TRANSACrriOiSS OF SECTION c. 695 



5. The Great Devonian Placoderms of Ohio, xvith Specimens. 

 By Professor E. W. Claypole, D.Sc. {Land.) 



The Upper Devonian Shales of Ohio have recently afforded a remarkable series 

 of fossil fishes rivalling in size and interest those found many years ago in the Old 

 Red Sandstoui.'s of similar age in Scotland, and described by Agassiz and Hugh 

 Miller. The earliest of these, DinicMhys, was closely studied, and its structure 

 was well explained by the late Dr. Newberr3^ It was an immense armour-clad 

 fish whose head measured from 2 to 3 feet in length. Titnnichtln/x, the second of 

 the group, though less massive, was of yet larger size. Gorgonichth/s, the third, 

 was described by the present writer in 1893, and, so far as is yet known, was 

 the most formidable of all, possessing jaws of enormous size and thickness, above 

 24 inches long, ending in teeth or points from 6 to 9 inches in length. The last 

 of the four, Brontichtkys, of which a description was also published by the writer 

 in the ' American Geologist' for 1894, is equally heavy and of equal size, but 

 differs from all the rest in possessing verj' massive symphysial portions in the 

 mandibles with sockets apparently for the reception of teeth, as in Titanichthys. 



Of the two last-named genera only the jaws are yet known with exactness. 

 Other portions have been found of Gorgoniclithys, but are still embedded in the 

 matrix. So far as can at present be determined, all the four are closely allied to 

 Coccosteiis, and belong to the same family. 



The set of casts exhibited in illustration of the fossils have been prepared by 

 their discoverer. Dr. "William Clark, and faithfully represent the originals, of many 

 of which only single specimens are yet known. The labour of extricating them 

 from the pyritous shale has proved very heavy, and much yet remains to be done 

 in this direction. 



6. Notes on the PhyJogeny of the Graptolites. By Prof. H. A. Nicholson, 

 M.D., D.Sc, F.G.S., and J. E. Marr, M.A., F.R.S., Sec. G.S. 



The authors note that the number of stipes possessed by graptolites has been 

 looked upon as a character of prime importance, many genera being based on the 

 possession of a certain number. Again, the ' angle of divergence ' has been looked 

 upon as an important factor in the diagnosis of families. They are, however, led to 

 believe that a character of essential importance in dealing with the classification of 

 the graptolites, and one which, in all probability, indicates the true line of descent, 

 is found in the shape and structure of tlie hydrothecse, the point of next import- 

 ance as indicating genetic relationship being tlie ' angle of divergence.' 



These views are illustrated by reference to forms belonging to the ' genera ' 

 Bryoyraptus, Dichograptus, Tetragraptiis, and Didymograptus, which appear in 

 turn in this sequence. 



Out of nine Tetragrapti (and the authors know of no other forms referred to 

 this genus which are represented by well-preserved examples), eight are closely 

 represented by I'orms of Didymograptus, which are closely comparable with them 

 as regards characters of hydrothecre and amount of ' angle of divergence,' whilst 

 the ninth is ccmiparable ^v•ith a Didymograjjtus as regards ' angle of divergence ' 

 onl}'. Moreover, four of the Tetragrapti are comparable as regards the two 

 above-named important characters with firms of Dichograpfii.^ and Brj/nyroptus 

 with eight or more branches, and the authors conlidently predict the discovery of 

 forms belonging to these or closely allied many-branched ' genera,' agreeing with 

 the remaining Tetragrapti in what they regard as essential characters. 



They give details showing the points of agreement of each group of the various 

 series, including a two-branched, a four-branched, and a mariy-branched lorm, and 

 point out hoM- difficult it is to understand how the extraordinary resemblances 

 between the various species of Tetragrapius and Didymograptus (to take one 

 example) have arisen, if, as usually supposed, all the species of the genera have 

 descended from a common ancestral form for each genus, in the one case four- 

 branched, and in the other case two-branched. On the other hand, it is compara- 

 tively easy to explain the more or less simultaneous existence of forms possessing 



