CHARLES L. BRISTOL. 215 
a porcelain dish (half-filled with water) held over the 
flame and this oxide film was then changed to an iodide 
film by exposure to hydriodicacid. In practice this was 
accomplished by setting the dish in the mouth of a salt- 
mouth bottle containing hydriodic acid and phosphorus 
acid obtained from the gradual deliquescence of phos- 
phoric triiodide. Another method was to volatilize 
iodine held on asbestos threads and condense it on the 
film. The film becomes an iodide film in both cases. 
Bunsen prepared a table differentiating these films, but 
as the results are very similar to those described later, I 
omit them. 
The next step was taken by Von Kobell, who pub- 
lished the use of potassium iodide as a means of dis- 
tinguishing bismuth from lead, in 1871. He mixed the 
powdered sample with an equal bulk of a mixture of 
potassium iodide one part, powdered sulphur five parts 
and treated the mixture ona long coal with a moder- 
ate reducing flame. If the sample is lead, a rather 
volatile coat of yellow iodide of lead is formed at quite 
a distance from the assay. In samples containing little 
lead, the yellow coating may be masked by the white 
coat of potassium iodide, and if mercury, arsenic or 
antimony be present, they may either yield similar coats 
or may completely mask it. Bismuth under this treat- 
ment yields a beautiful red coat of bismuth iodide, at a 
great distance from the sample ;—/f the test is properly 
made, says Cornwall. This may be masked ; either ap- 
pearing similar to lead or covered by lead itself. To in- 
crease the delicacy of the test, especially in the presence 
of much lead, proceed as follows: The assay is heated 
over an alcohol flame in an open tube about 100 ™™ long 
by 80 ™™ in diameter with somewhat more than its bulk 
of a mixture of potassium iodide one part, sulphur five 
parts. If bismuth be present, its sublimate will appear 
asadull brick red band or irregular streak about 8 ™™ 
99 
