CHEMICAL SCIENCE. 213 



exposed to the light with a sheet of white paper behind them. With 

 these precautions, I have found this test astonishingly delicate, in fact 

 ranking with those for iron, iodine, &c. Using fused nitre, I have 

 detected the presence of 1 pt. in 617,000 pts. of water, a by-stander, 

 wholly ignorant of the nature of the operation, pronouncing as to the 

 color. Yet this salt contained about one half its weight of indecom- 

 posed nitrate. It should be remarked, that the presence of a large 

 quantity of nitre has no influence upon this test, as with pure water 

 it gives no color. The same reaction also answers for the hyponitrates. 

 The next step is to convert this test into one for the nitrates. I find 

 that nitrate of ammonia is readily decomposed in presence of metallic 

 lead ; and, what seems surprising, nitrate of potassa is also decomposed, 

 though not so readily. To test for the nitrates, we have only to agitate 

 the slightly- warmed liquid for a few minutes with shavings of lead, and 

 proceed as before. By a longer digestion, more of the salt would be 

 converted into nitrate, and the color would be stronger. By using 

 distilled water, I have been able to detect the presence of one part of 

 nitre in about 60,000 of water, digesting with lead only a short time. 

 Oxalic, tartaric and dilute hydrochloric acids may be substituted for 

 the acetic, except when they produce precipitates, which would de- 

 stroy the clearness of the liquid. 



TEST FOR IODINE. 



LASSAIGXE recommends chloride of palladium as a test for iodine. 

 He states that two litres of water, containing two millionths of iodide 

 of potassium, is still distinctly colored brown by chloride of palladium. 

 After twenty-four to thirty-six hours, a floculent sediment of iodide of 

 palladium was formed. If there be but a few flakes, a little fine silica 

 is added after the water is poured off; the sediment is collected, dried, 

 and then heated in a glass tube, which becomes filled with the vapors 

 of iodine. Jour, de Phar. 



NEW METHOD OF TESTING OPIUM. 



M. DESJIEDT proposes the following new process of testing opium. 60 

 grammes of raw opium are treated by 240 grammes of boiling alcohol 

 at 71 C. It is decanted while hot, and the grounds put in the press 

 after cooling ; the grounds are then treated in the same manner with 

 160 grammes of alcohol of the same degree ; these several alcohates 

 are collected and hermetically sealed in phials, and left to deposit until 

 the next day. A magnificent crystallization of narcotine is then 

 found, which is separated from the alcoholic liquid. This liquid is put 

 into a wide-mouthed phial, and four grammes of ammonia are added ; 

 a considerable quantity of pure morphine will be thus obtained, which 

 is to be separated from the liquor. This done, a small quantity of dis- 

 tilled water is added to the alcoholic liquor, keeping the phial at a 

 temperature of 75 Fah. for two days. After this time a fresh quantity 

 of morphine will be observed, less pure than the first, but free from 

 narcotine. By the process of the exhaustion of opium by alcohol, 

 the authors have obtained five grammes of morphine from 60 grammes 

 of opium employed. Journal de Chemie, 1851. 



