130 SITIDON HARRIS 
the V4 bed at the point where the path through d intersects the line 
drawn from shot point B, and extending this line to intersect the 5.2° 
dip line drawn from shol poini A, the surface of the high speed bed 
was completed. 
A good method of checking the work is to select some point which 
lies on one of the V4 curves and trace its path from the station to the 
shot point. By measuring the sections of this path and dividing them 
by their respective velocities, the time of arrival of the first shock at 
the station located on this path is computed. This calculated time 
should check the seismic time on the time-travel curve. Several checks 
of this sort were made when the profile illustrated in Figure 4 was 
first computed. All the checks came within .oo5 of a second of check- 
ing the data. This accuracy was within the limits of error of the in- 
struments, which were Petty® condenser type seismographs equipped 
with a string galvanometer recording system. Chronometer marks 
were made on the seismograms which enabled the time of arrival of 
the first shock to be read to .o05 second. 
In summary, it will be seen that in order to arrive at any satis- 
factory conclusions as to the depths of consolidated deposits from 
refrdction shooting, it is necessary to shoot extensively over the area 
under consideration and then plot the entire data and arrive at ap- 
proximate true velocities of the deeper beds, before attempting com- 
putations of depths. Then each profile must be taken up separately 
and a rough contour of the subsurface bed of most interest (usually 
the deepest one) sketched from the time-travel ‘curves before actual 
calculations are started. This procedure gives the computer an outline 
to work from, and is of great aid in keeping the results within reason. 
In the case of the example illustrated in this paper, all six profiles were 
worked out and a subsurface contour map made of the limestone 
surface. The contour map showed a decided high in the surface of the 
limestone. This high was later drilled into and the difference between 
the well log and the calculated depth of the high was only 4% of the 
calculated depth. 
The fact that McCollum and Snell? found that the velocity of the 
sound waves is not the same in all directions in consolidated deposits 
introduces an error in the method described in this paper. However, 
if the velocity across the beds is determined from other methods, 
these values can be used in, determining the time required to cross the 
beds. It might be said that in regions similar to the one illustrated in 
6 Made by the Petty Geophysical Corporation, San Antonio, Texas. 
7B. McCollum, and F. A. Snell, Physics, 2, 174, 1932. 
790 
