088 PROFESSOR ANDERSON ON THE PRODUCTS OF THE 
length the bottom of the fiask may be heated nearly red hot, while a very insig- 
nificant quantity of oil distils up. In performing this process, glass flasks were 
corroded by the caustic potash long before the action was complete, and it was 
found very convenient to employ copper flasks made by the electrotype process. 
A plaster of Paris mould was taken from a glass flask of convenient size and 
shape; and from that a wax cast was made and electrotyped in the usual way. 
After about a week the copper was sufficiently thick for use. In such flasks 
pyrrol was boiled for a day or two with caustic potash, the heat being raised as 
high as an Argand or Bunsen’s gas-lamp would bring it. A bent tube was then 
fitted into the mouth of the flask, and the heat again applied, so as to distil off all 
the oil that could be obtained. The distillate had the smell of pyrrol mixed more or 
less distinctly with that of picoline, and the preponderance of the latter smell 
depended on the quantity of potash having been sufficiently large to retain the 
true pyrrol, which, however, it was not possible to do entirely, even when a very 
large excess of potash was used. When the whole of this oil had distilled, the 
bent tube was removed from the mouth of the flask, and the still fluid potash 
poured out on a copper plate. 
On cooling, it solidified into a hard white mass with a yellowish tinge, which, 
when perfectly dry had no smell, but it was only necessary to breathe upon 
it to cause it to exhale a delightful etherial and fragrant odour, not unlike 
that of chloroform, but softer and less pungent. When thrown into water 
the potash gradually dissolved, and a transparent and colourless oil col- 
lected on the surface of the solution, from which it was separated either 
by a pipette or by distillation. The potash solution on saturation with sul- 
phuric acid evolved the smell of a fatty acid, and when distilled, yielded a 
fluid which reddened litmus strongly, and had a smell resembling that of 
valerianic acid. The distillate was saturated with carbonate of soda, and 
the solution evaporated to complete dryness and extracted with absolute alco- 
hol. The alcoholic fluid was again evaporated, the residue dissolved in 
water, and a quantity of solution of nitrate of silver insufficient for com- 
plete precipitation added to it, and the precipitate was collected on a filter 
and washed. Another quantity of nitrate of silver was then added, and the 
precipitate collected, and finally, enough of the nitrate was used to throw the 
remainder of the fatty acids in the fluid. In this way three different silver 
salts were obtained, which were separately analysed. The first precipitate 
gave :— 
6-273 grains of silver salt gave 
6510  ... carbonic acid and 
2452.  .., water. 
5°192 grains of silver salt gave 
2697. «.. . silver. 
