.s FLATFISHES (HETEROSOMATA) 



almost fNailly Miiiilar to that of a generalised Serranid. and tiie jiectoral arcli,' the 

 attachment of the pelvic fins and the form of the vertebral column are all quite 

 I'ercoiil. ICven more important than these characters themselves is the combination 

 of 24 vertebra- (10 pra-caudal. 14 caudal). 17 principal caudal rays (15 branched), and 

 pelvic lins of a spine and 5 soft rays, with the pelvic bones directly attached to the 

 cleithra. Such a combination is commonly found among the Percoid fishes, but 

 occurs in no other group- 

 There is another character, the importance of wliich does not appear to have been 

 generally realised, which may be of some value, if not in actually indicating lines of 

 descent among Bony Fishes, at least in confirming already suspected relationships. 

 This is the form of the otoliths (Fig. 8), which, by a comparison of fossil forms with 

 those from existing fishes, may be shown to have undergone comparatively little 

 change in course of time.^ The sagitta of Psetlodes* is seen at once to be quite 



. S.— Sagltt.i of A, !■ puifphdus moriu {■ 2i} ; 

 liiigualula { >. 3); i>, Hippuglossus hippngloisiis 

 t. Hippoghssouli-s plalessoidcs (x 3) ; t;. Plciii: 



irumci ( ■ .;); c, hucitharui 

 r.phthulmu^ rh,.iul)Hs ( • jj) ; 

 ■ 3) ; H. .s„(,'„ s„Ua{ ■ ^<), 



unlike that of the Gadoids, and e\en more unlike the curious sagitta; of the Zeoids. 

 It agrees closely, ho%vever, with those of tlie rercoids, resembling that of Perca in 

 shape and that of Centropomus in the form of the sulcus. " The otolith of Psettodes 

 irumei," writes Frost, " resembles in every feature those of the suborder Percoidea." 

 Moreover, the sagitta of Eiicitharus is also markedly percoid in form, those of Hippo- 

 glossiis and of the members of the family Botliida? {Arnoglossus, Bolhns, Lepidorhomhus, 

 Scophthalnuis) are described as sub-Percid, and those of the Pleuronectidx' and Soleidae 

 can, for the most part, be shown to have been derived from this type in spite of their 

 specialised features. These facts not only provide striking confirmation of the l^ercoid 

 ancestry of the Flatfishes, but also provide additional evidence lu f.ivour of their 

 being a homogeneous group. 



^ Chabanaud (1931, p. 395) states that there is i 

 I have examined two skeletons of Psettodes eritmei, . 



- See Kegan (1929, P- 324)- 



' A valuable comparative study lA th.- ntditli 

 by Frost, whose series of paprrs w.is published in 

 (1935-1930). 



* See I'rost (ni3". p. i^z. pi. i.x. Iig. :;. 



inlv 



ngle post-cleithr 

 In botii two close 



in I'sellodes bclcheri. 

 united Ijoncs (Fig. 3). 



lit Xeoptervgian tishes h.Ts been niadi 

 i.ils .uid M.ngazinc of Natural Historv 



