FLOUNDERS OF GENUS PARALICHTHYS AND RELATED GENERA 



285 



to time, but their autlu'iitiiily was not checked. 

 All the mimeiouri si)ecimeiis exaiiiiiied lacked ac- 

 cessory scales. Of the American species, isosceles 

 lacks accessory scales and it is apparently a Pteu- 

 dor/ia»ibiis (p. 299) ; but in the species described 

 as Paralichthyx friocellatus bj' Miranda Rilwiro, 

 which is possibly also a Pseudorhambus, this chaj- 

 acter may vary with individual fish (p. 335). 



The accessory scales apjjear rather late in the 

 life of the fish and this character is probably a 

 late evolutionary acquisition. In P. dentatus and 

 /'. alhigutta they appear only after the fish has 

 attained a length of about 75-90 mm., and in these 

 two species they seem to develop at a smaller size 

 than in the others. In P. htasiUensis they seem to 

 develop first ii: specimens between 130 and 155 

 mm. On account of tlieir late appearance the 

 practical usefulness of this character is limited. 

 Tlie relative development of this character also 

 diifers much with the species. For instance, the 

 accessory scales are jjrofuse in large specimens of 

 dentatus and alhigutta, while in hrasiliemis they 

 are rather few in number. This character, there- 

 fore, is seemingly also not of transcendent impor- 

 tance in generic separation. 



This character was neglected by authors in gen- 

 eral and it is difficult to appraise its true value. 

 Norman (Monogr., p. 46, 1934) states: ". . . the 

 presence of supplementary scales . . . provide 

 features of taxonomic importance." He includes 

 tliis character in the definition of some genera but 

 fails to mention it in others. Later, in discussing 

 Paralichthys isosceles, Norman (Disc. Rept. vol. 

 1(), p. 135, 1937) states: "... I am not convinced 

 of the value of this character [the absence of sup- 

 plementary scales] in the definition of genera." 



However, irrespective of the value of this char- 

 acter in the family as a whole, it is evident tliat in 

 the species concerned it is of at least as much value 

 as the other character employed in the delimita- 

 tion of genera. Judged by the species studied by 

 me, it seems likely that it will prove to be of 

 greater value than the other characters for the 

 nui jor divisions of the species, in showing relation- 

 ship and in the separation of genera. Of course, a 

 final solution of the question must wait until tliis 

 character is determined in all the species involved. 

 its development with size, and its individual vari- 

 ability, especiall}' in triocellatiis which possibly 

 forms an exception. 



The foregoing consideration of the generic char- 

 acters makes it evident that the thi'ee genera as 

 now constituted are not sharply distinguished. 

 Witli the possible exception of the accessory scale 

 character, no other single character will delimit 

 any one of the three genera concerned. The de- 

 limitation of tlie genera de]iends rather on a com- 

 bination of characters and the lines drawn between 

 them are more or less arbitrary. As far as our 

 present knowledge of the morphology of the spe- 

 cies studied is concerned there are substantial 

 reasons for placing them in a single genus, Para- 

 lichthys, divisible into a munber of subgenera. 

 However, they evidently form groups of related 

 species and in view of the comparatively large 

 number of species involved, it is desii'able to split 

 them up into convenient genera. Another cogent 

 reason for adopting this course is that by doing 

 so the current nomenclature of the species will be 

 least disturbed. It is also possible that a further, 

 intensive study of the species will reveal satis- 

 factor}' internal charactei's to separate the genera. 



KEY TO AMERICAN SPECIES OF HIPPOGLOSSINA, PSEUDORHOMBUS, 



AND PARALICHTHYS ^ 



A. Accessory scales absent in large as well as small fish.' Anterior teeth only slightly or not enlarged, sometimes 

 moderately enlarged. 



B. Origin of dorsal behind anterior margin of eye: eyeball and orbit very large to moderately Large; interorbital reduced 

 to a mere ridge; the three characters occurring together. Accessory branch of lateral line rather poorly develojjed, not 

 reaching dorsal profile. Ctenoid scales either present on both sides or absent on both sides. Most prominent spots 

 either 4 or 6, depending on the subgenus, in two longitudinal rows, occupying nearly the same positions in all the species; 

 prepeduncular spot obsolescent or absent geniis Hippoglossina (p. 287) 



a. Orbit and eyeball strikingly large, 10.4 to 12.1 and 7.3 to 9.6, re.spectively, in large specimens. Origin of dorsal 

 usually over or nearly over middle of eye in large as well as in small sjieoimens. Preanal spine (first interhaemal) 

 usually well developed and visible externally. Typical color pattern with 6 prominent spots, incompletely ocel- 



3 S|K>cii'S of wliich no spocimons were examined are pla<-<'ti in hnickets. See also p. 334 for three specie.'! of doubtful relationship and position. 

 < Small specimens of all the species lack acci'ssory scales, have a narrow interorhitnl, a comparatively large eye, and the dorsal origin is more or le.ss be- 

 hind lh<' anterior niarcin of the eye. Consequently, this key should be used with care in placing small sitfciniens. 



