Hd^diSdcietf/, 



151 



to justify the belief that it might not only indicate the direction, but 

 measure the angular value of the deviation of the pendulum plane 

 from the meridian. Computing this value for an hour to be 12°+, 

 we watched the result and found the arc passed over to be 13°. When 

 this observation was recorded, the pendulum was supposed to have 

 commenced its vibrations on a north and south line ; but that was 

 an error ; it was really swung from east to west. 



In repeating this experiment, I have been more anxious to vary 

 the conditions, in a few arranged observations, than to accumulate 

 many simile results. We have observed in four azimuthal planes ; 

 one of our balls weighed eight pounds, the other twenty pounds : 

 one was an oblate, the other a prolate spheroid ; suspension was 

 effected at first by thoroughly softened catgut, afterwards by un- 

 twisted silk : we have compared small and large arcs, counted the 

 periodical times of vibration in three planes, noted the direction of 

 motion in the elliptic path of the pendulum, and estimated the 

 length of its minor axis. We have recorded results when no ellip- 

 ticity was remarked, and others in which its injurious effect was 

 manifest. 



The pendulum performs one complete vibration in 8": from 

 which its length is deduced =52'+ feet. The chord of the arc of 

 vibration was usually taken at 14« feet, but was on some trials re- 

 duced to 7 feet. The graduated circle was 12 feet in diameter. 

 Great care was used in starting the ball, which did not rotate, but 

 presented the same face to the same quarter of the room, in what- 

 ever direction it was swung. The apartment was subject to air 

 currents; the floor from which the suspension was effected though 

 strong was large ; and there was no method of securing exact verti- 

 cality in the iron tube which carried the flexible catgut or silk. 



From one or all of these causes it happened that ellipticity in the 

 path was noticed in almost every experiment, and it might exist in 

 all, and be unobserved if the minor axis did not exceed one-fourth 

 of an inch. After abandoning several trials in which the minor axis 

 was observed to increase rapidly, it was thought desirable to deter- 

 mine by experiment the effect of this elliptical swing on the angular 

 movement of the pendulum plane (see exp. 5). 



J^irst Set. Deviation observed after given intervals of Time. Sail 

 a prolate spheroid, weighing eight pounds. Suspension softened 

 catgut. 



Supposing no ellipticity to have existed in these experiments, the 

 results are — 



Deviation in one hour from E. and W. line to right 13°*0 



N. andS, „ 15°'3 



M2 



