280 Annals of the Carnegie Museum. 



stone as a well determined unit of the Paleozoic section of Colorado 

 must be considered as well established." It is adjudged by the 

 authority just quoted that the formation in question is of uppermost 

 Devonian age, but the fauna which it contains is but distantly related 

 to those of the New York area, or even to the more western Devonian 

 faunas of this country. It is, on the other hand, "somewhat strikingly 

 similar to the Devonian of Russia." 



The Ouray limestone, with a thickness ranging from lOO to 250 

 feet in the San Juan country, rests conformably upon strata a hundred 

 feet or so in thickness, which Dr. Cross has named the Elbert formation, 

 and in the opinion of this geologist the strata so designated "seem 

 unquestionably to form a lithologic, stratigraphic and faunal unit." 

 Intervening between the Elbert formation and the basal granite of 

 the region are beds of quartzite, supposed to be of Upper Cambrian 

 age. 



So much for the general Palaeozoic section of the San Juan region. 

 The stratigraphic equivalence of the Elbert formation is shown by 

 the evidence of its fish-remains, the only fossils yet obtained from it, 

 to be with the so-called "Parting Quartzite" of Leadville, in central 

 Colorado, and of Aspen, on the northeastern flank of the Elk moun- 

 tains; and the general aspect of these fish-remains has been pointed 

 out by the present writer to be indicative of Upper Devonian age.^ 

 Nevertheless, the ichthyic fauna of the Elbert was recognized as not 

 being closely similar to the faunas of the eastern and central United 

 States, in this respect agreeing with the Ouray invertebrate fossils, 

 which Dr. Girty has shown to "exhibit a closer parallel with the 

 Devonian of the Ural Mountains." 



The fish-remains thus far brought to light from the Elbert forma- 

 tion in Colorado, although numerically abundant, present a singular 

 lack of systematic diversity. Arthrodires are represented by dissoci- 

 ated tuberculated plates belonging to a fish about twice the size of 

 the type species of Coccosteus, but whose precise relations are not 

 determinable. Besides these fragments, only four recognizable 

 species have been thus far identified, as follows: Bothriolepis colo- 

 radensis, B. nitida,^ Iloloptychiiis gigQiiteus, and H. tuhercidatiis. 

 The second, third and fourth of the species just named occur typically 



' "On Upper Devonian Fish Remains from Colorado," Amer. Jour. ScL, Vol. 

 XVIII, 1904, p. 260. 



* This name antedates that of B. leidyi, which is synonymous with it. 



