CHAPTER XII. 



GRAVITATIONAL EXPERIMENTS. 



87. Introductory. In my last report * I inferred that it might be worth 

 while to study the motion of the usual gravitational needle in vacuo during 

 a brief interval after the attracting mass has been put in place. If this can 

 be done before the torque of the quartz fiber has become appreciable, the 

 weights at the end of the needle may as a first approximation be supposed to 

 be in uniformly varied motion. It would then be possible to deduce the 

 Newtonian constant in terms of the initial acceleration of the needle, or obser- 

 vationally in terms of its displacement during successive small times. It is 

 necessary, therefore, that the smallest possible displacements of the needle 

 should be measured, and hence the quadratic interferometer will be advan- 

 tageously used for this purpose. It is my purpose in the present paper to 

 pursue this investigation further. 



88. Apparatus. The apparatus used is essentially the same as that already 

 described and consists of a very flat box with large plate-glass sides parallel 

 to the needle inclosed and normal to the rays from the interferometer. In the 

 present experiments this box is made air-tight with cement and is capable of 

 being exhausted. 



The needle consisted originally of a shaft of straw 25.6 cm. long, 

 carrying a small shot of 0.61 gram at either end and suspended from a 

 delicate quartz fiber 17 cm. long. Its air-damped period was about 18 or 20 

 minutes. Symmetrically to the middle of the needle two parallel small light 

 mirrors were adjustably attached to receive the beams of the interferometer. 

 The needle weighed 1.49 grams. 



The attracting weight was a ball of lead 949 grams mass and could be 

 moved expeditiously on a circular track to 4.2 cm. on either side of the shots. 

 It was sufficient to act on one end of the needle only. This disposition was 

 maintained in the earlier experiments, in which no exhaustion was attempted. 

 The new modifications of the apparatus will be described in connection with 

 the observations. 



The arrangement of the gravitational needle relative to the interferometer 

 is adequately shown in figure 117. White light L from a collimator is guided 

 by the mirrors N, N', N", N'" and tn, m r , on the needle, into the telescope 

 at T. The mirrors N,N", N'", are half-silvered. The mirror N f is on a mi- 

 crometer with its screw in the normal direction n. The attracting mass M 

 is moved alternately from M to M' on the track t, adequately supplied by 

 a stout crank provided with adjustable stops. The breadth b of the ray 

 parallelogram was about 10 cm. and is the same as the mean distance from 

 m to m', the quartz fiber being between. In place of n a double offset 



* Carnegie Inst. Wash. Pub. No. 249, part III, Chapter IV. 

 116 



