• A. S. Woodward— Eocene Siluroid Fishes. 303 



certain centres. The second, third, or fourth series of spikelets attach- 

 ing themselves to one of the rays of a skeleton-crystal of magnetite 

 may be considered as independent of the similar additions, whether 

 simultaneous or not, made during consolidation to the other primary 

 arras. One arm, moreover, in such examples frequently attains to 

 a much greater development than the rest, and corresponding pheno- 

 mena may be expected among skeleton spherulites. 



I would submit these few observations as further evidence in 

 favour of regarding the pyromerides as the altered representatives 

 of the rhyolites of to-day, of which they reproduce, not only the 

 familiar, but also some of the rarer types. It is interesting to 

 observe that, equally with those of Jersey, the Wuenheim lavas are 

 held to be of Permian age. 



In the microscopic work relating to this subject, I have been 

 much indebted to preparations made in the Geological Laboratory 

 of the Normal School of Science and Eoyal School of Mines. 



III. — On Some Ebmains of Siluroid Fishes from British Eocene 



Formations. 



By A. Smith Woodwaed, F.G.S., F.Z.S., 



of the Britisli Museum (Natural History), 



AMONG the early Tertiary Fishes in the British Museum, there 

 are a number of detached spines and cranial fragments from the 

 Middle and Upper Eocene beds of Bracklesham and Barton, which 

 are undoubtedly referable to extinct members of the family of 

 Siluridae. With the exception, however, of three specimens figured 

 and briefly noticed by Dixon in his work on the fossils of Sussex,^ 

 all have remained hitherto undescribed ; nor is it an easy task to 

 base any satisfactory identifications upon such fragmentary materials. 

 But as it has sometimes been asserted that no traces of this family 

 have yet been discovered in European formations, — and, as moreover, 

 at least one important generalization has been based upon the 

 supposed fact,^ — it will perhaps be of interest to offer a few notes 

 upon these fossils, to show that their non-recognition is due rather 

 to the imperfection of the geological record, than to their actual 

 absence in the rocks. And some slight notice is all the more 

 desirable, since Dixon's reference of the Bracklesham specimens to 

 the temperate genus Silurus is obviously erroneous, and the mistake 

 has escaped correction in the second and revised edition of his great 

 work. 



It is impossible, of course, to attempt any precise specific diagnoses 

 equivalent to those recognized by the zoologist among living forms ; 

 for no evidence is yet forthcoming as to the arrangement and 

 proportions of the fins, which constitute so prominent a character 

 of note in systematic works. But there is very good reason to believe 



1 F. Dixon, " Geology and Fossils of Sussex," 1st edit. 1850, p. 204, pi. xi. 

 figs. 11-12 [2nd ed. p. 24i, pi. 11, figs. 11-13. J 



^ "The Siluroids .... came into existence after the Cj'prinoids, fossil remains 

 being known only from Tertiary deposits in India, none from Europe." (A. Giinther, 

 " Study of Fishes," 1880, p. 216.) 



