A. W. Wright—Distillation of Mercury in Vacuo. 488 
tubes by the pumping a in g, and after a short time each 
drop falls with a sharp click in the tube. The construction of 
the part 4 makes it easy to obtain sie mercury from the spi 
beginning of the operation, an advantage not furnished by the 
other forms of the apparatus mentioned. 
In adjusting the height of the en a, allowance must be 
made for the tension of the mercury vapor in the upper portion 
of the tube. The cooling effect of the condensing tube, ¢, is 
such that this is usually from four to six millimeters, and it 
rarely or never exceeds one centimeter he temperature of 
vaporization corresponding to the latter tension is less than 180°, 
as, according to Regnault’s results, this is the temperature at 
which the vapor has a tension of eleven millimeters. The low 
temperature is of itself a matter of importance, both as repiias 
economy in the application of the heat, and as diminishing the 
probability of volatilization of any substances which the mer- 
cury may contain as impurities 
The apparatus here escribed, when in use, consumes from 
one-third to one-half the amount of gas required for an 
ordinary Bunsen burner. The mercury does not come into 
active ebullition, but vaporizes quietly and entirely without 
shocks. The rate of distillation varies of course with the heat 
applied, but is from four hundred to four hundred and fifty 
grams per hour. After the burner is once adjusted the appa- 
ratus requires no attention and may be left to itself for hours, 
care being taken that the cistern contains sufficient mercury. 
hen out of use the tubes are left with the uae in them, 
remaining thus exhausted and ready for use at any tim 
s the mercury in the bulb and the tube } retains “all the 
impurities left behind in the process of distillation, these may 
at length accumulate in such quantity as to interfere with the 
proper operation of the apparatus, and to necessitate their 
= oval. This is not aly to occur for a long time unless the 
ercury used is excessively impure. But when the removal is 
“The apparatus in n operation has one cies satisfactory 
