498 
THE EARL OF BERKELEY AND MR. E. G. J. HARTLEY ON THE 
in millimetres per 15 minutes. The rates in this column are those caused by the 
“ guard-ring leak.” 
O Co 
Column (V.) gives the apparent turning-point pressure corrected for the “guard- 
ring leak ” (see explanation of column X.). As already pointed out, the pressures 
registered by the dead weights are not the total pressures on the solution ; the 
pressure of the atmosphere must be added to them. On the other hand, the water 
inside the tube is under atmospheric pressure, so that if there had been no “ solution 
leak ” the pressures noted in this column would be equilibrium pressures. 
Column (VI.) gives the length of time the solution was under pressure. 
Column (VII.) gives the total amount, in grammes, of solid which came through the 
membrane during the experiment. The letters j.v.t. mean “ just visible trace.” 
Column (VIII.). Assuming that the amount of solid given in (VI.) represents a 
corresponding volume of solution, then the numbers in this column give the rate at 
which the water would rise in the gauge, on the further assumption that the solution 
came through the membrane at a uniform rate. 
Column (IX.) gives the increment of pressure, in atmospheres, which was used in 
the experiment. Column (X.) gives the difference between the rates corresponding to 
the two pressures separated by this increment, the one just above and the other just 
below the turning point. The numbers are therefore the rates due to these increments 
of pressure near the turning point. It is from these rates that the correction for the 
“ guard-ring leak ” is derived by assuming that it bears the same ratio to the 
increment of pressure as the numbers in (IV.) do to those in (X.). 
