Chemistry and Physics. 51 



the applicability of the process. He finds that if a small amount 

 of an iodide is present in an ammoniacal solution of a cyanide, 

 silver iodide appears at the end of the reaction, although the 

 method, in its original form, cannot be used in the presence of 

 ammonia. He shows that the amount of silver solution used is 

 rigorously proportional to the cyanogen present, even when con- 

 siderable amounts of caustic alkalies, varying amounts of ammo- 

 nia, etc., are present. He recommends the following method for 

 alkaline cyanides : Weigh out one gram, dissolve in water, 

 making a volume of one liter, take 100 c.c. of the filtered liquid, 

 add 10 c.c. of ammonia solution, and about ten drops of a 20 per 

 cent, solution of potassium iodide, and run in decinormal silver 

 nitrate solution until a permanent precipitate appears. When 

 sulphides are present, 10 c.c. of a 10 per cent, solution of zinc 

 sulphate and a few drops of a basic lead acetate solution are 

 added to the liquid before it is made up to a liter. The author 

 proposes the use of a standard solution of potassium cyanide for 

 the determination of silver by titration, the end of the reaction 

 being determined by the disappearance, in an ammoniacal solu- 

 tion, of a slight precipitate of silver iodide. He considers this 

 method preferable to Volhardt's thiocyanate method, but the 

 greater instability of potassium cyanide solutions is evidently not 

 in its favor. — Ann. Chim. JPhys., VII, vi, 381. h, l. w. 



4. The Emission of Light during Crystallization. — Instances 

 of this phenomenon have long been known, and H. Rose, who 

 studied the subject, concluded that it was due to a change from 

 an amorphous to a crystalline condition. He believed, for exam- 

 ple, that vitreous arsenious acid when dissolved in hydrochloric 

 acid gave light when it crystallized, while the porcelain-like form 

 of the substance, dissolved in the same way, failed to give the 

 phenomenon. This view, that the amorphous or crystalline 

 nature of a substance could in any way affect the nature of a 

 solution of that substance, is not in harmony with the modern 

 ideas concerning solutions. Bandrowski has now shown, as 

 was to be expected, that both kinds of arsenious acid produce the 

 same results under like conditions. He finds the result, in the 

 case of arsenious acid, to depend upon the presence of hydro- 

 chloric acid, as no emission of light occurs when the crystalliza- 

 tion takes place from water or from alkaline solutions. The 

 author made a series of experiments by dissolving 15 g. of arse- 

 nious oxide in 150 c.c. of hydrochloric acid, gradually varying 

 the strength of the acid.. He found that with very weak and 

 very strong acid no light was emitted upon cooling to crystal- 

 lization, and that the maximum effect was produced when the 

 acid contained from 16*5 to 18 per cent, of hydrogen chloride. 

 The phenomenon is described as very brilliant when observed in 

 darkness, sometimes furnishing sufficient light for telling the time 

 by a watch. The light appears in the form of innumerable 

 bright sparks which are accompanied by a peculiar sound, and 

 the whole phenomenon is suggestive of an electric discharge. 



