348 



PROCEEDINGS OF SECTION B. 



Table II. — A Comparison of Maercker-Morgan and Lietz Methods for 

 Starch, a\d of Komig's and Acid-Alkali Methods for Crude Fikre. 



I 



6.— THE DETECTION AND ESTIMATION OF 

 FORMALDEHYDE IN MILK. 



Bij J. C. BRUNNICH, F.l.C, Agricultural Chemist. 



[Abstract.] 



The author draws attention to the constantly increasing number 

 of reactions for formaldehyde, and gives a resume of a number of recent 

 tests, viz. : — 



Morphine Test (F. Bonnet, jun., J. Amer. C.S., 1905, pp. 601-605) 

 is recommended both for ease of application and sensitiveness. The 

 test is conveniently conducted in covered Petri dishes. As claimed by 

 Bonnet, no coloration is obtained from pure fresh milk, even after 

 standing over night ; but sour milk, to which no formalin has been 

 added, gives a pink coloration after three hours. One part formalin in 

 in 1,000,000 parts of fresh milk was indicated by a pink coloration, 

 obtained in one to one and a half hours, at 26° to 28° C. In milk 

 kept for two weeks only one part in 100,000 could be detected. Through- 

 out, temperature conditions were found to influence the time requisite 

 for the obtaining of reactions. Uniform coloring of the morphine 

 reagent, rather than the color rings mentioned by Bonnet, were ob- 

 tained. 



Leach's Hydrochloric Acid Test and Gallic Acid Test (R. H. Williams 

 and H. C. Shermann, J. Amer. C.S., 1905, pp. 1497 to 1505). 



Schiff's Reagent.- — Contrary to the experience of Hehner ("Analyst," 

 1896, p. 94), who obtained positive reactions with pure water after half 

 an hour's standing, this test is highly satisfactory. Special care should, 



