144 T. C. Mendenhall — Use of Planes and Knife-edges 



in doubt, containing like or similar faunas. Notably among 

 the latter are the Denver and Arapahoe beds in the vicinity 

 of Denver, Colorado, and their probable equivalents in other 

 portions of the same state, recently described by Mr. "Whitman 

 Cross, as " The Post Laramie Deposits of Colorado."* Mr. 

 Cross refers these beds to a period later than the Laramie on 

 account of unconformities existing, in some places, at least, 

 between them and the underlying recognized Laramie and 

 older Cretaceous formations, and in opposition to the evidence 

 afforded by -the vertebrate fossils thus far obtained in them. 

 Whether it is better to ignore the evidence afforded by the 

 vertebrate fossils, or to treat these unconformities as local, 

 remains to be decided by future investigations. It is quite 

 possible that Dinosaurs continued into the early Eocene, but 

 they were represented there, if at all, only by degenerate 

 types, and less- specialized forms. It would certainly be 

 remarkable, not to say impossible,- that a group of Dinosaurs 

 showing so great a development and marked specialization as 

 are to be seen in Triceratops and Claosaurus should continue 

 uninterruptedly from near the base of the Laramie up into 

 the Tertiary. Nevertheless at least two species of Triceratops 

 have been described from the Denver beds referred by Mr. 

 Cross to the Post-Laramie. Regions affording such contra- 

 dictory evidences should be thoroughly examined, and, where 

 possible, their actual stratigraphical relations should be deter- 

 mined. Results thus attained might be sufficient to harmonize 

 observations now in apparent opposition. 



Tale Museum, New Haven, Conn., December 5, 1892. 



Art. XVIII. — On the Use of Planes and Knife-edges in 



Pendulums for Gravity Measurements ;f by 



'T. C. Mendenhall. 



In the theoretical discussion of the pendulum it is usual 

 to assume that it vibrates about an axis which is a straight line. 

 In practice it is generally attempted to realize this condition 

 as nearly as possible and the method almost if not quite uni- 

 versally adopted has been to attach what is technically known 

 as a knife-edge to the pendulum and allow it to rest upon a 

 suitably supported plane horizontal surface. The axis about 

 which rotation takes place is at any moment determined by 

 the contact of the edge with the plane surface. To satisfy 

 theoretical conditions it is evident that this edge must be 



*This Journal, vol. xliv, p. 19. July. 1892. 



f Read at the meeting of the National Academy of Sciences. Nov. 3, 18S2. 



