Oxidation of Iron in the presence of Vartry Water. 149 
It seemed worth ascertaining whether the presence of already- 
formed rust determined a further production of oxide under the 
conditions of the last described experiments. Accordingly, I sub- 
stituted for the bright piano wire, pieces which had been allowed 
to rust by exposure to moist air, and of which one half was sub- 
sequently cleaned. When these were placed in solutions of nitrate 
and nitrite, and submitted to a temperature of 156° C., no further 
production of oxide could be detected. 
I have also investigated the behaviour of iron in these solutions 
at higher temperatures, and I have ascertained that at tempera- 
tures not exceeding 270° C. (the highest temperature I could 
conveniently command) iron is not oxidized in air-free solutions 
of potassium nitrate or nitrite. This temperature greatly exceeds 
that of steam boilers, as it corresponds to a pressure of nearly 700 
lps. on the square inch. 
These results fully justify the conclusion that iron does not 
reduce potassium nitrate or nitrite in solution at temperatures 
below 270° C. That a contrary opinion could have been arrived 
at is only explicable on the supposition that the method employed 
for excluding atmospheric oxygen from the experimental tubes 
was not effectual. When water containing a small quantity of 
saline matter is used in steam boilers no scale is produced, conse- 
quently the plates are quite unprotected from the action of the 
oxygen with which they come in contact, under circumstances 
highly favourable to oxidation. Boilers that are in constant use 
are proportionately less injured than those that are frequently 
allowed to cool; for, in the latter case, the dissolved oxygen is 
renewed by the exposure of the water to the air which enters the 
boiler, Vartry water taken from the service tap in the Laboratory 
of the Royal Dublin Society in December, at a temperature of 
4° C., contained per litre 26°46 cc. of dissolved gases, having the 
following composition :— 
Nitrogen, : : 5 . 5 . 2 67-46 
Oxygen, : : : : - : : 30°50 
Carbon dioxide, . ‘ F : ; a 2-04 
100-00 
It would take nearly 4,000 gallons of this water to yield enough 
dissolved oxygen for the conversion of 1 Ib. of iron into rust; but 
