512 



YOSHLJIRO KATO 



Calculations. 



The equation which I have assumed for the relation between the 

 intensity of magnetisation and time under a constant field is 



3 = a- —le~ lt -me-» 1 — ne~ vt -pe~ nt ', 



where ^>/^>>v> 



Now, on drawing the figures exactly as they were given by 

 the observations, I found that the curves had some irregularities, so that 

 the values could not be made use of directly in the calculations of the 

 several constants. I drew, therefore, a smoothed curve, which may be 

 called the mean curve, lying as closely as possible to the actual curve 

 in each case, and from the values obtained from this mean curve, 

 calculations were made. As the process is rather laborious, I took 

 only six of the above experiments. In Exp. 7 and 8, the mean of two 

 successive observations is taken in drawing the curves, but for all the 

 rest the first observation alone is used. 



