447 
TABLE V. 
No. of experiment. Kind of Oxide. Pressure. 
1 errs B 204 
2 B 191 
3 B 200 © 
4 A 205 
' Experiment 2 differed. from the others in that the silver oxide was 
almost entirely used up. ‘The result is undoubtedly too low. Discarding 
this value therefore and taking the average of the other three the eu 
is 203 atmospheres. 
The preceding calculations have been based on the assumption that tiie 
air in the manometer obeys Boyle’s law. According to the work of 
Amagat® the true pressure of air in the neighborhood of 200 atmospheres 
is, at 0°, about 24 per cent higher than that calculated from the volume 
by Boyle’s law. At 30°, the temperature of the manometer, this correc- 
tion would be somewhat less, say two per cent. Making this correction, 
therefore, we get 207 atmospheres as the equilibrium pressure of silver 
oxide at 445°. 
DOES SILVER SUBOXIDE EXIST? 
M. Guntz°® heated silver oxide to a temperature of 358° for fifty hours 
and obtained a pressure of 49 atmospheres. ‘This he considered the pres- 
sure of equilibrium in the reaction via 
2 Ag,O = 8 Ag+ O, 
and cited as evidence for this view the following experiment. 
In a closed tube were placed known weights of silver and silver oxide 
and a quantity of potassium permanganate sufficient to yield enough 
oxygen to produce a pressure higher than 49 atmospheres at 358°. The 
whole was heated at this temperature for three days. The tube was then 
cooled rapidly and it was found that the silver had gained in weight and 
the silver oxide had lost in weight by amounts which indicated the forma- 
tion of Ag,O in both cases. 
I shall presently show that in all my experiments carried on at tem- 
peratures from 300° to 445°, and with two different samples of silver 
oxide, the suboxide of silver was never formed. It seems difficult to 
reconcile my results with those of M. Guntz. Of course, it is conceivable 
that the suboxide is capable of existence and that it appeared for some 
reason in one case and not in the other, just as occasionally some hydrate 
may be suddenly precipitated from an aqueous solution although it may 
fail to appear many times under apparently similar circumstances. 
Ee Tam ped to belicve that the ex erimanis of M. Ganiz can 
