4(' FLATFISHES (HETEROSOMATA) 



which normally covers it as a post-mortem condition. Nevertheless, if the appearance 

 of the outer point cannot always be relied upon, the form of the spine in the difierent 

 species may be a useful character. 



Among other purely external characters, the form of the scales on the two sides 

 of the body, their size, their transformation into tubercles or membranous processes, 

 the presence of supplementary scales, etc., as well as the structure and course of the 

 main branches of the lateral line, all provide features of taxonomic importance. In 

 Psettodes and in most genera of Pleuronectidae the main lateral line is well developed 

 on both sides of the body, but in most of the Bothids this is either feebly developed or 

 absent altogether on the blind side. The presence or absence of a supratemporal 

 branch, running upwards and forwards in the region of the nape, is another useful 

 character. A remarkable modification of this branch is found in certain genera of 

 Pleuronectina", curiously enough all occurring in the North Pacific. Here the supra- 

 temporal branch runs forward from the anterior part of the main lateral line as usual, 

 but It then forks into two branches, a short one clirected anteriorly and a more or less 

 lengthy posterior one which runs just below the dorsal fin (Figs. 234-245). In Lepidop- 

 selta this posterior branch is not very elongate, and in other respects the genus is 

 exactly like Litnanda, to which it is closely related ; in Inopsetta it is very short indeed. 

 In their key to the genera of Pleuronectinje, Jordan and Evermann (1898, p. 2607) 

 have made use of this character to separate two groups of primarj' importance, but it 

 would seem to be of less value than they supposed. In the genus Platichthys the 

 supratemporal branch is not normally provided with this postenor prolongation, but 

 there is a Flounder (P. flesus) in the collection of the British Museum in which this 

 firanch is forked and the posterior prolongation is of fair length. 



Kyle (1900B, p. 34b) has discussed the value of the alimentary canal as a taxonomic 

 character at some length, but concludes that this marks " a specialisation in structure 

 and habits and cannot be used in classification." Examination of the abdominal 

 cavity in a large number of species, however, has convinced me that, like the lower 

 pharyngeals, the form of the alimentan,' canal provides a useful character for dis- 

 tinguishing the species in the more specialised genera of small-mouthed Pleuronectina, 

 and may even be used for the definition of genera or higher groups. The two most 

 important features are the intestinal tract (referred to througliout as the intestine) — 

 its length and the manner in which it is coiled in the abdominal cavity — and the 

 ' ' pylonc ' ' appendages — their presence cir absence, size and number. These appendages 

 may be grouped round the junction between those parts of the intestinal tract usually 

 referred to as the stomach and duodenum, or there may be two or more in this situation 

 with others placed further down the intestine. The group of Pleuronectine genera 

 including Microstomus, Embassickthys, Tanakiiis and Ghptocephalus have generally 

 been marked off from the remainder of the small-mouthed members of the subfamily 

 as a primary- division, distinguished by a generally more elongate body and by an 

 increased number of vertebra-, fin-rays and of scales in a longitudinal series. Such an 

 arrangement is clearly an artificial one, and it is doubtful whether these genera really 

 form a natural group. Microstomus and Embassickthys appear to have been derived 

 from some form very like Pseiidopleuronectes, and it is possible that Tanakius is most 

 nearly related to Dexistes. It is of some interest to find that the general elongation 

 of the body, with the corresponding increase in the number of vertebra? and fin-rays, 

 has been accompanieii in all four genera by a striking modification of the alimentary 

 canal. This is usually elongate, and the second coil extends into the secondary body- 

 cavity of the ocular side. In other genera (e.g. Plcuroncctes] the intestine may project 

 shghtly into the secondary body-cavity, but in all the above-mentioned genera the 

 second coil, which lies close to the reproductive organ of that side, runs well backwards 

 alongside the interhamal spines on the ocular side. A somewhat similar condition is 

 found in the Soleidse, in which two to four coils of the intestine may enter the body- 

 cavity of the ocular side 



Finally, perhaps the commonest and most widely-used characters for the distinction 

 of species of Flatfishes are the numerical ones — the number of rays in the dorsal, anal. 



