206 MR. E. T. NEWTON ON SOME [Apr. 2, 



ornamentation is peculiar and exactly corresponds with that found 

 on the pectoral plate originally described by Dixon. The tubercles 

 on all these specimens are sharp, conical, and more or less connected 

 together by ridges which give a reticulated appearance to the plates 

 when closely examined. Spines which cannot be distinguished from 

 those of A. egertoni are found at Barton ; but the skull above de- 

 scribed differs from A. egertoni in several particulars, altbough it is 

 only the supraoccipital bone which is available for comparison. 

 This bone, so far as it is preserved, has a different form from that of 

 A. egertoni, being proportionally wider and with less deeply grooved 

 mucus-canals. The ornamentation of the bones, likewise, is of 

 another character ; the tubercles are more numerous, more rounded, 

 and with little or no indication of the reticular structure between 

 them ; moveover, they have a greater tendency to run together in 

 radiating lines, and to become less distinct towards the margins of 

 the bones. It will be obvious that this skull cannot be referred to 

 A, egertoni, and there seems no good grounds for referring it to 

 A. ? bartonensis, which is a smaller form and not certainly belong- 

 ing to the genus Arius. On the other hand, there can be no question 

 as to the otolith found in this skull being specifically identical with 

 the one figured and described by Herr E. Koken (6) as Otolithus 

 {incertce sedis) crassus, and this specific name must therefore be 

 adopted for our specimen, which will henceforth be known as Arius 

 crassus. Should the spines called A. bartonensis prove eventually 

 to belong to the same species, the name of A. crassus having priority 

 will have to be retained, although it may be a less appropriate cog- 

 nomen. According to Herr Koken this form of otolith has been 

 found at Headon Hill, Isle of Wight, and also in Oligocene strata 

 at Lattorf, Cassel, Westeregeln, and Waldbockelheim, in Germany. 

 The specimen from the Miocene of Tortonese, referred to by Herr 

 Koken as possibly belonging to this species, which is figured by 

 Dr. Sismonda (Mem. Accad. Sci. Torino, 1849, ser. 2, vol. x. pi. 2. 

 fig. 71), does not seem to me to belong to the genus Arius. 



It is quite likely that one or other of the three forms of otoliths 

 from Barton may belong to A. egertoni or A. bartonensis ; but it 

 is likely to be lon» before the means of correlating them will be 

 found, and I have thought it best to distinguish them provisionally 

 as Arius (otolithus) sp. A (fig. 4), Arius {otolithus) sp. B (fig. 5), 

 and Arius {otolithus) sp. C (fig. 6). 



Should the otolith from Ankoala, Madagascar (fig. 7), prove to 

 belong to an undescribed species, I would suggest that it be named 

 after the gentleman who brought it to this country, Arius baroni. 



Works that mag be consulted on Fossil Siluroid Fishes. 



1. Cope, E. D. — The Vertebrataof the Tertiary Formations of the 

 West. Rep. U.S. Geol. Surv. vol. iii. Book i. p. 62 (1884). 



2. Dixon, F. — Geology and Fossils of Sussex. 1st edit. 1850, 



p. 204; 2nd edit. 1878, p. 244. 



3. Gtj^THER, A. — " Contributions to our Knowledge of the Fish 



