On tJw Calibration of Capaciti/'ho,i'es. 101 



was made tioht by a mercury seal. The tube, with the 

 radium beside it, was placed in a large box lined with lead. 



In making the experiments the tube was sealed to a three- 

 fall Sprengel pump and exhausted, being gently warmed at 

 intervals during the })rocess. When the McLeod gauge 

 showed a high vacuum had been obtained dry air was slowly 

 admitted through a series of drying-tubes until atmospheric 

 pressure was reached. The glass tube giving connexion 

 with the external air was then sealed off and observations of 

 the potential-difference were commenced. The pump was 

 started and kept working for about six hours, readings of the 

 electrometer being taken from time to time. When the 

 highest vacuuui attainable had been reached dry air was 

 once more admitted to the testing-vessel, but no alteration in 

 the deflexion could be perceived. The series of operations 

 was repeated with the same result. It was estimated that an 

 alteration of 5 per cent, could be detected. 



In conclusion we must express our great indebtedness to 

 Lord Kelvin and Dr. J. T. Bottomley for their interest and 

 advice in the course of these experiments. 



Blvthswood Laboratory, 

 Renfrew, N.B. 



LXXXIX. On the Application of Alter natiJig Currents to the 

 Calibration of Capacity-hoxes^ and to the Comparison of 

 Capacities and Inductances. By W. Stkoud^ M.A., D.Sc, 

 Cavendish Professor of Physics, and J. H. Oates, JB.Sc, 

 Assistant -Demonstrator in the Physical Laboratory at the 

 Yorkshire College, Leeds* . 



TN the May number of the Philosophical Magazine for the 

 X current year there appeared a paper by Prof. Fleming 

 and Mr. Clinton upon the measurement of small capacities 

 and inductances. By means of a substantially built and 

 rapidly revolving commutator and a special form of movable 

 double-coil galvanometer, capacities amounting to only a few 

 micro-microfarads and inductances amounting to only a few 

 millihenries are measured. For some time prior to the pub- 

 lication of this paper we had been engaged upon the solution 

 of a similar problem, viz. the accurate comparison of a 

 capacity with either an inductance or a second capacity ; we 

 had not, however, any idea of measuring such very small 

 capacities as those dealt with by Prof. Fleming^s method, 

 neither had we any idea of the practical importance of such 



* Communicated by the Authors. 



