236 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.52. 



During the first half of the year 1914, in pursuance of arrange- 

 nieuis made by Dr. Charles D. Walcott, the fossil fish remains in the 

 United States National Museum were systematically arranged and 

 put in order by the present writer, and a number of undeseribed or 

 scientifically interesting specimens were set aside for special study. 

 The greater part of these were afterv\^ards placed in the hands of Mr. 

 R. Weber for the purpose of preparing suitable illustrations to accom- 

 pany a report upon the collection, which is now printed m the follow- 

 ing pages. To Dr. G. P. Merrill, head curator of geology, and to Mr. 

 J.W. Gidley and Mr. C. W. Gilmore, of the Section of Vertebrate Pale- 

 ontology^, cordial thanks are returned for the enjoyment of many 

 privileges and courtesies extended while the work of studying the 

 collection was in progress. 



A. ORDOVICIAN SYSTEM. 



The earliest remains of vertebrate life anj^vhere found are those 

 occurring at several Ordovician localities in Colorado, the Bighorn 

 Mountains of Montana, and the Black Hills uplift of South Dakota. 

 The longest and best known of these localities is Harding Quarry, a 

 short distance west of Canon City, Colorado, where vast quantities 

 of detached scales and other fragmentary hard parts of primitive 

 fishes are contained in sandstone (now known as the Harding sand- 

 stone) usually correlated with the Lower Trenton of the eastern 

 United States and Lower Bala of Wales. R. S. Bassler in his biblio- 

 graphic index of American Ordovician and Silurian fossils places the 

 Harding sandstone in the Black River groups because it underlies the 

 Kinnswick limestone of uppermost Black River age. 



Three species, supposed to represent true fishes, were established 

 by Dr. C. D. Walcott,* after an examination of hundreds of fragments 

 collected by liimself and Dr. T. W. Stanton at Canon City in 1890 

 and 1891, and the ichthyic nature of the remains was confirmed by 

 Dr. Otto Jaekel's study of microscopic sections of dermal plates 

 belonging to two of the species. 



Cope, however, in a review of Walcott's paper, expressed the opin- 

 ion that it is " extremely unlikely that these forms are fishes, but they 

 are more likely Agnatha." Under this latter term the author just 

 quoted included the great extinct group of fishlike vertebrates to 

 which he applied the name of Octracodermi, and to which he assigned 

 a lower rank than that of Pisces proper. 



• Walcott, C. D. Notes on the discovery of a vertebrate fauna in Silurian (Ordovician) strata. Bull. 

 Geol. Soc. Amer., vol. 3, 1892, pp. 153-172. Doctor Jaekel's view.s as to the nature of the remains are appen- 

 ded to this article, and are again stated in a review published by him in Neues .Talirb. f. Mineral.. 189."i, p. 

 162. Cope's review is found in the American Naturalist for March 1S93, pp. 268-269. See also the same 

 journal for February. 1891, p. 137, for a notice of Doctor Walcott's fust communication. 



