50 W. D. KELLER 
dip of the rocks, and the net result is the lowering of the rate of in- 
crease of total resistivity. 
The circuit was also partitioned by using the auxiliary potential 
electrode, P3, located at the formation contact. The data are repre- 
sented by curves WN and S, which show the values of the potentials 
on the north and south halves, respectively, of the potential circuit. 
They indicate nicely that the north half of the circuit is only about 
two-thirds as resistant as the south side, which is wholly in accord 
with the geology. 
There is no indication of the prevailing structural condition in 
curve 69, and if the rocks were covered, this contact would be passed 
over unknowingly. A partitioned circuit, however, would give warn- 
ing, at least, that the geology was changing, if not a fair indication of 
the nature of the components as a whole. 
BEHAVIOR OF MULTIPLE ROCK SYSTEMS 
After a consideration of the resistivity of a single type rock, the 
next logical step is toward a system of two. Suppose that 25 feet of 
bed A underlie 75 feet of bed B and a depth determination is made 
to 100 feet. What will the resistivity figures for the 76- and 100-foot 
depths be? Will the 100-foot figure be the same as that for 100 feet of 
B, 100 feet of A, 75 feet of B, 25 feet of A, an average of 25 feet of 
A and 75 feet of B, or any other combination? The field answer is 
shown in the following paragraphs. 
HIGH OVER LOW RESISTIVITY 
Graph No. 11 carries two resistivity curves of 20 feet and 75 feet 
of limestone over 80 feet and 25 feet of shale, respectively. In curve 45 
the resistivity starts at 46 units and rises in a normal limestone slope 
to an 18-foot depth and then drops to a shale level. The electrical 
prediction of shale at 18 feet, that is, the break in the curve, with 
actual occurrence at 20 feet, is satisfactory. 
In 44 the curve is typically that of limestone to a depth of 70 feet, 
where it falls consistently to that of shale. The electrical prediction of 
70 feet is apparently 5 feet in error, but the field notes on this set-up 
state that the current electrodes were 3 feet below the instrument 
from the 100-foot mark out. Prediction is again within 2 feet of actual- 
ity. 
The two curves are clearly illustrative of the fact that the break in 
the curve follows rather closely the change in the rocks. Instead of 
designating the maximum or minimum in the curve as the location of 
the geological change, it is often customary to split the slope on a curve 
572 
