Dr. E. J. Mills on Elective AiiracHon. 



513 



1 



I. 



II. 



Thallous nitrate taken 



Residue 



1-8993 

 20870 

 118°-5 



37-77 



8-89 



54-4(5 



21431 

 2-4041 

 116^-5 



42-01 



9-97 

 48-50 



Temperature 



Composition of the residue,viz. : — 

 Thallic dichloride 



Pliosphoric oxide 



Tliallous nitrate 





101-12 



100-48 



Value of a 



4-40 



4-36 





The excess of the above percentages ovei* tlie proper amount 

 is doubtless owing to the whole of the chlorine found having 

 been calculated as dichloride, whereas, on account of the slight 

 loss of chlorine already referred to, the residue must have con- 

 tained a little sesquichloride. In the second of these experi- 

 ments, where the temperature was purposely restrained to the 

 lowest advisable limit, less chlorine was probably missing, and 

 calculation agrees more nearly with fact. Mr. Crookes informs 

 me that, although he has never placed himself precisely in the 

 conditions under which I worked, the behaviour of the dichloride, 

 as here described, is such as he would have been induced to 

 expect. 



Sodic Nitrate, 



This body was prepared, in a state of satisfactory purity, by 

 the two following methods. In the first, a sample of common 

 sodic chloride was repeatedly digested with strong aqueous hydric 

 chloride assisted by heat ; and the insoluble portion, after having 

 been drained, was evaporated with an adequate excess of hydric 

 nitrate w^hich had been distilled below its boiling-point. In the 

 second, some sodic hydrate, prepared directly from the metal, 

 was evaporated with excess of the same hydric nitrate. Either 

 product v^^as dried, fused, dissolved in water, and evaporated to 

 crystallization. Nitrate obtained by the first process was used 

 in experiments I. and II. On evaporation with aqueous hydric 

 chloride it furnished 68*78 per cent, of sodic chloride. Theory 

 requires 68*76 per cent. In experiment III. a salt from the 

 second source was employed. Desiccation was effected with 

 facility at 165°-180° in a few hours. Contact with the oxychlo- 

 ride gave rise to immediate action and perceptible warmth. At 

 40° the liquid was turbid; at 50° -65° there was very consi- 

 derable action ; and at 70° the contents of the tube had a pecu- 

 liar fleshy colour and were semisolid. Liquefaction gradually 

 ensued; and complete decolorization occurred at 97°*5. The 



Phil, Mag, S. 4. No. 296. Suppl. Vol. 41. 3 L 



