134 CAKNEGIE INSTITUTION OF WASHINGTON. 



Some results indicative of the behavior of several different furnaces and 

 set-ups are presented for their comparative value and also with the view of 

 indicating the magnitude of some of the effects obtained. 



(18) The melting-points of cristobalite and tridymite. J. B. Ferguson and H. E. Merwin. 



Am. J. Sci., 46, 417^26 (1918). 



The melting-point of cristobalite has been redetermined and found to be 

 1710 ±10° C. This value is consistent with the experimental evidence which 

 was obtained in a study of a portion of the ternary system CaO — MgO — SiOj 

 and which had caused the earlier investigations to be viewed with suspicion. 



Tridymite is unstable at its melting-point, and this unstable melting occurs 

 at 1670 ±10°. Artificial tridymite made from quartz could not be melted, 

 owing to the rapidity of the tridymite-cristobalite inversion, but a sharp 

 melting was obtained with natural material. Since this unstable melting- 

 point is below that of cristobalite, there can no longer be room to doubt that 

 cristobalite is the high-temperature form of silica. 



The determinations were carried out in a furnace specially designed for 

 the purpose, which can be heated to and maintained at temperatures slightly 

 above 1700° C. The furnace is constructed on the cascade principle: the 

 inner coil is of an alloy of platinum with 20 per cent rhodium, the outer coil 

 of pure platinum. The two coils are insulated from each other by well-burned 

 magnesia powder, and the inner coil is wound on a helically grooved magnesia 

 tube. 



(19) The equiUbrimn between carbon monoxide, carbon dioxide, sulfur dioxide, and free 



suKur. John B. Ferguson. J. Am. Chem. Soc, 40, 1626-1644 (1918). 



This paper contains an extended account of the investigation of the reaction 

 CO-f 3^S02<^^^C02+MS2, a partial summary of which appeared last year. 

 (See "The equilibrium between carbon monoxide, carbon dioxide, sulfur 

 dioxide, and free sulfur," reviewed on p. 147, Year Book No, 16,) The study 

 was undertaken primarily as a part of a comprehensive study of certain gas 

 reactions and their role in volcanic activity, and this particular reaction was 

 selected because it afforded a direct means of determining the free energy 

 or thermodynamical potential of sulphur dioxide, one of the most important 

 of the volcanic gases. 



Two methods of attack were employed in the research : the stream method 

 and a semi- stream method. The former was found to be inadequate chiefly 

 because of subsidiary reactions which occur at low temperatures, and the 

 latter was devised to overcome this source of trouble. In this latter method, 

 a volume of gas is maintained at a constant uniform temperature for sufficient 

 time to enable the mixture to attain an equilibrium condition, and then a 

 part of the gas is quickly withdrawn into a cold evacuated vessel. The sam- 

 pling takes from 0.1 to 0.2 of a second, and yields samples ranging from 15 to 

 25 c.c. in volume for analysis. Equilibrium can be reached at 1000° C. with 

 the aid of a contact mass of platinized crushed porcelain, and barely reached 

 without such a catalyst at 1180° C. In all equilibrium mixtures obtained at 

 or above 1000° C, the presence of COS could not be detected. This fact not 

 only greatly facilitated the analytical work in the later stages of the research, 

 but established beyond a doubt the efficacy of the sampling method, since 

 the velocity of the reactions producing the COS greatly exceeded the velocity 

 of the reactions under investigation. 



The composition of the equilibrium gas-mixtures obtained, with the corre- 

 sponding initial gas-tnixtures, both calculated and actual, and the resulting 

 equilibrium constants, are given in tabular form, and the thermodynamical 

 calculations based thereon given in detail. The latter give an average mean 

 value of 3,99 for the thermodynamical constant I and —22510 calories for 



