TRANSACTIONS OT THE PHARMACEUTICAL SOCIETY. 
365 
filters is always nearer to the latter than the former rate, as I shall now 
show. 
To ascertain the velocity of filtration of water through different kinds of filter¬ 
ing media, each medium being placed under the influence of columns of water 
of various lengths beneath it, a Schacht’s filter was made in the following 
manner :—into a glass vessel of about the size and shape of a common tumbler, 
but having a hole in its base,—into a vessel resembling, in short, a flower-pot,— 
a disk or disks of flannel was securely fixed by a cork having many perforations 
at about half an inch from the bottom ; a few inches of narrow glass-tube was 
now fitted by a cork into the hole of the vessel, and a ten-feet length of three- 
sixteenth india-rubber tubing attached to the glass tube. By immersing the 
vessel, mouth downwards in a pan of water, and sucking air out of the distal 
extremity of the flexible tube, the whole apparatus was filled with water in a 
more efficient though certainly more troublesome manner than that proposed by 
Mr. Schaclit. I shall describe a still better method presently. The glass was 
now elevated, mouth upwards of course, to a height of about ten feet from the 
floor and, the level of the water in it being kept constant by a proper supply, 
water allowed to flow off by the india-rubber tube, at different distances from 
the surface of the liquid, the tube being, for this purpose, simply held in the 
shape of the letter U, and its orifice raised or lowered so as to be one, two, three, 
four, five, six, seven, eight, or nine feet from the level of the water in the filter¬ 
ing vessel. The time of flow was at each distance exactly sixty seconds, and 
the following were the amounts obtained :—- 
Through. flannel 
1 foot . 
. . 9 ounces . . 
. 7 ounces. 
9 
J-l 
feet . 
. . 15 
11 
• 11* 
n 
3 
ii 
. . 20* 
11 * 
. 15* 
ii 
4 
ii 
. . m 
11 
• 19-| 
ii 
5 
n 
. . m 
11 * 
. 23 
ii 
6 
ii 
. . 30* 
11 * 
• 25* 
ii 
7 
ii 
Q9A 
• • ^2 
11 
. 27* 
ii 
8 
ii • 
• • O'J r> 
„ • 
. 30 
ii 
9 
ii 
• • 37* 
n 
. 32 
ii 
Through close black coat-cloth. 
A glance at this table is sufficient to show that in these experiments the rate 
of flow is far more nearly what it would be if due only to the acceleration of 
motion in gravitating bodies, obtained with liquids, by the use of a long column, 
than if it were simply due to the excess of atmospheric pressure above, over that 
below the filtering medium. The following table shows the quantities that 
would be obtained if the flow were hydrodynamic, if it increased in an aerostatic 
proportion and the quantities actually yielded in the case of the flannel filter, 
the actual rate at one foot being the assumed unit for hydrodynamic and aero¬ 
static rates of increase:— 
Hydrodynamic rate. Aerostatic proportion. Actual rate. 
At 1 foot ... 9 ounces ... 9 ounces ... 9 ounces. 
2 feet . . . 12f- ,, .... 18 ,, ... 15 „ 
q Hu 97 uaa 
U ,, ... 11 ... If 
height of about thirty-four feet. But the pressiwe on the outside of the plate or the tub is 
still fifteen pounds on the square inch ; and thus is demonstrated the fact, that the only in¬ 
fluence of a column of water on the closed and upper end of a long tube or other vessel, is 
to so oppose aerostatic pressure, exerted within the vessel, as to enable the aerostatic pressure 
outside the vessel to manifest itself. Roughly speaking, every foot of such a column devc- 
lopes half a pound of aerostatic pressure on the outside of a vessel. 
