34 SECTIONAL ADDRESSES 



they would be proportional to the variation of field intensity over the 

 limited space occupied by the suspended system, instead of to the full 

 intensity at the station. In the gravity torsion balance they are, in fact, 

 negligible, and they could be made equally so here. 



In order to realise in a magnetic torsion balance the full advantages of 

 the corresponding gravity instrument, we must make the magnet system 

 in the suspension completely astatic, so that, in a precisely uniform field, no 

 couple acts upon it. It is here that the practical difficulties of construction 

 present themselves. The polar nature of magnetism, as contrasted with 

 gravitation, added to the fact that the earth's magnetic field is not every- 

 where nearly vertical, demands a precision of construction which may be 

 unattainable. For example , to reduce to negligible proportions the directive 

 action of a horizontal magnetic intensity of o • 1 8 gauss upon a suspension 

 of the Coulomb balance type, we should require two magnets equal, and 

 remaining equal, in moment to within one part in a hundred million, and 

 with their magnetic axes aligned in opposite directions with an accuracy 

 of one-hundredth of a second of arc. The conditions applying to an 

 instrument ^ for measuring only the horizontal gradient of vertical inten- 

 sity are not, however, so severe and unpromising, and it is in this direction 

 that there appears to me to lie hope of practical realisation. One magnet 

 only is required, suspended with its magnetic axis nearly vertical from one 

 end of a torsion balance beam, and suitably counterbalanced by a non- 

 magnetic load at the other end. This is, indeed, a modification of the 

 form used by Eotvos, which he operated with partial but not complete 

 success. The outstanding difficulties to be overcome relate to the elimina- 

 tion of the effects of torsion in the fibre used to suspend the magnet from 

 the beam arm. If this can be done we shall have, as indicated earlier, 

 means of extending greatly the scope of the magnetic method of surveying. 



Electrical Methods. 

 I have left until last reference to electrical methods, not because they 

 are of less importance, but because I am less familiar with them, and could 

 not speak with any of the authority which comes from practical experience. 

 Accordingly I shall use this opportunity of calling special attention to the 

 work of the Imperial Geophysical Experimental Survey ^ which operated 

 in Australia from 1928 to 1930. This survey, under the leadership of 

 Mr. Broughton Edge, whose extensive experience of electrical surveying is 

 well known, was concerned chiefly with electrical investigations. Until its 

 work began, although it required no great insight to recognise that the 

 basis of electrical surveying was the differences of the electrical conduc- 

 tivity of underground bodies, and that the procedure was to measure 



* The equivalent of the gravity gradiometer, a modification of the Eotvos 

 torsion balance, due to Shaw and Lancaster- Jones. See Mining Magazine, 

 May 1929. 



^ The survey was instituted as a result of the suggestion of the Geophysical 

 Sub-Committee of the Committee of Civil Research, and was carried out under 

 the joint auspices of the British Empire Marketing Board and the Australian 

 Commonwealth Government. The report, ' The Principles and Practice of 

 Geophysical Prospecting," was published in 1931 (Cambridge University Press). 



