Velocity of the Electric Current. 107 



In the case in question, a current-velocity of 554,680,000 cm 

 per second would nave been indicated by l mm change of deflec- 

 tion when the coil reached 380 revolutions per second. In 

 these preliminary measurements, the highest degree of sensi- 

 bility attainable by the method in question had been approached 

 only in so far as the velocity of the revolving coil was con- 

 cerned. It was evident that the number of ampere-turns in 

 the revolving coil could be greatly increased without corres- 

 ponding loss of speed, and that the question of further increas- 

 ing the sensitiveness of the astatic pair, depended only upon 

 our power to eliminate the disturbance due to -the lack of com- 

 plete balance in the differential windings of the coil. 



To meet these ends a new coil was made for us by Mr. F. C. 

 Fowler, the mechanician of the department of Physics, to 

 whose skillful workmanship the excellent performance of the 

 coils already described was due. This new coil was of the 

 same diameter as the original one, and it was constructed in 

 the same general manner. It was wound upon a box-wood 

 spool, however, which was thick enough, axially, to admit of 

 390 turns of wire without changing the mean area of the wind- 

 ings. To avoid the Yerj great difficulty of constructing a dif- 

 ferentially wound coil so perfect that when carrying large cur- 

 rents, its effect upon the delicate astatic needle should be neg- 

 ligible, we resolved upon a modification in our method which 

 should make the complete electrical balancing of the coil un- 

 necessary. For the current from the storage battery used in 

 ou.r preliminary observations we substituted that generated 

 by a small alternating-current dynamo giving 40,000 reversals 

 a minute. The advantages of this change were very great, for 

 there was no appreciable effect upon the needle, even when a 

 much heavier current than we had attempted to use in our first 

 experiment was traversing the coil. Under these conditions 

 the sensitiveness of the astatic pair could be increased to the 

 highest degree compatible with the maintenance of a perma- 

 nent zero point, and the current traversing the coil was limited 

 only by the heating of the wires. 



The current from the alternating-current machine was meas- 

 ured by its heating effect upon a phosphor-bronze "wire about 

 fifty centimeters long, stretched vertically within a cylinder 

 surrounded by a water-jacket. The elongation of the wire was 

 made to move a small mirror by means of a simple device 

 which need not be described here, and the angular movement 

 of the mirror was read with a telescope and scale. This wire 

 had been previously subjected to extended investigation and its 

 reliability as an indicator of current was well established.* It 



*See the Thesis of P. P. Barton and P. R. Jones: "The Measurement of Al- 

 ternating Currents." MS. in the Library of Cornell University. 



