94 OLD RED SANDSTONE FISHES. 



longitudinal angle along which the lateral plates are bent, as indicating the 

 remains of an obsolete lateral canal. Here I must frankly confess that I see 

 nothing at all suggestive that this line, a mere " Kante " with some longitudinal 

 arrangement of the tubercles, has anything whatever to do with the lateral line 

 system. But to my mind the similarity in character between the cephalic lateral 

 line grooves and that in the lateral body-plates of each side and their demon- 

 strable continuity in the lateral occipital region quite settles the question. 



In a paper on Pterichthys, published in 1891 by Dr. J. V. Rohon, 1 a peculiar 

 structure is mentioned in connection with a specimen of PL Bhenanus (Beyr.) from 

 the Middle Devonian of the Eifel. This appears in the form of a horizontal 

 osseous septum placed below the hollow of the posterior median dorsal plate, and 

 thus, as it were, dividing off an upper chamber from the rest of the cavity of the 

 body-carapace. I have not seen the specimen myself, though Dr. Smith 

 Woodward, who has, assures me that Dr. Rohon's figures of the appearance in 

 question are quite correct. I must therefore content myself with stating that such 

 a structure has not yet been detected in any British specimen of the genus. 



I must now add a few words as to the validity of the generic name Pterichthys 

 for the fishes now under consideration. This name, given by Agassiz to specimens 

 collected by Hugh Miller at Cromarty, was, as already mentioned, first published 

 by Murchison in 1840, and has been in general use among geologists and 

 palaeontologists for this genus up to the present time. It is therefore a matter of 

 considerable regret that it was not noticed from the beginning, that Swainson 2 had 

 already given the same generic name to a fish of the family Scorpsenidse, namely, 

 the Apistus alatus of Cuvier. For, according to the strict rules of biological 

 nomenclature, a generic name once applied to a particular type cannot be passed 

 on to any other; and so, in 1859, Bleeker 3 proposed to cancel Pterichthys of 

 Agassiz, and to substitute for it the modified term Pterichthyodes. 



Apparently unaware of Bleeker's action, Mr. S. A. Miller, of Cincinnati, 

 proposed in 1893 the name Millerichthys, "in honour of Hugh Miller, who was 

 really the first to fully characterise and illustrate the genus in his Old Red Sand- 

 stone." * However, Dr. 0. P. Hay of New York decides that " Millerichthys " is 



1 TJeber Pterichthys. ' Verliandl. russ-kaiserl. mineralog. Gesellschaft. St.Petersb.' (2), vol. xxviii, 

 pp. 292—316. As pointed out by Jaekel, op. cit., p. 46, Dr. Rohon has confused the anterior with the 

 posterior aspect of his fossil. 



2 ' Natural History of Pishes, Amphibia, and Reptiles, or Monocardian Animals,' London, 1839, 

 vol. ii, p. 65. 



3 ' Enumeratio specierum Piscium hujusque in Archipelago Iudico observatarum,' prefaced by 

 a general classification of fishes, entitled ' Systematis piscium naturalis Tentamen.' ' Verhandelingen 

 der Naturkundige Vereeniging in Nederlandsch Indie,' Daal vi, 1859. Pterichthyodes is proposed 

 for Pterichthys, Agassiz, at p. xxi of the ' Tentamen.' 



4 ' North American Geology and Palaeontology,' First Appendix, Cincinnati, 1892, p. 716. 



