542 i^Erguson: oxidation of lava 



For the steam experiments the steam-generating apparatus 

 was attached to that end of the porcelain tube which projected 

 5-6 cm. from the furnace. This apparatus consisted of a boiler 

 and pre-heater. The latter also acted as a trap for liquid water. 

 The boiler had two valves so that the water in it could be boiled 

 for some time to expel all oxygen before an experiment was 

 started. The rest of the apparatus was so arranged that it 

 could be initially swept out with pure nitrogen and then, after the 

 steam had been shut off at the close of an experiment, could be 

 again filled with nitrogen. The water that condensed in the cold 

 portion of the porcelain tube was wiped out before the boat was 

 withdrawn. During the wiping-out process and the cooling of 

 the charge an atmosphere of pure nitrogen was maintained. 



Typical experiment. — Five grams of lava powder were placed 

 in the platinum boat and the boat placed in the cold end of the 

 porcelain tube. Nitrogen was then admitted to the apparatus 

 and all the air swept out. The boat was then pushed into the 

 hot portion of the porcelain tube. The nitrogen was partly 

 shut off and steam admitted. When a good flow of steam was 

 obtained through the furnace the nitrogen was entirely shut off. 

 The porcelain tube was so tilted that the water condensing in 

 the colder portion of the tube, after passing the hot zone, would 

 drain away from the furnace. After the experiment had run for 

 the desired time the steam was gradually replaced by nitrogen. 

 The liquid water in the end of the porcelain tube was wiped out 

 (a good flow of nitrogen prevented the access of air during this 

 operation) and then the boat was withdrawn to the cold portion of 

 the tube and, when cold, removed. A part of the charge was ana- 

 lyzed for ferrous iron by the modified Pratt method as used by 

 Washington. ^° Calcium phosphate as recommended by Gage 

 was used to aid in the determination of the end points in the 

 permanganate titration. ^^ It should be noted that if any of the 

 reducing gases present in the rock dissolved when the rock 

 sample was brought into solution, they would be titrated and 

 would appear as ferrous iron. 



1° Washington, H. S., The chemical analysis oj rocks, 2d Ed., p. 136-138. 1910; 

 3d Ed., p. 186-190. 1919. 



" Journ. Amer. Chem. See. 31: 381-385. 1909. 



