PHYSICS: L. B. LOEB 
347 
with phosphorous pentoxide in the later measurements. The variations 
of the values of the mobilities as shown in the table were in a large 
measure attributable to insufficient drying in the earlier determina- 
tions. Fluctuations in the operation of the arc, together with sHght 
uncertainties as to the value of the constants of the alternating po- 
tential difference, might possibly have caused an error of ten per 
cent in the determination of the absolute value of the mobiHties, The 
actual error, however, is probably much less than this. By means of 
check determinations, using the 120 volt 60 cycle city alternating po- 
tential, which gave strictly normal values of the mobihties, the possi- 
bility of any instrumental errors was precluded. 
The results obtained are shown in Table 1. In the first column are 
the frequencies of alternation in cycles per second. In the next 
