MAGNUS ON THE EXPANSIVE FORCE OF STEAM, 225 
an inch internal diameter, so that no correction was requisite 
on account of capillary depression ; the same was the case with 
the tube a 6 d in the box, the parts a d and 0 e, fig. 2, in which 
the surfaces of the mercury were preserved, being more than 
half an inch wide. 
When I began these experiments, the mercury in the arm of 
the gauge p qg, in which the air was rarefied, altered so much 
after a short time that it adhered to the sides, and its surface no 
longer reflected. I soon perceived that this change was caused 
by the aqueous vapour which got there on diluting and again 
letting in the air; on this account I inserted a chloride of cal- 
cium tube at / m, fig. 1, so that this evil was entirely avoided. 
To avoid having to arrange the telescope of the kathetometer 
specially for each reading off, the various objects to be observed 
must be placed at equal distances from it. This equally refers 
to the tube a 0 d in the box, as well as to the gauge and the 
barometer. It would however have been difficult to have sta- 
tioned the air thermometer at the same distance. In order to 
be able likewise to read this off by means of the kathetometer, 
I have applied to the object-glass of the telescope a convex 
glass, which may be readily prefixed. In this manner it was 
possible to observe the air thermometer, which was stationed 
nearer, without any further alteration in the telescope*. 
The mode of proceeding in the experiments was as follows: 
—As soon as the temperature in the case P Q R had become 
constant, which was indicated by the air thermometer, and like- 
wise by two quicksilver thermometers, v and w, fig. 1, affixed for 
this purpose, the air in the tubes / ¢ g b In was slowly rarefied. 
An assistant looked into the case and noticed when the mercury 
stood at the same height in both arms of the tube ad 6. 
An accident frequently occurred here which cost me much 
time and trouble ; the tube, for instance, had always to be moved 
by knocking, in order that the water and mercury should be- 
come loosened in the closed arm of the glass, nevertheless it fre- 
quently could not be effected even when the rarefaction of the 
air had become several, sometimes even ten inches less than the 
expansive force of the vapour at the temperature existing. The 
* It is indeed possible that by prefixing this glass the direction of the axis of 
the telescope is slightly altered; however, perfect parallelism is only requisite 
for the notations which are made either with prefixed glass, or for those made 
without it. 
