298 
MR. W. CROOKES OK THE ATOMIC WEIGHT OE THALLIUM. 
spheric pressure forces the glass in, forming- a fine hole. In this way no fragments of 
glass get into the water, as might be the case if the end of the tube were broken off 
after scratching with a file in the usual manner. As much water as is needed is driven 
out by warming the globe, and the water remaining unused may be sealed up again. 
Water purified in this manner was employed in the final crystallizations of all the 
salts used in the investigation, in rinsing out the apparatus, and generally in all opera- 
tions where its employment was likely to increase the accuracy of the result. 
Nitric Acid. 
The apparatus in which the nitric acid was prepared is represented in the accom- 
panying figure (fig. 4, Plate XVI.). a, b, c , d are glass globes, a being about 4 inches 
and the others about 2 inches diameter ; they are connected by fusion with glass tubes 
bent as represented in the drawing, and contracted at the points o, q>, q, r, s. The cylin- 
drical tube e is connected by means of a flexible joint, iiii (fig. 4 b), to the Sprengel 
pump: this joint consists of several pieces of glass tube held closely together by means 
of india-rubber tubing ; the pieces of tubing are fastened with two or three turns of 
silk-covered iron wire. A little glycerin smeared over a joint of this description renders 
it quite air-tight, whilst the apparatus is capable of considerable movement. A mixture 
of glacial phosphoric acid and nitrate of silver in atomic proportions, and coarsely ground 
together in an agate mortar, is introduced into the globe a until about half-full, the 
cylinder e having previously been filled through the opening g with pieces of caustic 
soda. The tubes/ 1 and g are now sealed up at the contracted portions, and the Sprengel 
pump is connected to h. Exhaustion is proceeded with, the apparatus being gently 
warmed at the same time until moisture is no longer visible, and the mercury is as 
high as the vapours present will allow it to rise. The bulb b is then immersed in a 
freezing-mixture, and heat is cautiously applied to the bulb a. The nitrate of silver 
and phosphoric acid soon fuse together to a clear liquid, and vapours of nitric acid mixed 
with nitrous acid are copiously evolved. The mixture froths considerably, and care 
must be taken that none rises so high as to pass into the bulb b. In this and the other 
bulbs nearly all the vapours condense, the small quantity that escapes being caught by 
the caustic soda in e. If, through spirting or inadvertence, some of the solid matter is 
carried over from a into b, the contents of b are easily decanted back into a by tilting the 
apparatus into such a position that the line Jc l would become horizontal : owing to the 
curvatures of the connecting-tubes, no liquid which might be in the bulbs c or d, and 
none of the pieces of caustic soda in e, can get out of their place. When the reaction 
between the phosphoric acid and nitrate of silver has been pushed as far as convenient 
(too strong a heat must not be applied, or the nitric acid is in part decomposed as it is 
liberated), the tube connecting a and b is sealed in a spirit-lamp, and the globe a drawn 
off. The bulb c is now thoroughly washed out with nitric acid by distilling a little over 
from b, letting it condense in c, and then pouring it back by inclining the apparatus so 
that the vertical line in n would become horizontal. The bulbs b and d are now warmed 
