86 Dr. Thomson on Arsenic. [Feb. 



change. The portions of the salt next the vessel, however, 

 become whiter a little, and probably lose part of their water of 

 crystallization, though the salt loses no sensible weight. In a 

 red heat, it melts, and becomes as liquid as water. In that state 

 it is nearly colourless, having only a slight tinge of yellow, and 

 sometimes of green, both of which shades disappear when the 

 salt becomes cool. On concreting, it cracks in all directions, 

 indicating that it occupies a smaller bulk when solid than when 

 liquid. The congealed salt is opaque, or only translucent, and 

 white: 100 gr. of the salt by this treatment lose 7'5 gr. of their 

 weight. The salt is completely soluble in water ; therefore, the 

 7*5 gr. may be reckoned the water of crystallization. 



Arseniate of potash is insoluble in alcohol : 100 parts of 

 water at the temperature of 42° dissolve 19*047 gr. of the salt. 

 The specific gravity of this solution (at 60") is 1-1134. It is 

 much more soluble in hot than in cold water. Hence a saturated 

 solution in hot water crystallizes veiy readily on cooling. 



The constituents of this salt, according to my analysis of it, 

 are as follows : 



Arsenic acid 65*426 



Potash 27*074 



Water 7*500 



100*000 



Now the weight of an integrant particle of potash is 6, and 

 27*074 : 65*426 :: 6 : 14*5 very nearly. As far, therefore, as this 

 salt is concerned, the weight of an atom of arsenic acid may be 

 14*5. 



If you mix together a solution of arsenic acid and soda in the 

 proportion of 7*25 arsenic acid + 4 soda, the liquid acts hke an 

 alkali upon vegetable blues, and cannot be made to yield crys- 

 tals. We have, therefore, no accurate means of determining 

 whether the acid and alkali form a salt in these proportions or 

 not. The same observation applies to a mixture of 7*25 arsenic 

 acid with six parts of potash . 



III. Arseniate of Copper. 



No fewer than five varieties of this salt all crystallized, but 

 varying in the colour and shape of the crystals, have been 

 described by Count Bournon, and analyzed by Chenevix. Unfor- 

 tunately the analysis was made at a time (1801) when the 

 necessity of minute accuracy was not so much felt as at present. 

 Hence we are not to expect exact agreement between the atomic 

 theory and the analytical results of Mr. Chenevix. Four of these 

 five varieties occur crystallized in the copper mine of Huel 

 Garland, in Cornwall. The fifth variety was formed artificially 

 by Mr. Chenevix. He poured arseniate of ammonia into nitrate 

 of copper, and filtered the liquid to get rid of a green-coloured 



