8 PAUL WEAVER 
fore, the geologist can restrict the interpretation of such a sharp 
gravity minimum to a relatively simple picture of a salt dome, or an 
anticline, depending on the shape and sharpness of the minimum. 
Again, although the seismic method is supposed to give a unique solu- 
tion, unknown variations in the formation traversed by the sound 
wave may give several alternative interpretations. 
At the present time we are asking that the geophysical report for 
a salt dome shall include an estimate of the depth to the salt, the area 
of its cross section, and the existence of an overhang on some part of 
the periphery. When these additional questions are asked, the diffi- 
culty of obtaining the information by potential function methods be- 
comes very great. We do not refer to the practical difficulty of making 
very accurate measurements, but to the theoretical difficulty, or 
what we shall call the resolving powers of the method. To refer to our 
desert analogy, as we look across the valley to the hills on the far 
side, we can appreciate that these-hills have certain salients and al- 
luvial fans, but we can not, even with a powerful telescope, see the 
minor topographic details because the air through which the light 
travels prevents obtaining a proper stereoscopic picture. We can see 
the same difficulty if we go up in an airplane on a day when there are 
no shadows. At slight elevations we can estimate visually the height 
of buildings, but at 10,000 feet, we can calculate the height of build- 
ings only by photographs taken several hundred feet apart, and can 
do this accurately only when the photograph is very clear and there 
are no local hazy spots in the air between us and the ground. 
The potential function methods, therefore, will always have a 
limit when we try to make profiles and contour maps of the top surface 
of buried structures because the resolving power of these methods is 
inadequate. Even where the structures which are being studied are 
shallow, this difficulty appears. For example, a single bed which is 
thin and a very good conductor can be mapped by electrical methods, 
but where there are three good conductors at fairly short intervals, 
even if all of them are shallow, the electrical problem becomes ex- 
tremely difficult. Now the limit in resolving power of the potential 
function methods applies not only where there are a number of beds 
which we are trying to map, as in the electrical case just mentioned, 
but it applies also to the kind of surface which a single bed may have. 
For example, to distinguish between a fault and a short steep dip be- 
comes an impossible problem in certain measurements. 
There is a further difficulty in geophysical surveying by potential 
function methods. It is necessary to extend the surveys to a consider- 
able distance because the rate of change of the values measured is im- 
530 
