13 



indicating the grouping of the various species. But subsequent authors frequently 

 raised these divisions to the rank of distinct genera. Bonaparte, 1841, separated the 

 last subdivision of Cuvier only as a distinct genus, namely, Plagusia. 



Dr. Gimther, in his important work, " The British Museum Catalogue of Fishes," 

 gives a comprehensive classification of the Pleuronectidae, which includes all the 

 forms up to that time described in the literature or represented in the great national 

 collection. He distinguishes in all thirty-four genera, many of which are entirely 

 new, while the limits and definitions of the rest are revised. His definition of Solea 

 is as follows : 



Eyes on the right side, the upper being more or less in advance of the lower. Cleft 

 of the mouth narrow, twisted round to the left side. Teeth on the blind side only, 

 where they are villiform, forming bands : no vomerine or palatine teeth. The dorsal 

 fin commences on the snout, and is not confluent with the caudal. Scales very small ; 

 ctenoid. Lateral line straight. 



Thus the chief difference between this definition and Cuvier's is that it excludes all 

 the forms in which the longitudinal fins are continuous with the caudal. All the British 

 forms of sole described in the present work are included by Giiiither in the genus Solea. 

 But almost every author defines the genera of Pleuronectidae, even at the present day, 

 in some degree after his own fashion. There is a general agreement, together with 

 differences of opinion on certain points. In the Pleuronectidae, as in nearly all families 

 of animals, there are certain well-marked species which are recognised by all naturalists. 

 There are others, especially those founded upon a small number of specimens, which 

 are more difficult to separate, and concerning which differences of opinion exist. But 

 the question of the arrangement of the species into genera gives rise to much greater 

 differences of opinion. The arrangement of course depends on which of the characters 

 possessed by several species in common, are taken as characterising a genus. If one 

 character is taken, a certain number of species are united by it ; if another is taken 

 the same species are scattered among other genera. Thus the fish called in this work 

 Arnoglossiis mtgastoma is variously placed. By Giinther it is placed in the genus 

 Rho?nbu$, because it possesses teeth on the vomer. By Moreau, whose arrangement 

 seems to me more original than natural, it is placed in the genus Pleuronectes. The 

 characters in which the species agrees with Ai*noglossus laterna s'eem to me to be more 

 numerous and important than the presence or absence of vomerine teeth, and I have 

 therefore followed those authors who include it in the genus Arnoglossus. 



In describing the British species I shall give not merely the range of variation 

 of each numerical character, but the actual numbers observed in several individuals. 

 It must be pointed out that the number of scales in the lateral line is always 

 obtained not by counting the scales in the line itself, but the series of oblique trans- 

 verse rows of scales which cross the lateral line ; the scales are so arranged as to 

 form rows which run obliquely downwards somewhat from before backwards, other 

 rows which run more obliquely downwards and from behind forwards, and others, 



