336 



Messrs. Guthrie and Bovs on 





Speed. Torsion. | Zero. 



Speed 



Torsion 



a and b 



17-79 



206 



128-2 

 128-4 

 1 28-9 



•804^ 



•857 )■ -860 

 •867; 



b and c 



66-19 



1G-91 



77-2 

 10-5 









e 



14-04 



45-6 

 16-3 



128-7 



•4795 



293 



-——r= 1-794, the corrected resistance of the liquid ; 

 55-80, the corrected conductivity. 



The zero in the next column is obtained from the two ob- 

 servations by a simple proportion sum ; and the numbers in the 

 last column are obtained as shown, and are a measure of the 

 resistance of the liquid, but uncorrected by the comparison 

 plate. The three figures obtained differ ; but the greatest 

 weight is given to the observation at the highest speed, and 

 •860 is taken as a fair mean. To make use of the observa- 

 tion e, the probable zero must be found by examining the 

 numbers already obtained. As they indicate a gradual rise 

 in its position, a number higher .than the mean is taken. 

 This shifting in the zero caused more uncertainty than any 

 thing else ; and it is this that limits the accuracy of the in- 

 vestigation. The probable zero subtracted from the reading 

 174-3 gives the torsion 45*6, due to the liquid and to the plate 

 together ; but the torsion due to the liquid alone, of the appa- 

 rent resistance *860 already found, and with the speed 14*04, 

 is 16*3 ; and this, subtracted from 45*6, leaves 29*3, the tor- 

 sion due to the plate alone ; and this divided into the speed 

 14*04 gives the quotient -4795, an arbitrary number, the 

 changes in which from time to time represent changes in the 

 magnets or wire. From this the corrected resistance 1*794 is 

 found as shown ; and its reciprocal is the conductivity. All 

 these numbers are in themselves entirely arbitrary ; but they 

 are all comparable with one another ; so that any number of 

 liquids may have their conductivities compared with one ano- 

 ther, and eventually with that of some standard metal. 



The conductivities of sulphuric acid and of sulphate of 

 copper, the only liquids examined at present, are given in 

 the following Table. The conductivity-curve of sulphuric 

 acid agrees in every important particular with Kohlrausch's, 

 which was obtained with alternating currents, the chief dif- 



