NO. 2177. FOSSIL FISHES IN NATIONAL MUSEUM— EASTMAN. 239 



fore us, we should prefer to compare it with the large dorsomedian 

 shield of Psammosteus and Drepanaspis, more particularly with such 

 forms as Psammosteus taylori Traquair, from the Scottish Old Red 

 Sandstone, and the Russian P. paradoxus. For an account of the 

 structm-e of the shield in the two last-named species we may refer to 

 recent papei's by Traquair, Woodward, and Preobrajensky in which 

 the organization of Psammosteus is described.^ 



The large compound plate of Astraspis differs from the dorsome- 

 dian and ventromedian shields of known Psammosteidae not only in 

 outline, so far as may be judged from the portion preserved, but also 

 in the presence of a median and two lateral ridges, which extend over 

 the anterior half of the plate in a longitudinal direction. The outer- 

 most of these folds is apparently very close to and parallel with the 

 external margin of the plate. The ridges in question recall somewhat 

 the rows of enlarged tubercles seen in the head-shield of Thyestes and 

 Tremataspis. Owing to their presence, and to the pecuhar form of 

 the sliield, which resembles somewhat a dorsomedian plate of Coccos- 

 teus, it seems necessary to place the genus Astraspis in an indepen- 

 dent family, closely alHed to the Psammosteidae. Some further 

 remarks on the genus Psammosteus will be found under the caption 

 of Silurian fishes. 



Formation and locality. — Hastings sandstone, Black River group; 

 Canon City. 



B. SILURIAN SYSTEM. 



Family PTERASPIDAE Smith Woodward. 



Genus PALAEASPIS Claypole. 



PALAEASPIS AMERICANA Claypole. 



Palaeaspis americana Claypole, Amer. Naturalist, vol. 18, 1S84, p. 1224. 

 Palaeaspis bitruncata t'LAYPOLE, Amer. Naturalist, vol. 18, 1884, p. 1224. 

 Palaeaspis elliptica Clatpole, Proc. Amer. Assoc. Adv. Sci., 1885, p. 426. 



This species is represented in the collection by a number of frag- 

 mentary remains, none of which adds to our knowledge of the struc- 

 tural organization. It deserves to be recalled, however, that the 

 reported presence of paired ap|>endages in this species has been dis- 

 proved by Dr. BashJord Dean, his investigation of the type material 

 having shown that the structures regarded as such by the original 

 describer are in reality Elasmobranch spines resembling OncTius, 

 which have become accidently associated with the dermal shields of 

 Palaeaspis. The dorsal shield consists apparently of a single plate, 



> Traquair, R. H. Additional notes on the fossil fishes of the Upper Old Red Sandstone of the Moray 

 Firth Area. Proc. Roy. Phys. Soc. Edinb., vol. 13, 1897, p. 379. Woodward, A. S. On the Upper Devon- 

 ian Ostracoderm, Psammosteus taylori. Ann. Mag. Nat. Hist., ser. s, vol. 8, 19n, pp. 648-652. Preobra- 

 jensky, J. A. Ueber einige Vertreter der Familie Psammosteidae Ag. Sitzber. Naturforsch. Gesellsch- 

 Univ. Dorpat, vol. 19, 1910, pp. 21-36. (Text in Russian.) 



