J)>\ 11. II. Traquair — On Brepanaspis. 153 



II. —Notes on Dkepaxaspis GMUNDEifENSis, Schluter. 

 By Dr. R. H. Traquaik, F.R.S. 



'■PHIS most remarkable fisb, from the Lower Devonian Eoofing 

 X Slate (Hunsriickschiefer) of Gmiinden.. in Western Germany, 

 was named, but very imperfectly described, by Schliiter in 1887, as 

 liis material was at that time of a very fragmentary nature. He 

 apparently considered the creature to be allied to Cephalai^pis. In 

 Dr. Smith Woodward's "Catalogue " (pt. ii, 1891, p. 311) it is only 

 mentioned by name along with a number of other imperfectly known 

 forms (AspidicJithys, Anomalichthijs, etc.) which he considered as 

 "perhaps for the most part" referable to the Coccosteidse. 



In 1890 I began to obtain specimens, at first very fragmentarj', 

 through the agency of Mr. Stiirtz in Bonn, and in exhibiting these 

 before the Koyal Society of Edinburgli, I expressed the opinion that 

 the affinities of Drepanasiris lay rather with the Pteraspidas {Nature, 

 vol. liv, p. 263). Finally, having got together a considerable 

 nraount of more perfect material, I incidentally described the fish in 

 my recently published memoir on the Silurian I'ishes of Scotland, 

 giving a restored figure of its dorsal aspect, which Dr. Smith 

 Woodward has reproduced in his critical notice of that memoir 

 which appeared in the Geological Magazine for February last. 

 The conclusion I came to was that Drepanaspis formed the type of 

 a distinct family, related not only to the Pteraspidas but also to the 

 Psammosteidse and Coelolepida?, and which must consequently along 

 with these three families be included in the Heterostraci, of which 

 the Coelolepidaj appear to form the starting-point. 



But beyond the fact that Drepanaspis possessed a large median 

 ventral plate as well as a dorsal one, and that the two could readily 

 be distinguished by their shape, ray knowledge of the topography of 

 the inferior surface of the carapace was still imperfect. Nor had 

 I ascertained the position of the mouth, though 1 said that " as it is 

 impossible to conceive of the absence of a mouth, we must conclude 

 that it was placed exactly at the anterior margin." 



However, since that memoir was published, I have received from 

 Dr. Krantz, of Bonn, a specimen of the carapace of Drepanaspis 

 (Fig. 2), which not only clears up previously unsettled questions 

 regarding the arrangement of the plates on the ventral surface, but 

 also shows the mouth itself in situ. 



But before describing the ventral surface, I may as well recapitulate 

 the facts already known regarding the configuration of the fish. 



The carapace (Fig. 1) is broad, depressed, very obtusely rounded 

 in front, and terminating behind in a prominent though rounded 

 angle on each side. There is on the dorsal surface a large median 

 plate (m.d.) of an ovate-hexagonal shape, the anterior margin being 

 short and nearly straight, while the still shorter posterior one is 

 acutely notched. Each postero-lateral angle of the carapace is 

 formed by a narrow, elongated, somewhat falciform plate (p-l-), the 

 postero-lateral or cornual, Avhich tapers to a sharp point in front and 

 forms most of the external marg-in of the shield. The rest of the 



