1894.] 



on the Newtonian Constant of Gravitation. 



359 



couple increasing with the angle up to a certain point (65° in tho 

 particular case), after which the couple falls off again and becomes 

 zero when it has turned 180°. 



Since the effect is a maximum at 65°, very great accuracy in the 

 measurement of this angle is of little consequence. By means of a 

 small telescope at a distant table, and the divided edge and vernier, I 



Fig. 1. 



By permission of the Engineer. 



Zr^y 



can tell the angle with certainty to 1/20 degree ; an uncertainty of 

 one-quarter of a degree would be of but little consequence. Again, 

 if the pair of gold balls twist about an axis which is not exactly 

 that round which the lead balls are carried, if there is any small 

 eccentricity of the gold or lead balls, then eccentricity in the common 

 plane removes the gold balls from a position of minimum effect, 

 eccentricity across the plane removes them from a position of. 



