ORE DEPOSIT THEORIES. 213 
higher level in the main channel than in the smaller open- 
ings. This excess of upward flowage in the main openings, 
which is the result of expansion aldne, and may at first be 
very small, would constantly traverse rocks of lower tem- 
perature, so that the temperature of the water would 
always be slightly higher than that of the surrounding rock ; 
that is, it would be higher than the temperature of any 
water, stationary, descending, or more slowly ascending in 
those surrounding rocks, and there would be induced a 
hydrostatic head due to difference of temperature in the 
waters contained in openings of the first order and those of 
higher orders. Thus, when the smallest difference of hydro- 
_ static head was induced between the main channels and the 
smaller openings, there would be induced a constant flow of 
water from these smaller openings to the main channel, 
which would aid in causing the upward movement to cooler 
rocks and further accentuate the difference of temperature, 
and therefore of pressure. The whole cycle would be cumu- 
lative in its action, and given any movement whatever in 
the underground water system, an upward circulation in the 
direction indicated must ensue. Whether the main 
openings communicated directly or indirectly with the surface, 
the flowage in the ascending channels would find relief by a 
surface discharge, or, in some cases, possibly by a gradual 
distribution of the ascending water into openings of higher 
orders, the result of which would be to gradually cool down 
these waters, and return them to the descending channels. 
Thus, we may see that the openings of the first order 
would be occupied with ascending currents under the in- 
fluence of a difference of hydrostatic head induced by the 
application of heat to the water system. The channels of 
the higher orders would, however, be drawn upon to supply 
the flow induced in the larger channels, so that the whole 
body of underground water would be set in motion, the 
larger channels being occupied by ascending waters, and the 
smaller by descending waters. 
It is further evident that the current ascending the few 
large openings of the system of flowage must be balanced 
in amount by the waters descending through the small 
openings. If the total areas of the respective systems were 
the same, the average velocity of flow would be the same 
also, but though it would be hard to give definite proof of 
the fact, it is apparent, almost to demonstration, that the 
sum total of the small openings must be immensely larger 
than the area of the comparatively few larger openings 
occupied by ascending currents. The flow through the 
smaller openings we know must be extremely slow, and we 
