a new Lophioid Fish from Qreenland. 339 



principle that when in two allied species there are expressed 

 two ditferent " ideas"*, two independent thoughts of the crea- 

 tive power of nature (if I may so express myself), they should 

 be placed in different genera, or express the rule more practi- 

 cally thus, " si quaedam species ab aliis, quam maxime ipsi 

 affinibus, characteribus tamen ejusmodi differt, qui in aliis, ad 

 genus stabiliendum valent, non conjungenda est cum aliis, 

 sed generice distinguenda "f, we shall certainly recognize in 

 Oneirodes Esclirichtii a type different from Melanocetus John- 

 sonn, although very nearly allied to it ; and in order to have 

 something more definite to hold to, something that is not 

 merely a matter of more or less, we may appeal to the differ- 

 ences in the direction of the mouth, to the presence of the 

 peculiar (second) dorsal ray in the one species and its absence 

 in the other, and to the characteristic development of the 

 frontal ray in Oneirodes. On the other hand, it would seem 

 at present (probably as an immediate consequence of the 

 Darwinian ideas which are spreading so rapidly) that science 

 is passing through a reaction against a generic differentiation 

 which has been carried too far — a feeling with which I can 

 entirely sympathize (although I cannot altogether accept its 

 motive), simply because I must always see in the idea of the 

 genus an expression of a nature-thought, for which reason I 

 can by no means sympathize unconditionally with the modern 

 notion of the merely relative value of the idea of the genus. 

 The present case is one of those upon which opinions may be 

 divided. There are no fixed criteria as to which " characters" 

 have and which have not absolute validity as generic distinc- 

 tions : what experience proves to be good generic characters 

 in one family are of no value in another (" scias characterem 

 non constituere genus, sed genus characterem !") ; and we are 

 thus referred to a subjective, and therefore to a certain extent 

 less certain, judgment as to what is the right conception 

 in a given case. Now, as regards especially the relation 

 between Oneirodes and Melanocetus^ I am not blind to the fact 

 that there is so thoroughgoing a resemblance between them in 

 all the more essential features, that one might perhaps feel 

 hesitation about weakening the impression of their intimate al- 

 liance by placing them in different genera. Would it not be 

 very natural to include two such nearly allied forms under 

 the same generic name ? What would there be against species 

 within one genus of such abnormal fishes as the Lophioids 



* See Bnmner von Wattenwyl, ' Revue et Magasin de Zoologie,' 1870, 

 pp. 118, 119. " A genus is a divine idea " (E. Forbes). 



t Van del" Hoeven, ' Philosophia Zoologica ' (18G4), p. 27G. 



