2 l-'l A ri'ISllKS (HI'.ri-'KOSOiMATA) 



iik-.i whuli Ik- lu-vi-r olabor.itcil AyasM/ (iS.)j. p, jOo) placed tlie Flathshes near the 

 Chirtoiluiitula' ami Scorpiilida', and Holt (181)4, p. 43S) hinted at an affinity with 

 such deep-bodied hshes as I'lattix or DascyUiis, or even with Zens. 



In looj. Houlenger (1902) and Thilo (lyoj). working independently, came to the 

 conclusion that the John Dories (Zeida?) were nearly related to the Flatfishes, and 

 Boulcnf^er associated the two groups with Amphistium, a fossil fish from t^c I'pper 

 ICocene, in a division of the Acanthopterygii to which he gave the name Zeorhombi. 

 Kegan (lyioB, p. 484) first drew attention to the perch-like characters of Psettodes, 

 which he regarded as the most generalised member of the Heterosomata and " simply 

 an asymmetrical I'ercoid " ; he suggested that the rest of the Flatfishes had arisen 

 from a form not very unlike Psettodes. Later, Regan (1913) placed the Heterosomata 

 as a specialised oftshoot from the order Fercomorphi, a position which it occupies in 

 his latest classification of the Bony Fishes (1929). In his most recent arrangement 

 of the Bonv Fishes. Jordan (1923, p. 167) still places the Heterosomata near the 



,<c<^ 



O'} 



7 ) 



(xj.; 



Anacanthini and Allotriognathi (Kibbon-fishes, etc.). but remarks that " the fiounders 

 and soles, having no spines and the ventral fins thoracic with an increased number 

 of rays, should not be placed far from the percomorphous series ". 



All the authors so far mentioned appear to regard the Flatfishes as a natural 

 group derived from a single stock, whether Gadoid, Zeoid or Percoid. Recently, 

 however, Kyle (igai, p. 118) has concluded that the origin of the group has been 

 polyphyletic. " With regard to Origin," he writes, " the conclusion is reached that 

 the Flat-fishes are not a homogeneous group. Sympluinis represents the earliest 

 origm, and has sprung from a stock which has given rise, amongst others, to the 

 Macrurids and Trachypterids. The Bothus type is related to the Psettida; ; the 

 Rhomboids have a near relation in Stromateoides, and Zeus is an advanced relative ; 

 the Fleuronectoids are distinct from both. Psettodes, the ' Percoid ', appears to 

 have sprung from a different line of evolution, and is a modern accession to the ranks 

 of Flat-fishes." The evidence on which these conclusions are based, however, is at 

 times somewhat difticult to follow, and does not always appear convincing. 



As will be shown in the section devoted to classification, there seem to he good 

 reasons for regarding the Heterosomata as a homogeneous group, of which Psettodes 

 (Fig. I) IS the least specialised member, with the proviso that a more detaileil study 



