438 On Tests for ArSenle. 



Let us therefore compare the two tests, (for the ammonia 

 is assuredly not to be called a test for arsenic) and see if there 

 be anvotherc5.9e7zf/a/diSerence between tbeni than thatwhich 

 was first noticed by myself, iu the Philosophical Magazine, 

 nearly four vears ago, in these words : — " The test I propose 

 as a substitute (for sulphate of copper) appears to be more 

 efficacious, inasmuch as it produces a more copious precipi- 

 tate from a given quantity of arsenic." 



1. Both these tests have a melallichAse ; they are applied 

 to the same purpose ; and they admit of the same variations, 

 especially with respect to the aLkaline agent, in their appli- 

 cation for the detcciion of arsenic. 



2. The addition of an alkali, of some kind or other, is as 

 necessary with one test as it is to succeed with the other; 

 and no less caution is required iu either of the exijeriments 

 to avoid an excess of the alkali. 



3. Any of the alkalin;" earths will serve with sulphate of 

 copper, as I have proved they will with nitrate of silver. 

 Though I have not made some experiments which 1 have 

 in conteinplation, I suspect the sulphate of co])per will 

 produce a more dense and conspicuous precipitate, when 

 ike arsenic is combined with harytes ; if so, this notice will 

 prevent any one arrogating to themselves the discovery of a 

 new test of this species. 



4. Both the copper and silver tests act exactly upon the 

 same chemical principles, or, in the language of Bergman, 

 by double elective attraction. 



5. It is here altogether in my favor that these metallic 

 salts have the same property in respect to ammonia^ that of 

 forming triple or ternary compounds. Of this fact both 

 Dr. Marcet and Dr. lloget seen) to have been totally igno- 

 rant, before I prescribed the ammoniaco-nitrate of silver^ 

 which you did me the honor to place among your pages in a 

 late number. Dr. Roget need not, therefore, be surprised 

 that I should become the ''panegyrist" of such a finished 

 composition. As we may with the same facility prepare a 

 triple salt with copper, to serve as a test for arsenic, I trust, 

 it would be unnecessary to carry this parallel any further, or 

 advance additional arcunicnts to shew that these tests are, 

 mutatis mutandis^ identical in their application and mode of 

 operating, in respect to arsenic. 



G. Ijoth these tests agree in producing more aluiidant 

 precipitates; the silver is, however, as I have already 

 noticed, by far the most delicate and conspicuous. 



Lastly? there is a striking analogy in these tests, in respect 

 to the colour each produces, so that the operator may 



instantly 



