254 Rev, C. S. Lymaii on the Pendulum Experijnent. 



it 36 inches. Owing to the closeness of the entire apartment, 

 little or no trouble was experienced or apprehended from aerial 



currents. . 



Without giving the experiments in detail, we will only men- 

 tion some of the results. 



The pendulum was usually set vibrating with an arc of about 

 four feet, which in half an hour became reduced to 28 inches, in 

 an hour to 19 inches, in an hour and a half to 14 inches, in two 

 hours to lOJ inches, in three hours to 6^ inches, and in four hours 

 to 4J inches. In general the duration of the experiments was 

 not much over an hour, often much less, occasionally three or 

 four hours. 



As to the amount of ellipticity, there were but two instances 

 out of over thirty in which none could be perceiv^ed. In several 

 other cases it was barely perceptible after the lapse of two or 

 three hours, and in but one instance exceeded two-tenths of an 

 inch for the minor axis. In general it was less than one-tenth of 

 an inch at the close of the experiment. 



As to the motion in the ellipse, it was by no means always in 

 the same direction, and hence could not be due exclusively, at 

 least, to the influence of the earth's rotation. It seemed rather 

 to depend on some imperfection of the wire at the point of sus- 

 pension, or some other defect of the apparatus, inasmuch as it 

 appeared to have a pretty uniform connection with the direction 

 across the circle in which the pendulum vibrated. For example, 

 six times when that direction coincided with the meridian the 

 motion in the ellipse was with the hands of a watch. The same 

 number of times when put in vibration at right angles to the 

 meridian, the motion was in the contrary direction. When vibra- 

 ting from N.E. to S.W, the motion was with the hands; when 

 from N.W. to S.E. the reverse. And these results were nearly 

 uniform while the point of suspension retained a given position 

 in reference to the points of the compass, but when that position 

 was altered, the points at which a given direction of elliptical 

 motion was observed were changed also; showing, as before re- 

 marked, that the direction of elliptic motion was connected not 

 with the rotation of the earth, nor with the points of the com- 

 pass, but with something in the apparatus itself; and it probably 

 depended not on any imperfection in the ball of the pendulum, 

 but in the wire above at the point of suspension. The necessary 

 ellipticity of rotation was not detected in these experiments, being 

 doubtless over-balanced and lost amid the more efficient causes of 

 elHpticity arising from accidental sources. 



As to the rate of angular deviation of the plane of vibration 

 on the graduated circle, this was observed to be obviously affected 

 by the direction in which the pendulum moved in its elliptic 

 orbit — being accelerated when that motion coincided with the 



