ON THE LIMITING THICKNESS OF LIQUID FILMS. 
649 
resistance coils containing a resistance of one megohm. The binding screws forming 
the terminals of this resistance could be connected with the electrometer. The gold 
needles could also be connected with the electrometer. Thus the difference of poten¬ 
tial between the two needles in the film could be compared with that between two 
other points in the same circuit separated by a known resistance, and the resistance of 
the film between the two needles thence determined. The independent difference of 
potential between the needles, when they were in the film, but when no current was 
passing, was troublesome, but was reduced to a minimum by carefully cleaning them 
with nitric acid at the beginning of each day’s work. Every part of the circuit was 
carefully insulated, and the insulation of the two gold needles was specially tested 
at the end of each set of observations. If the result of the test was not satisfactory, 
the preceding observations were discarded. In the case of no experiment given below 
did the insulation-resistance of the needles fall below 300 megohms. 
The observations were carried out in the following manner. The glass case having 
been thoroughly cleaned and the hygrometer adjusted to stand at from 40° to 50° on 
the scale, when in a saturated space, the cover with its fittings was introduced into 
its place, and the apparatus thus enclosed was placed in the centre of the glass water- 
tank. The ebonite plug X, fig. 3, was then removed, a tube terminating in a small 
funnel introduced, and about 20 centims. of the soap solution poured in. This 
quantity was sufficient to cover the bottom of the box to a height of about 3 or 4 
millims. The tube was withdrawn and the plug replaced. The screw-head V was then 
rotated so as to bring every part of the endless band, into the liquid. This operation 
was subsequently repeated from time to time to ensure the linen remaining thoroughly 
wet. The apparatus was then left to itself for 30 or 40 minutes, which interval 
was as a rule sufficiently long to allow the hygrometric state to become constant. 
The hygrometer index rose at first rapidly and afterwards more slowly to a limiting 
position which it steadily maintained. On one occasion the indication was observed 
at intervals during 36 hours, and did not vary more than a tenth of a scale division. 
The new apparatus has, in fact, enabled us to overcome the chief difficulties encountered 
with the old. We can now maintain the temperature and hygrometric state of the 
space round the film constant as long as we wish. To proceed:—The electrical 
connexions having been completed, the lower ring F was turned to one side by the 
button H, and the pinion C rotated until the mouth of the upper cup just touched 
the liquid. When it was raised, a plane film was formed over its mouth, and this, 
after the ring F was restored to its position, was blown out into a cylinder in the 
usual way. The needles were inserted in the film and adjusted so that their position 
could be easily observed by the cathetometer. The length of the cylindrical film was 
usually about 34 or 35 millims., and the upper needle was placed about 5 millims. 
below the cup B. The distance between the points where the needles pierced the film 
varied in different experiments between 4'6 and 5‘3 millims. There were, indeed, as 
has been stated, three needles on the same support, but only the first and second were 
