284 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



gradation than in the eastern species, and fiirtlier- 

 more, in the case of small specimens of the west 

 coast species, there are no characters which may 

 be correlated with the fin ray count, the number of 

 gill rakers and scales and the color being nearly 

 alike in those two species. 



GENERIC LIMITS 



External characters altogether satisfactory for 

 the division of the species treated in this paper 

 into major groups or genera have not been pro- 

 posed hitherto. None were elaborated during the 

 present investigation, except one which although 

 not entirely satisfactory is apparently more so 

 than those hitherto proposed. These characters 

 are discussed in reverse order of their apparent 

 importance. 



The presence of an anterior accessory brancli of 

 the lateral line has been used for distinguishing 

 Psendorhombus; but this is the least important of 

 all characters and is of very little usefulness in 

 generic division. This character and the num- 

 ber of vertebrae are discussed at greater length 

 on page 298. 



The structure of the scales, cycloid or ctenoid, is 

 of some moderate use. The known species which 

 apparently belong to ParaUehfhys either have all 

 scales cycloid or when ctenoid scales are present 

 they are typically confined to the eyed side. Some- 

 times very few ctenoid scales are present on the 

 blind side as a rather infrequent individual vari- 

 ation. If a species typically does have ctenoid 

 scales on the blind side, it is highly probable that 

 it does not belong to Parcdichthys. In the known 

 species of Hippoglonslna ctenoid scales are either 

 present on both sides or absent on both sides. In 

 the known species of Pseudorhombus the scales 

 are ctenoid on the eyed side and ctenoid or cycloid 

 on the blind side. This character is useful for sub- 

 generic division. The species of Paralichthys may 

 be divided into two apparently natural groups liy 

 the presence or absence of ctenoid scales on the 

 eyed side. In Pseudorhombus the same difference 

 on the blind side may possibly be used for sub- 

 generic division. In one subgenus of Hippoglos- 

 sina, however, this difference is only of specific 

 importance. 



The structure of the scales is very constant in- 

 traspecifically with two exceptions. In H. ohJomja 

 the number of ctenoid scales differs greatly with 



the individual (p. 294). In P. 'aeshiarhis the 

 ctenoid scales lose their spinules with growth and 

 all scales are cycloid in large specimens. How- 

 ever, in spite of the relative constancy of this 

 character it is of limited use for generic division. 

 It is evident that the disappearance of ctenoid 

 scales occurred independently in all three genera. 

 (The presence of cycloid scales appears to I'epre- 

 sent a more recent development in the species 

 concerned.) Consequently, to use this character 

 by itself for the major division of the species into 

 genera would run counter to their natural rela- 

 tionship and lead to the formation of polyphyletic 

 genera. 



The size of the eye and the interorbital width 

 are of importance in sej>arating the species placed 

 in H'rppogJossina, but the transition between ex- 

 treme species in these respects is rather gradual. 

 The same is true of the size of the teeth. The 

 species placed in Paralichthys have markedly long, 

 fanglike teeth, while those placed in Hippoglos- 

 shui have rather small teeth; but here also there 

 is a gradual transition, some of the sjDecies placed 

 in Pseudorhombus having the teeth intermediate 

 in size. No sharp lines may be drawn between the 

 major groups, or genera, on the basis of these 

 three characters. 



The position of the dorsal origin is of consider- 

 able importance. In the species of HippogJossina 

 the dorsal begins approximately over the middle 

 of the eye, while in nearly all other species it begins 

 over the anterior margin of the eye or a little 

 more forward. However, this character does not 

 separate all the species. In microps (p. 301) 

 which, judged by other characters, apparently 

 belongs to Paralichthys, the dorsal origin is over 

 about the middle of the eye. 



The presence or absence of accessory scales was 

 found to be a good criterion for the generic sep- 

 aration of the species concerned. All the species 

 of Hippoglossina examined lack accessory scales. 

 All those of Paralichthys have such scales. (They 

 are very few in squamiltenfus, p. 333). The fol- 

 lowing exotic species of Pseudorhombus, labeled 

 as such or under their synonyms in the National 

 Museum, were examined for this character, 

 namely, a7\sius, javanicus, jeiujnsn, pentophthal- 

 mus, cirmamoneus, oligodon and oligolepis (Nor- 

 man, Monogr., 1934). These identifications were 

 made by a number of previous workers from time 



