532 



PROCEEDINGS OP THE AMERICAN ACADEMY. 



(270) and obtained a long series of records for gap-widths up to about 

 30 mm. When the gap was closed and the current had been sent first 

 in one direction and then in the other alternately for a number of times, 

 I got the results given in Table XI for the growth of a current (A) 

 which had the same direction as the next preceding current, and the 

 results given in Table XII for a current (B) which had a direction oppo- 

 site to that of the preceding current. 



