1388 WILLIAM M. BARRET 
indicated in Table III, the respective differences between the observa- 
tions being included to facilitate comparison. 
TABLE III 
Station Number DZ in Gammas DZ! in Gammas DZ-DZ' 
1928 1931 
I ° ° ° 
2 aa —10 — 6 
3 5 —12 eal 
4 —10 36 ar é 
5 Sa) ar © ° 
6 +36 +16 via 
of +32 +41 ar ©) 
It is seen that, with the exception of station No. 6, the variation 
between the surveys does not exceed the probable instrumental error, 
the difference at 6 being of little consequence, as this position is well 
removed from the area of potential disturbance. It is interesting to note, 
in connection with these data, that a circle having a diameter of 3 
miles may be circumscribed about the field so as to enclose every well, 
and that the mean variation between the two series of observations, for 
the three stations within the circle, is only 1 gamma. Certainly, these 
results do not indicate that the original pattern of the vertical field has 
been altered by the presence of the additional wells. 
CONCLUSION 
We have seen that it is possible, with our derived formulae, to de- 
lineate with fair approximation the distribution of the magnetic elements 
within the surficial zone that surrounds an element of magnetized casing. 
These equations were not developed for the purpose of determining the 
appropriate correction belonging to an effected observation, but rather 
to explain and amplify our experimental evidence relating to the phe- 
nomena. By a more elaborate process we might define the observed 
effects with precise fidelity, although it is well to remember that the 
geophysicist is primarily interested in the attenuation of the disturbance 
and for this purpose the present derivations are entirely adequate. 
In concluding, the writer wishes to summarize the following perti- 
nent deductions. 
1. The magnetic disturbance caused by an isolated string of mag- 
metized casing is of a highly localized character. 
2. The maximum effect, which occurs in the immediate proximity 
of the casing head, varies as the physical proportions of the pipe, increas- 
ing very nearly as the cross-sectional area of metal at the casing head. 
106 
