152 HENEY A. EOWLAND 



difference of positive and negative rotations is over seven per cent, 

 we find 



sec. 



The rejection of all the higher powers of w renders the correction 

 uncertain, but it at least shows that the Ohm is somewhat smaller 

 than it was meant to be, which agrees with my experiments. 



It is to be regretted that the details of these experiments have 

 never been published, and so an exact estimate of their value can 

 never be made. Indeed we have no data for determining the value 

 of the Ohm from the experiments of 1863. All we know is that, in 

 the final result, the 1864 experiments had five times the weight of 

 those of 1863, and that the two results differed -16 per cent, but 

 which was the larger is not stated. Now the table of results pub- 

 lished in the report of the 1864 experiments contains many errors, 

 some of which we can find out by comparison of the columns. The 

 following corrections seem probable in the eleven experiments : No. 4, 

 second column, read 4-6375 for 4-6275. No. 10, fourth and fifth 

 columns, read 1-0032 and + 0-32 in place of 1-0040 and +0-40. No. 

 11, fourth and fifth columns, read 1-0065 and + 0-65 in place of 0-9981 

 and 0-19. Whether we make these corrections or not the mean 

 value is entirely incompatible with the statement with respect to the 

 1863 experiments. With the corrections the mean value of the 1864 



experiments is 1 Ohm = 1-00071 earth ^ uad \ and without them, using 



sec. 



the fourth column, it is 1-00014. With the corrections the difference 

 between fast and slow rotation is 6 per cent. 



In the year 1870 Professor F. Kohlrausch made a new determination 

 of Siemen's unit in absolute measure, the method being one formed 

 out of a combination of Weber's two methods of the earth inductor and 

 of damping, by which the constant of the galvanometer was eliminated, 

 and is the same as Weber used in his experiments of 1862. His formula 

 for the resistance of the circuit, omitting small corrections, is 



approximately, 



where 8 is the surface of the earth inductor, T is the horizontal inten- 

 sity of the earth's magnetism, K the moment of inertia of the magnet, 

 t the time of vibration of the magnet, ^ the logarithmic decrement, 

 and A and B are the arcs in the method of recoil. 



