170 Trans. Acad. Sci. of St. Louis. 



placed a sheet of glass and three layers of asbestos 

 paper. 



The shield was mounted upon a heavy piece of timber 

 and rested directly upon a sheet of glass. It was also 

 wrapped with asbestos paper. The long suspension tube 

 was held in stable position by three heavy silk cords at- 

 tached to surrounding cases, and loaded with a series of 

 distributed masses, about 25 in number, and having a 

 mass varying from 50 to 100 grams each. The metal 

 shield was loaded with two ten-pound masses of iron 

 and its sides were clamped with wooden clamps in order 

 to quiet any vibrations in the shield. The large iron 

 masses and the metal shield were connected with each 

 other by large copper wires, leading to a spark-knob 

 mounted upon the table. Another spark-knob, mounted 

 on an independent support, was provided with a dis- 

 charge terminal, consisting of 150 pin-points, the heads 

 of which were soldered to a copper disc. The entire ap- 

 paratus was mounted upon a massive table. The floor 

 of the room was of reinforced concrete. The discharge 

 terminal was connected with the influence machine in an 

 adjoining room, by means of brass rods hung upon silk 

 cords. The machine was operated by a synchronous elec- 

 tric motor. The discharge knobs at the machine were 

 separated so that no disruptive discharges were possible. 

 One terminal of the machine was grounded. There were 

 no condensers on the machine terminals. 



The time interval of a to and fro vibration of the sus- 

 pended masses was 485 seconds. When at rest there 

 appeared to be no disturbance of any kind. The reading 

 would somtimes remain constant for hours. At other 

 times the reading slowly changed. It often varied 

 throughout a day through 20 scale divisions. It was 

 found that the large masses appeared to attract the sus- 

 pended masses with a varying force, even when the 

 former were hung from the ceiling above. Sometimes they 

 appeared to repel the suspended masses. These vary- 

 ing effects were finally traced to very moderate changes 



