I90 SCIENCE PROGRESS 



similar reaction limits to those of the alloys. Sodium chloride 

 and silver chloride are completely miscible ; crystals with i 

 to 0-75 mol of AgCl scarcely yield any sodium chloride to 

 water, while from 0*75 to 0-625 the series yield part of the 

 sodium chloride, and from 0-625 to o the whole of the salt is 

 removed. 



Galvanic resistance limits are also observed ; thus mixed 

 crystals between ^2 and o per cent, give the same polarisation 

 potential and possess the same precipitating ability as the 

 more noble metal. The thermodynamical theory of the depen- 

 dence of polarisation potential on the composition of an alloy 

 evidently stands in opposition to the existence of galvanic 

 resistance limits. This theory can only be employed when the 

 interchange of atoms in the crystals proceeds sufficiently 

 rapidly to enable the metallic phase to be in equilibrium with 

 the liquid. These conditions are not fulfilled at ordinary 

 temperatures in the case of the Au-Cu and the Au-Ag alloys, 

 so that galvanic resistance limits corresponding to the chemi- 

 cal limits could have been predicted. The polarisation poten- 

 tial of .the Au-Cu series of alloys in the element 



Cu-Au Alloy 



■o'5 mol 

 CUSO4 or yCu 

 .CuNOs 



has a constant value of i -40 volt for mixed crystals containing 

 I -00 to 0-50 mol of gold, while below this concentration of gold 

 the voltage falls off rapidly, and at 0-25 mol the voltage has 

 fallen almost to zero. Measurements of the polarisation 

 potentials at higher temperatures show no discontinuity at 

 0*50 mol gold, but vary continuously for the whole series of 

 alloys. At these temperatures the atoms in the metallic 

 phase are in equilibrium with the liquid. 



Working the metal either by rolling or hammering imparts 

 increased chemical activity to the metal ; the uniform arrange- 

 ment of the atoms in the space lattice is destroyed, and the 

 crystals more readily admit the attack of the reagent. 



Tammann develops a theory of the distribution of the 

 atoms in the mixed crystals from the results of this work. 



Isotopes of Chlorine. — According to the work of Aston 

 {Nature, 1919, 104, 393; Phil. Mag., 1920, 39, 620), who apphes 

 the positive-ray spectrograph to the analysis of the elements, 

 the mass spectra obtained with chlorine appear to prove that 

 this element consists of two isotopes of atomic weight 35 and 

 37. This is confirmed by W. D. Harkins {Science, 1920, 51, 

 289-91), who has carried out preliminary experiments on the 

 separation of hydrochloric acid by diffusion. His analyses 



