406 RECORD OF SCIENCE FOR 1886. 



of sixteen analyses gave ratios approximating three atoms of silver to 

 one of nitrogen, whicli gives tlie formula NAgs. 



The substance was also prepared by warming the ammonia solution 

 of the silver oxide on a water bath, and by precipitating it with alcohol, 

 and these samples gave the same results on analysis. Berthollet's ful- 

 minating silver explodes with a very slight concussion when dry, and 

 even when moist must be handled with precaution. The explosive 

 character of each sample analyzed was determined. It dissolves in 

 potassium cyanide solution almost immediately, probably giving the re- 

 action : 



NAg3 + 3KCy + SH.O = NH3 + 3KH0 -f 3AgCy 



(Liebig's Annalen, ccxxxiii, 93.) 



Compounds of the Nitrates of the Alkalies icith Nitrate of Silver, by A. 

 Ditte. — The author describes the preparation and characteristics of 

 the following double salts: AgN03,KN0;i; AgNOs, RbNOj; Nn4N03, 

 AgN03; and shows that with sodium and lithium analogous double 

 salts arediflQcult to obtain in definite compounds. No less than twelve 

 reasons are presented for dividing the alkaline group of metals into two 

 sections, one embracing K, Rb, Cs, NH4; and the other, Li and Na. 

 (Ann. Chim. Phys. [6], viir, 418.) 



Decomposition of Potassium Chlorate, bj" Frank L. Teed. — In a previ- 

 ous paper the author arrived at the conclusion that the decomposition 

 of potassium chlorate by heat was represented by the equation : 



IOKCIO3 = 6KCIO4 + 4KC1 + 30^ 



but later experiments lead him to believe that the following is more 

 nearly correct : 



22KC1()3 = I4KCIO4 + 8KC1 4- 502 



A majority of the author's results fall within the limits calculated 

 from these two equations. When the chlorate is heated with manganese 

 dioxide it decomposes ai)pareutly without formation of pcrchlorate. 



In the discussion which followed the reading of this paper at the 

 Chemical Society of Loudon, Dr. Percy Frankland said experiments 

 made in the South Kensington laboratory had lead to the equation : 



8KCIO3 = 5KCIO4 + 3KC1 + 2O2 



(Chem. News, liii, 56.) 



For a further discussion of this subject see article by E. J. Maumeu6 

 in Chem. News, liii, 145. 



The Solvay Process of Manufacturing " Soda.^^ — In our reports for 1883 

 and 1884 we chronicled tlie decline of the Leblauc process and the rise 

 of the so-called " ammonia process" of manufacturing soda 5 we now 



