210 PROCEEDINGS OF SECTION C. 
available cross-section of the openings connecting the two 
regions. These openings are not confined to the space 
directly between the two regions, but are distributed more 
or less throughout the whole of the space between, surround- 
ing, and below the two regions. Van Hise illustrates 
this law by supposing water to be poured into a well A and 
to flow to a well B, and quoting Professor Schlichter’s theo- 
retical investigations of the motion of underground waters, 
shows that the flowage is not by a direct path between 4 
and B, but by a large number of diverging paths from A 
during the first part of the journey, and by a large number 
of converging paths to B during the latter part of the 
journey. Thus, it would appear that, given an undulating 
. Surface of the ground water level, the whole of the water 
below that level forming a part of thesea of deep underground 
water must be in motion. Secondly, circulation may be 
produced or promoted by differences in temperature be 
tween the ascending and descending columns. The expan- 
sion of water with increase of temperature amounts to over 
4 per cent. between 0° and 100° C. at ordinary pressures. 
Assuming this difference to be the same at higher pressures, 
then, if the difference between the ascending and descending 
columns were 100° C., 4 feet of head in favour of the 
descending column would be produced for each 100 feet of 
depth. The natural increase of temperature in the earth’s 
crust as depth is gained is one of the causes of the differ- 
ence in temperature between the ascending and descending 
columns. For the ascending column flowing from a hotter 
to a colder region is always hotter than the rocks which 
surround it, while the descending column flowing from a 
colder to a hotter region is always colder than the rock 
which surrounds it. At any given depth, therefore, the 
ascending column must be higher in temperature than the 
descending column. The difference in temperature is, of 
course, largely dependent upon the rate of flow of the water ; 
a rapid rate of flow would produce great differences in tem- 
perature, while a slow rate of flow would produce slight 
differences, sinte it would enable the water to assume the 
temperature of the surrounding rock. 
As will be shown hereafter, there is a tendency for the 
descending water to flow comparatively slowly through a 
large number of small openings, and for the ascending 
water to flow comparatively rapidly through a small number 
of large openings. It follows from this that the temperature 
of the ascending water will differ more greatly from the 
temperature of the surrounding rock than that of the 
descending water. It is impossible to form any estimate 
