FERGUSON: OXIDATION OF LAVA 54I 



represent the reaction between water and the ferrous iron in the 

 silicates might be quite different in character from the equation 

 given for the pure oxides, and an assumption of analogous be- 

 havior in these cases appears to be a somewhat risky matter.^ 

 Indeed, Thaddeef's^ results upon the oxidation of olivine by 

 ignition in air indicate that the oxides are much more easily 

 oxidized than is the ferrous iron of the olivine. He found that 

 only two-thirds of the iron present in this silicate could be oxi- 

 dized by such ignition, whereas it is common knowledge that 

 ferrous oxide would be readily changed to ferric oxide under 

 similar treatment.^ 



My results qualitatively confirm the observations of Thad- 

 deef in this respect and were carried out as follows: 



Materials. — A piece of lava from the crater floor of Kilauea 

 was crushed in a hardened steel mortar and the part that passed 

 the twenty but not the forty mesh sieve (0.38-0.86 mm.) and 

 also the part that passed the two hundred mesh sieve (less than 

 0.074 mm.) were used. 



The nitrogen was made from sodium nitrite, or by removal 

 of oxygen from air by copper, and was passed through the usual 

 purifying and drying train. 



Apparatus and procedure. — The furnace was a platinum- 

 wound resistance furnace of the type usually employed in this 

 laboratory. The temperatures were determined by means of a 

 platinum-platinrhodium thermoelement and a suitable potentiom- 

 eter set-up. A long porcelain tube was placed in the furnace 

 so that it projected twenty or thirty cm. out of the furnace at 

 one end and five or six cm. at the other. The charge was placed 

 in a platinum boat in the porcelain tube, and the boat had at- 

 tached to it a stiff platinum wire which enabled one to insert the 

 boat into the hot portion of the tube or to withdraw it into the 

 colder portion at will. 



^ Such an assumption appears to have been made by Brun, who states that water 

 vapor du-ectly oxidizes the ferrous siHcate to magnetite. His statement is not con- 

 firmed by analyses of the soHd products after reaction with water, but rests on the 

 change in color of the rock and the presence of hydrogen, together with CO and CO2, 

 in the gases evolved. Arch. sci. phys. nat. 41: 404. 1916. 



8 ThaddeEF, K. Z., Kryst. 26: 77. 1896. 



9 SosMAN, R. B., and HosTETTER, J. C, Joiu-n. Amer. Chem. Soc. 38: 820. 1916. 



