CHEMISTRY: RICHARDS AND COOMBS 
405 
only to be observed when the meniscus appeared to be precisely tangent 
to the edge of the screen. 
Both of these precautions have been more or less fully heeded by 
others; but another precaution, the determination of the diameter neces- 
sary for the larger tube in order to secure perfectly flat surface, has been 
often overlooked. We found that a tube over 35 mm. in diameter was 
required, and that even into this wide tube it , 
was not permissible to insert a capillary; for such 
an insertion acted as another basis of support for 
the liquid and caused appreciable rise. By ac- 
tual measurement we found that the capillary 
rise in a 20 mm. tube, counting only from the 
middle of the bottom of the meniscus, was over 
0.5 mm., and the addition of a capillary tube in 
the middle of this raised it at least 0.3 mm. more. 
As apparatus of this sort has been used by most 
experimenters on surface-tension, most of the 
capillary rises which have been reported are in 
the neighborhod of 1 mm. too low — an error 
which accounts for a large part of the discrepan- 
cies betw^een different methods. 
Another error which does not seem to have 
received sufficient attention is that due to the 
weight of the Kquid in the finer meniscus above 
its lowest point. The equation of Poisson, which 
is usually used for calculating this weight, gives 
an absurd result with tubes as wide as 1 cm., 
and therefore must be rejected. Another equa- 
tion, that of Desains, gives a result for fine tubes 
which is not plausible; hence this also seemed un- 
worthy of confidence. A careful measure of the 
height of the meniscus between its lowest point 
and its line of contact with the fully wetted walls 
showed that in very fine tubes this height is al- 
most exactly equal to the radius, and that, therefore, the meniscus is 
here essentially hemispherical. As the tube widens, the hemisphere be- 
comes somewhat flattened, and for moderate radii it appears entirely 
safe to apply, as a correction to be added on account of the menis- 
cus, one-third of the meniscus height as actually measured. This 
method of correction was shown to give consistent results with tubes of 
different diameters. 
APPARATUS IN DIA- 
GRAMMATIC SECTION. 
(THE LOADED SINKER IS 
TO DIMINISH THE NECES- 
SARY VOLUME OF 
LIQUID.) 
