1909-10.] Method for Determining Vapour Pressures. 439 
of 0’07°. The irregularity of the individual observations is due to the 
crudeness of the gauge and the lack of special precautions in ascertaining 
the temperatures — both intentional. 
There are three sources of error. That due to capillary ascension of 
the bath-liquid ( + 0‘5 to +0 - 6 mm.) and that due to dilatation of the 
bulb of the thermometer (equivalent to — 05 to —1*2 mm.) act in opposite 
directions. When correction on these accounts has been made, the average 
divergence from the standard values becomes 0T6 mm. or 0*03°. In the 
third place, correction may be made for the local value of the gravity 
constant (here —0*25 mm. per 760 mm.). Taking account of this reduces 
the average divergence to 0‘06 mm. or 001°. 
Boiling-Points at exactly Normal Pressure and at Definite Reduced 
Pressures . — It is very desirable that the chemist, instead of publishing 
boiling-points as observed, and often omitting to give the exact pressure, 
should ascertain the boiling-point in each case at 760 mm. 
Again, boiling-points, under reduced pressure, determined in the ordinary 
vacuum distillation apparatus, are especially liable to very great errors. 
With little extra trouble the submerged bulblet apparatus may be used 
to secure exact boiling-points at definite pressures. Only a few measure- 
ments of vapour pressure near to 760 mm., or near any temperature at 
which the substance is stable, are required. When the temperatures and 
pressures have been plotted graphically, the boiling-points at 760 mm., or 
at convenient reduced pressures such as 20 mm., 50 mm., or 100 mm., may 
be read off. 
Chicago, February 1910. 
r 
{Issued separately June 16, 1910.) 
VOL. XXX. 
29 
