er
bearing rocks of the well. Mr. Carll, in III. Pennsylvania Second
Geological Survey, has discussed this subject very fully, and has made
estimates of the quantity of oil that the sand rock can hold and
deliver into a well; also, T. Sterry Hunt, in his _Chemical and
Geological Essays_, has made deductions as to the petroleum contained
in the Niagara limestone that outcrops about Chicago.

While the experiments and conclusions of these geologists go to prove
that these rocks are capable of holding the oil, there are on record
no facts as to the phenomena of its flow, other than by capillarity,
through the rock. To obtain some data of the flow of liquids under
pressure through certain oil-bearing stones, series of tests on small
pieces were made. These tests were carried on during this spring, and
many results quite unlooked for were obtained. When crude oil,
kerosene, or water (river or distilled) was forced through the
specimens, the pressure being constant, the rate of flow was variable.
At first, the amount flowing through was large, then fell off rapidly,
and when the flow had diminished to about one-quarter of its original
rate, the decrease was very slight, but still continued as long as
measurements were made, in some cases for three weeks.

When using crude oil, this result was not surprising, for, as the oil
men say, crude oil "paraffines up" a rock, that is, clogs the minute
pores by depositing solid paraffine (?); but this so-called
paraffining took place, not only with crude oil, but with refined oil,
and even with distilled water.

The only explanation as yet is, that liquids flowing under pressure
through rock on which they exert little or no dissolving effect,
instead of washing out fine particles, tend to dislodge any minute
grains of the stone that may not be firmly held by cement, and these
block up extremely fine and crooked pores in which the fluid is
passing.

Several tests indicated that this blocking up was largely near the
surface into which the fluid was passing. When this surface was ground
off, even 1/50 of an inch, the flow increased immediately nearly to
the original rate.

Reversing the flow also had the effect of increasing the rate, even
above that of any time previous.

With the moderate pressures used--from 2" to 80" of mercury--the
results show that the rate of flow, other things being equal, is
directly proportional to the pressure.

The porosity of rock is not always a cri