580 



SCIENCE 



[Vol. LVI, No. 1455 



ciated with the absoi'pti%'e capacity of the organic 

 compounds for ultra-\'iolet light. A striking cor- 

 relation between these factors has been obtained 

 in the case of benzene, several esters, acids, 

 amides, ketones and alkaloids. The retarding ac- 

 tion of amines and alcohols requires an explana- 

 tion based on other causes than adsorption of 

 light; (3) It has been shown that the inhibitors 

 act more efficiently when in the peroxide solutions 

 than when in a screening solution of similar 

 thickness and concentration. 



Tlie reduction of copper oxide hy carton mon- 

 oxide and the interaction of carbon monoxide 

 and oxygen in the presence of copper and of 

 copper oxide: H. A. Jones and Hugh S. Taylor. 

 (1) The reduction of copper oxide by carbon 

 monoxide has been shown to be an autoeatalytio 

 process, copper being the autocatalyst ; the reduc- 

 tion occurs at a copper-copperoside interface; (2) 

 The retarding action of carbon dioxide and of 

 oxygen on the primary reaction of the reduction 

 process, i. e., the formation of the copper nuclei, 

 has been pointed out; (3) The mechanism of the 

 carbon monoxide- oxygen catalysis over copper 

 oxide has been shown to be alternarte reduction 

 and oxidation of the copper oxide; (4) The 

 mechanism of the catalysis in the presence of 

 copper has been shown to be oxidation of an ad- 

 sorbed layer of carbon monoxide; (5) It has 

 been established that oxj'gen is a poison in the 

 combination of carbon monoxide and oxygon over 

 copper. 



The calculation ,of critical values for binary 

 mixtures: A. G. LooMis.' One of the greatest ex- 

 perimental difficulties in the study of the pressure- 

 temperature composition surface for binary mix- 

 tures in the accurate determination of the critical 

 values for the various mixtures. It is shown in 

 this paper that by emploj'ing the equation of 

 Dieteriei, which is quite accurate in the critical 

 region, the entire critical line may be very closely 

 calculated; this equation leads to better results 

 than the equation of van der Waals. The com- 

 position of the mixture with minimum critical 

 temperature can be very accurately calculated by 



ax 

 finding the expression which makes — a minimum, 



bx 

 where ax and bx are the attraction and volume 



constants, each e.xpressed as a quadratic function 



of the composition. By emploj-ing the conditions 



for phase equilibrium on the critical line and 



introducing the equation of state in the expression 



for the free energy of the system, the composition 



of tlie mixture with maximum vapor pressure is 



accurately calculated when the values Fj and Fo 



of each phase are equated and put equal to 26, as 

 demanded by the equation of Dieteriei. 



Compound formation and ionization in fused 

 salt mixtures: James Kendall, E. D. Critten. 

 DEN and H. K. Miller. A study of the freezing- 

 point curves for a large number of systems of 

 the types aluminum chloride — metal chloride, 

 aluminum bromide — metal bromide, antimony tri- 

 chloride — metal chloride has shown that here, as 

 in other fields, the extent of compound forma- 

 tion between the two components is primarilj' 

 dependent upon the diversity of their constituent 

 radicals. The effect of subsidiary factors, such 

 as valence, unsaturation, internal pressure and 

 atomic volume, has also been investigated. 

 Ionization is found to run parallel with com- 

 pound forma"tion in the few systems for which 

 data are now available. The work is being con- 

 tinued. 



A low temperature electrolyte: W. H. Eode- 

 BUSH and Theodore 0. Yntema. Hydrogen 

 ehlorid and nitric oxid form a compound at low- 

 temperatures of an intense purple color. This is 

 of interest because G. N. Lewis has pointed out 

 that nitric oxid is the only molecule, containing 

 an odd number of electrons that is not colored. 

 It was predicted that this compound would show 

 conductivity in the liquid state. This prediction 

 was verified, a conductivity of 10-3 being easily 

 obtained at 130° K. This is believed to be the 

 lowest temperature at which electrolytic conduc- 

 tion will be obtained. Speculations are offered as 

 to the nature of the compound. 



Transference numbers of sodium and potassium 

 in mi^ed chloride sohition: R. F. Schneider and 

 S. A. Braley. The transference numbers of 

 sodium and potassium have been determined in 

 solutions with a total concentration varying from 

 0.1 to 1.6 N and with varying salts ratios of 

 from 3 KCl and 1 NaCl to 1 Kcl and 3 NaCl. 

 While the conductance of such solutions conform 

 to calculated values on the basis of the isohydric 

 principle the ratios of the transference are widely 

 different from the calculated. At a total concen- 

 tration of 0.2 N and a salt ratio of 3 Kcl to 

 1 NaCl the transference number of the sodium 

 becomes practically zero, indicating that' the 

 sodium is not only transported as the cation but 

 is also tied np in a complex anion to such an 

 extent that it moves in both directions at the 

 same rate. The da,ta show that the theory of 

 complete dissociation as advocated by Ghosh can 

 not possibly hold. 



Charles L. Parsons, 



Secretary 



