TRANSACTIONS OF TIIE PHARMACEUTICAL SOCIETY. 
369 
Thinking that the friction of the water in such a length of tubing might in¬ 
fluence the numbers, I performed nine experiments with a ten-feet length of 
tubing, with the following results:— 
At 1 foot .... 
. 184 
ounces. 
9 
feet .... 
. 31 
Q 
o 
D .... 
. 40 
4 
D .... 
. 49 
•) } 
5 
,, .... 
. 57 
6 
i j .... 
. 66 
•) J 
7 
• 
,, .... 
. 77 
8 
,, .... 
. 87 
n 
9 
,, .... 
. 94 
,, . .... v ,, 
ITow ~ five minutes in each experiment. 
The results at three, six, and nine feet are sufficiently near to those in the 
previous series of experiments to indicate that the amount of the U-shaped por¬ 
tion of the apparatus does not materially influence matters. 
From these last two series of experiments it is again obvious that the nature of 
the action of such a filter, even though it present an unusually large amount 
of resistance to the passage of liquids, is still, even apparently, partly hydro- 
dynamic. (I believe that actually it is wholly hydrodynamic for reasons I shall 
give presently.) 
I was not altogether prepared to find that the rate of flow continued to in¬ 
crease beyond thirty-four feet; indeed, I was somewhat surprised that the rate 
should increase, as it does, after about twenty-five feet. For the air which 
comes out of solution in the water at these diminished pressures expands of course 
considerably at the moment of evolution, and destroys (apparently) the con¬ 
tinuity of the long column with the filtering medium. It does this as the 
length of the column approaches thirty feet, even though the capacity of the 
air-chamber above the medium be nearly one pint. And I had supposed that, 
with the destruction of the continuity of the column, hydrodynamic increase of 
rate would cease. On watching the flow, however,,in the glass tube a few 
inches below the medium during the use of the forty-two-feet column, I became 
convinced that, although the column did not at that point occupy the whole 
area of the tube, yet its continuity was not sufficiently impaired to totally 
interfere with the advantages resulting from the use of that length of column 
below the filter. In short, the column Avas not broken, the rapidity of flow at 
this great length being sufficient to enable it to maintain its integrity. If this 
explanation of the increase in the rate of flow beyond aerostatic-pressure dis¬ 
tances be the correct one, that increased rate is a fresh means of proof of the 
part played by hydrodynamic action in the instrument. 
With regard to filtration in which gravitation is aided by hydrodynamic 
pressure, I would say, finally, we must regard the action of instruments for this 
purpose as, normally 7 , identical with that of the water supply of those houses in 
which there is a cistern in the roof and a descending supply 7 -pipe. The rate of 
flow from such a pipe is, we all know, in proportion Avith the square root of the 
distance from the orifice of outflow to the surface of the Avater in the cistern. 
Place noAv a filtering medium in the upper part of the Avhole column of w T ater 
(say near the bottom of the cistern), and you have a “ Schacht’s filter.” Di¬ 
minish the porosity of the medium, and you interfere Avith the normal rate of 
flow, until the medium is impervious. The pressure on the upper surface of the 
medium is now Avholly static ; to a small extent hydrostatic, but chiefly aerostatic. 
Remove this medium and place it at the bottom of the column of Avater, and you 
have the so-called, but Avrongly-termed “hydrostatic filter.” Diminish the 
porosity of the medium, cripple it as a filter, and you interfere Avith the normal 
