to draw inductions 

 from the mass of discoveries already made. 



Let us take then this immense group, such as MM. de Blain- 

 ville and Deshayes understood it a few years ago ; this group, 

 whose animals they then believed to be generically identical; let 

 us begin by lopping off, finally, with M. Deshayes, the Iridina, 

 with M. D'Orbigny the Castalice, of which the shell alone did 

 not permit a rational distinction. Then let us reduce, with Mr. 

 Lea, all the other genera of the Naiades* to the simple im- 

 portance of subgenera or sections, waiting always to withdraw 

 from these, all those species whose anatomical characters when 

 known, shall be found to differ from those of the studied type, in 

 like manner as the Iridina and Castalice, and we shall have a 

 genus which will possess only a supposed unity in its essential 

 characters, but which on that very account will be conditionally, 

 and therefore theoretically, exact. 



Arriving at species, we would proclaim as irrevocably autono- 

 mous, those which offered no specific anatomical differences, such 

 as the form of the ovaries studied by Mr. Lea in the Uiiio irro- 

 ratus, ochraceus, cariosus, retusus, in the Anodonta undulata and 



* I say nothing here of the genus Mijcetojwda of M. d'Orbigny, which I have 

 not at present an opportunity of studying. Mr. Lea does not admit it, any more 

 than that of Castalia. 



Vol. xLi, No. 1.— April-June, 1841. 15 



