224 Messrs Heyoock and Neville, Superficial Colour, etc. [Nov. 23, 



That the air has nothing to do with this gradual change of 

 colour can be proved by heating some of the turnings to 150° C. 

 and then exposing them to the air. A month's exposure gives 

 no trace of red. In fact, provided the alloy has been annealed at 

 150°, the white colour is quite stable in air, in moisture or in 

 oxygen. 



The red colour also seems to be quite stable at ordinary tem- 

 peratures. The red resists dilute sulphuric and hydrochloric acids 

 and the alkalies, and what is more remarkable, it is not affected 

 by sulphuretted hydrogen or by ammonium sulphide reagents 

 which blacken silver. The action of ozone on the red is very 

 slow but it does finally blacken the alloy. 



These are the main facts we have observed. We do not knoAV 

 the chemical change associated with the red colour, and on account 

 of its being a surface action involving minute quantities of matter 

 it will probably be very difficult to settle this. Oxygen may play 

 a part in the reaction, for there is almost certaiuly some zinc oxide 

 in the alloy, and possibly some oxygen occluded by the silver. 

 The upper limit of temperature 300° is that at which Ag 2 is 

 completely decomposed under atmospheric pressure. 



We hope however to be able to replace speculation by facts in 

 a further communication. 



(Note added Dec. 31, 1896.) 



Since reading the above paper before the Society, we have 

 prepared some of the alloy from metals that had previously been 

 scrupulously freed from oxygen or oxides, the silver by heating to 

 a red heat for some hours in an atmosphere of hydrogen, and the 

 zinc by distillation in a hydrogen vacuum. The alloy made in 

 hydrogen from metals thus prepared turns a fine red when heated 

 and suddenly cooled in the unopened sealed tube in which it was 

 made. The colour change cannot therefore be due to oxygen or 

 to the formation or decomposition of oxides. 



We have also found that a silver-cadmium alloy AgCd under- 

 goes somewhat similar colour changes. 



(2) Note on Thermometric " Fixed Points." By E. H. 

 Griffiths, M.A., Sidney Sussex College. 



It is commonly assumed that our present definitions of the 

 thermometric fixed points are theoretically sound and in practice 

 satisfactory. This may possibly be the case, but, at the same time, 

 there are certain ambiguities having their origin in experimental 

 difficulties which may render profitable a consideration of the whole 

 matter ; especially as recent advances in the science of exact 

 thermometry make it impossible to tolerate any doubt as to the 

 fundamental definitions on which that science depends. 



