422 MR. E. W. L. HOLT STUDIES IN [MttjM, 



abdominal cavity of a flat-fish, we find that the peritoneum of the 

 ocular side is beset with dark pigment, while that of the blind side 

 is not. It seems clear enough that whatever cause governs the 

 presence or absence of external pigment in these fishes is 

 equally potent with regard to the dark pigment of the internal 

 region l . If this cause be the light, we have in the condition 

 of round-fishes an apparent anomaly, not in the absence of 

 pigment from the peritoneum in littoral fishes, since the difference 

 of position involves a much greater interruption of the light from 

 above, but in its conspicuous development in the deep-sea fishes, 

 the peritoneum of which can be in no way subject to any action 

 of the sun's light. 



In conclusion it may be remarked that M. dbyssorwm is a form 

 such as might well be descended from a parent stock not greatly 

 differing from the Common Ling of the present day, and that the 

 extensive vertical range of that species Mould appear to offer 

 facilities for the establishment of such an abysmal offshoot -. 



II. On the Kecessus orbitalis, an Accessory Visual Organ 

 in Pleuronecttd Fishes. 



In making an examination of the cephalic anatomy of the 



1 Since this was written, I find that Messrs. Cunningham and MacMunn, in 

 their monograph "On the Coloration of the Skins of Fishes" (Phil. Trans. 

 1894, p. 809), consider that the " difference can only be explained as the effect 

 of light falling on the upper side of the fish, and not on the lower." I omit 

 further reference to their results, since they are discussed at some length in 

 No. II. of this series of papers. 



2 It is interesting to find that Professor Smitt (op. cit. p. 525) arrives at 

 a precisely opposite conclusion, viz., that " from a systematic and genetic point 

 of view, M. dipterygia must be regarded essentially as a predecessor of the 

 Common Ling, though the former has evidently adopted in certain respects a 

 distinct direction of development from the common original type which we 

 are entitled to assume." " This," he remarks, appears from an appended table 

 of averages of certain measurements in the two species. No further ex- 

 planation being forthcoming, I am unable to say in what way the table may 

 be supposed to support his conclusions, nor am I inclined to attach any great 

 value to a series of averages based in no case on more than three specimens. 



The right of the author to form any opinion he chooses, as to the relative 

 antiquity of the types of structure exhibited by the two species, is undoubted ; 

 but it seems impossible to limit its application, and the logical conclusion is, 

 that if Smitt holds this opinion in the present instance, he also regards all 

 Gadoids which exhibit an abysmal type of structure as more primitive than 

 their littoral brethren. This view is, of course, in direct opposition to that 

 held, as I suppose, by all other ichthyologists, viz., that the abysmal forms are 

 descended from littoral ancestors. 



Elsewhere (p. 520) Smitt notes that the trace of division in the second 

 dorsal fin is more marked in the Birkelange than in the Common Ling, but it 

 does not appear that this was taken into consideration in formulating the 

 conclusion quoted above. To most ichthyologists it would appear as rather 

 important evidence as to the greater antiquity of the type exhibited by the 

 commoner species. Again, when dealing with Gadus csmarkii, the author 

 considers that the persistence of the barbel brings that species nearer to the 

 common origin of the Cods than the Pollack, which has no barbel (p. 499), 

 so that the same facts would appear, in the hands of Professor Smitt, to 

 be capable of diametrically opposite interpretations. 



