August 17, 1906.] 



SCIENCE. 



203 



In the earlier work of one of us, it was 

 found impossible to procure crystals from 

 solutions of the sulphate having a basic 

 ratio greater than BeO/SOg. Crystals 

 have now been obtained from solutions 

 with a ratio as high as 3BeO/2S03. These 

 crystals are in every case the normal tetra- 

 hydrate, and by their separation the 

 mother-liquors are rendered more basic. 

 Repeated attempts to obtain the hexa-hy- 

 drate described by Levi-Malvano (Ztschr. 

 anorg. Cheinie, 48, 446) have resulted in 

 failure. Although the conditions described 

 by that author were faithfully followed and 

 other methods used, the tetrahydrate in- 

 variably separated. A series of experi- 

 ments on dialyzing the sulphate solutions 

 of a basic nature showed a tendency for 

 the solution to become much less basic by 

 dialysis and the hydroxide was left behind. 



The Theory of the Dissociation of Gases 

 around Highly Heated Wires: Irving 

 Langmuir. 



In a previous paper the author gave the 

 results of experiments on the dissociation 

 of water vapor and carbon dioxide, made 

 by passing the gases over glowing platinum 

 wires. The present paper shows that un- 

 der ordinary conditions dissociation phe- 

 nomena take place so close to the surface 

 of the wire that convection currents do not 

 influence the dissociation. The tempera- 

 ture of the gas near the wire is then calcu- 

 lated from the heat conductivity and the 

 heat given off by the wire. Taking into 

 account the diffusion and the variation of 

 the dissociation constant and of the velocity 

 coefficient with the temperature, a formula 

 is derived which enables one to calculate 

 the difference between the dissociation for 

 equilibrium at the temperature of the wire 

 and the dissociation actually observed after 

 passing the gas over the wire. Applying 

 the formula to the results of the experi- 

 ments, it is shown that this difference can 



not exceed 10"^° per cent, or only 10"^ of 

 the actual dissociation. Therefore the only 

 remaining source of error in the experi- 

 ments was in the determination of the tem- 

 perature of the wire and in the analysis 

 of the gases. The results for the dissocia- 

 tion may be considered quite accurate. 



The Lime-Silica Series of Minerals: 

 Arthur L. Day and E. S. Shepherd. 

 There are two definite compounds of 

 lime and silica which can exist in contact 

 with the melt: (1) the pseudo-hexagonal 

 metasilicate, melting at 1,512° and invert- 

 ing to wollastonite at about 1,200° ; (2) the 

 orthosilicate of calcium, melting at 2,080° 

 and possessing tbree polymorphic forms, 

 which have been given the names a, ft and 

 y, in the order of their formation. The 

 a form is monoclinic, density 3.27, hardness 

 5 ; the /? form is orthorhombic, with about 

 the same density ; the y form has a density 

 of 2.97 and also crystallizes in the mono- 

 clinic system. The inversion point a to ^ 

 occurs at 1,415° ; ;8 to y at 675°. There 

 are three eutectics in the series, tridymite 

 -j- metasilicate at 35 per cent. CaO, 1,417° ; 

 metasilicate + orthosilicate at 54 per cent. 

 CaO, 1,430°; orthosilicate + lime at 67^ 

 per cent. CaO, 2,015°. The constants of the 

 original components are these : pure fused 

 lime has a density of 3.32, hardness 3 -f- 

 The fusion temperature is unknown. Lime 

 crystallizes in the isometric system and no 

 polymorphic forms were found. Silica 

 melts very gradually, beginning at about 

 1,600°, to an ultraviscous liquid. The 

 melting point, like those of albite and 

 orthoclase, is, therefore, indeterminate. At 

 all temperatures above 800° quartz changes 

 to tridymite and quartz glass crystallizes 

 as tridymite, so that above this tempera- 

 ture tridymite is unquestionably the stable 

 phase. The density of pure artificial tri- 

 dymite is 2.320 (25°) ; that of quartz glass, 

 2.213 (25°); the purest natural quartz 



