178 
Proceedings of Eoyal Society of Edinhurgh. 
of narrow tubing, shaped cj, was next fused to the free arm (see 
fig. 1, a). Afterwards pure sulphuric acid was sucked in, and 
allowed to drain off ; and the narrow part of the tube was cleaned. 
After trying various expedients, the tubes were stopped by a glass 
cover held in its place by a narrow band of india-rubber, which 
prevented any leakage (see fig. 1, h). The tubes were joined by 
keeping the free ends close together by means of a piece of tightly- 
fitting india-rubber tubing (see fig. 1, c). 
A 14-litre bottle filled with water was used as an aspirator, 
and also 5-litre reversible metal aspirator made by Muencke of 
Berlin. When the temperature was below the freezing-point, the 
latter alone was employed, filled with a mixture of water and 
methylated spirit in the proportion of three of the former to one of 
the latter. The balance was a short-beamed one made by Yerbeek 
and Peckholdt of Dresden. It was kindly lent to me for the winter 
by University College, Dundee. The wet and dry bulb thermo- 
meters were those used in the Ben Kevis observations hung in the 
Stevenson screen. Several check readings were taken from a sling 
thermometer. In the Table all temperatures have been expressed 
in centigrade degrees. 
Method . — Three capped tubes were weighed, one against weights,, 
and each of the other two against this standard tube and weights- 
combined, except in experiments marked f in the table. The stan- 
dard tube went through the same experiences as the other tubes, 
except that the caps were never removed. The other two tubes 
were then joined in series, and a third tube attached which pre- 
vented any vapour from the aspirator reaching the weighed tubes. 
To Aspirator 
Fig. 2. 
