622 LAMONT: MAGNETIC RESULTS AT MUNICH. 
the magnets, calculation gives for the series of observations in 
page 218%, 
p= +0135 
g = — 2°360 
r = — 1167 
These coefficients are too large, and we come nearer the truth if 
we assume the magnetism to be concentrated in a point of 
the whole length from the end; we have then to multiply the — 
values of p, g, and 7 by the second, fourth, and sixth power of 
(a = a): In order to find the factor (a — = we have 
only to make the calculated values of p and g equal with those 
found by experiment, page 219; the result is that we may call 
(2 _ ~) = 0°74 nearly. Hence we obtain 
q=- 0°708 
7 — —O191. 
The value of p departs only jth, and the value of g jth from 
the true values, so that we may suppose the value of r also to be 
incorrect only to a small amount. 
If we now compute the corrections for log 2 = we find that 
0:00007 would have to be put for the nearest, and 0°00001 for 
the second distance; for the remaining distances the correction 
falls beyond the fifth place of decimals. We conclude therefore 
that in cases which occur in practice the fourth member in the 
development may be neglected. 
* The observations here referred to are contained in a notice printed in the — 
same part of the dnnalen, and’entitled, “On the Determination of the Law ac- 
cording to which Magnetism is distributed in Steel bars.” 
