80 R. THRELFALL AND FLORENCE MARTIN. 
oxygen becomes unstable as to its pressure volume relation, and 
that: the equilibrium condition is not attained until after seventy- 
eight hours rest. The slightest change of pressure or volume then 
upsets the equilibrium, and time has again to elapse before a steady 
state is attained. It appears likely either that the oxygen forms 
an allotropic modification, or that it forms some compound with 
mercury or other material present and with which it is in contact. 
It will be noted that according to Bohr, the volume of the gas 
tends to increase below 0:7 mm. indicating that the molecules of 
oxygen are partly split up. In this case, therefore, it would be 
reasonable to infer an increase of oxydising power, and it is 
possibly to this cause that the soiling of the fall tubes of Sprengel 
pumps is to be attributed. It appears worth while therefore to 
try to arrange some chemical.test capable of showing the presence 
of active oxygen. 
The two following test solutions were found to satisfy the 
conditions, though one was more sensitive than the other. One 
condition of course is that the test solution must not have @ 
vapour pressure comparable with 0-7 mm. The first indicator 
tried was a solution of indigo in pure sulphuric acid. This is 
bleached by ozone, but experiment showed that the reaction does 
not afford a very delicate test of the presence of that gas. 
Another solution was therefore tried, consisting of potassium 
iodide and starch dissolved in glycerin. The glycerin was 
carefully dried at a temperature of 260° C. When cool, some of 
it was mixed with a small quantity of powdered potassium iodide. 
A very small quantity of starch was added to the remainder of 
the glycerin, which was then slowly heated till the starch was 
quite dissolved and the liquid again became transparent. When 
cold, this portion was mixed with the potassium iodide solution— 
a solution so prepared is not affected by ordinary oxygen, but on 
bubble of the gas which has passed through an ozone tube turns 
it bright yellow, and three bubbles give it a dark blue, almost 
black colour. This seemed sufficiently sensitive, and was accord- 
ingly adopted. Of course the starch is not absolutely necessary, 
: 
Nie Ea heen Sent 
