734 report — 1884. 



Pennsylvania. This shale in New York immediately overlies the Niagara lime- 

 stone, which is correlated on satisfactory evidence with the Wenlock limestone of 

 England. Ten or twelve species are common to the two heds. 



It seems, therefore, tbat the great mass of coloured shale, near the top of which 

 these fossils were found, and which is a continuation of the Onondaga group of New 

 York, has no representative in the British series, but corresponds to an interval 

 between the Upper Wenlock and the Lower Ludlow. (For details regarding the 

 correlation of these heds in Pennsylvania with those in New York, see a paper by 

 the author in ' Proc. Amer. Phil. Soc.' for 1884.) 



It is consequently a necessary inference that the beds yielding PaUea&pis and 

 Onclms in Pennsylvania are somewhat older than those containing Scaphaspis and 

 Onclius in England. 



Microscopic examination of the specimens, and a comparison of their structure 

 with that of Scaphaspis and Cephalaspis, are in progress, and the details will be 

 given below. Other fossils in the author's possession indicate the possible exist- 

 ence of tish at a still earlier date, but the material is not yet worked out. 



The microscopic structure of Palaaspis corresponds in all important points with 

 that of Scaphaspis as figured by Huxley (' Q. J. G. S.' vol. xiv., pi. xv., fig. 1). The 

 basal cellular and vascular layers are present in both, but the latter presents a 

 regularity of structure to which Scaphaspis affords no parallel. A section parallel 

 with the striation presents a beautiful forest-like appearance produced by the rise 

 and divergence of the tubules from the larger vessels. The most remarkable 

 anatomical peculiarity of these fossils is also one in which they coincide with the 

 Pteraspids of England. All the species belonging to this group are distinguished 

 by the total absence of the bone cells (lacuna) and the connecting tubules (cana- 

 liculi) characteristic of all other bone-structure in all classes of the animal king- 

 dom. This absence was the greatest objection to the admission of the fossil to 

 this class of fishes. Though now overruled, its importance is in no wise diminished. 

 To receive these forms the group of Heterostracaus was established by Professor 

 Lankester in 1SG9 as a sub-divisinn of Huxley's family of Cephalaspids. So pro- 

 found a difference, however, forming a distinct line of demarcation between the 

 fossils of the family seems to demand bolder recognition, and it was proposed to 

 raise Lankester's sub-family of Ileterostracans into a distinct family under the 

 name of Pteraspids (Pteraspidai) and the name of Cephalaspids (Cephafaspidce) be 

 retained for those forms exhibiting true bony tissue, and composing the sub-family 

 of the Osteostracans of Lankester. 



3. On American Jurassic Mammals. By Professor 0. C. Marsh. 



The first Jurassic Mammals discovered in this country were found in 1878, in 

 the Atlantosaurus beds of Wyoming Territory, and described by the author. Other 

 discoveries in the same region soon followed, and a systematic exploration of this 

 geological horizon has been continued by the author up to the present time. 



This horizon in the upper Jurassic, with its characteristic fossils, has been 

 traced by the author along the flanks of the Rocky Mountains for a distance of 

 more than three hundred miles. Its position is shown in the geological section on 

 page 735. The deposits are lacustrine, and the accompanying fossils are dinosaurs, 

 pterodactyles, crocodiles, lizards, and fishes. 



Remains of Mammals have been found by the author at several points along 

 the line of this horizon for two hundred miles or more, but the most productive 

 locality is in the region where the first discovery was made. At one point, from 

 a space scarcely larger than the room in which the author was speaking, remains 

 of more than three hundred individuals of Jurassic Mammals have been obtained, 

 a good indication of the wonderful richness of the fauna here entombed. 



Other localities have yielded many important specimens, so that up to the pre- 

 sent time, the author has secured the remains of nearly or quite four hundred 

 individuals of Jurassic Mammals. So far as known, no other specimens of this class 

 have been secured from this formation in America. 



