Nipher — Gravitational Repulsion. 181 



When T was determined the temperature of the air in 

 the room was 15°.5 C = t a 



The temperature of the screen was t s = 15°.2 C. 

 When T x was determined, 



t a - 17°.4 C 

 t s = 17°.4C 



The temperature of the two large masses was 16°.4 

 and 16°.6. 



These values were determined by means of four ther- 

 mometers reading to tenths of a degree. The bulbs were 

 placed in contact with the masses, and covered with a 

 layer of cotton batting. 



When the readings for T 2 were finished 



t a = 18°.0 C 

 t,= 17°.9 C 



The temperature of the large masses was 17°.7 and 

 17°.8. 



On a former occasion, when the bifilar suspension was 

 different, it was found that placing metal vessels contain- 

 ing a litre of water having a temperature of 28° C. at the 

 beginning of the experiment, and 24° at the close, in the 

 positions occupied by the large masses, the temperature 

 of the air and shield being 13°.8, the value of T was in- 

 creased from 644.2 sec. to 652.5. The warm vessels were 

 in contact with the shield. 



When the metal vessels contained water and ice, the 

 temperature of the air and shield being 12°.0, the value 

 of T was decreased by 25 seconds. 



These results seem to indicate clearly that convection 

 of the air within the shield had no material effect upon 

 the values of the quantities used in computing the value 

 of n in Eq. 4. 



It apparently follows that the value of the gravitation 

 constant was made negative by the process described. 

 Attraction was converted into a repulsion. The influence 

 machine was unusually active by reason of fresh drying 



