Benedenius deneensis, Traquair. 449 



scales become more equilateral towards the dorsal and ventral 

 lines and towards the caudal extremity ; close to the ridge- 

 scutes they assume a more rounded shape and concentric 

 striation. I am unable to detect a lateral line. 



The inner surface of the scales is seen in specimen B 

 (PI. X. fig. 3) to be extremely similar to that of EurynotuSj 

 viz. with a strong keel parallel with and near the anterior 

 border, and a point at the upper end the length of which is 

 about half that of the largest scales. 



In specimen A, as it lies on the right side, the ventral 

 scutes anterior to the ventral fins scarcely appear, being 

 covered with small scales forming oblique series directed 

 forwards ; the two or three front ones, which are partially 

 exposed, appear to be directed backwards, but this is probably 

 due to crushing, what is seen on the outline of the pectoral 

 region being the right branch of the V which is formed by 

 these scutes. 1 must regard the arrangement described and 

 figured by Van Beneden and Traquair as the natural one, 

 and am therefore quite unable to account for the position 

 these scutes assume in Fraipont^s specimen, which has led to 

 the interpretation that they formed a single series, each scute 

 being disposed transversely and perpendicular to the sides. 

 On the contrary, these elements, as shown isolated in speci- 

 men B (PI. X. fig. 4), were V-shaped pieces, as are known in 

 many Ganoids, and formed of two distinct halves joined on 

 the median line. Seen facing, they are thickest at the line of 

 meeting of the branches of the V, which taper to a point. 



As I have stated above, the dorsal ridge-scutes extend 

 forwards to the occiput ; 33 can be counted in front of the 

 dorsal in specimen A, and 23 are preserved in regular succes- 

 sion in specimen B. Each of them corresponded to an 

 interneural bone, which is to be detected in relief under the 

 scales in specimen A. 



The result of this examination entirely confirms Prof. Tra- 

 quair's original conclusions, viz, that Benedenius is closely 

 allied to Eurynotas, Wardichthys, and Mesolepis. Until we 

 know something definite of the dentition, we may hold that 

 the only fundamental diflference between Benedenius and 

 Eurynotus resides in the lesser elongation of the dorsal tin 

 and the larger ridge-scutes in the former, characters shared 

 by Mesolepis and the imperfectly known genus Wardichthys^ 

 allied fishes with very deep body. 1 cannot understand what 

 could have induced Prof. Traquair to modify his views in 1890, 

 as appears from a note in these ' Annals ' (ser. 6, vol. vi. 

 p. 492), to the effect that the author has " become convinced 

 that, though it presents many resemblances to the Platysomidae, 



