COTTON-SEED nil.. 



affected to any great extent hy tills heating, as is shown in Table 

 XXIII, taken I'nun their work. 



TABLK XXIII.- ]'(t/it,x <>/ //, <t/< >/ /<,//, ,/,-*, , <l ,,</ < 'l'<>rt< III <m</ A'/"/' 



The iodin number of the litjiiid fatty aei<U is afl'eeted ly heating to 

 a greater degree than any other value, being considerably lowered. 

 Tortelli and Ruggeri propose the following method for the detection 

 of the heated oil which depends on the fart that while, such an oil will 

 not give the Halphen or Bechi test, the reacting substance is not 

 entirely destroyed, and by concentrating this a reduction of silver is 

 obtainable. 



The liquid fatty acids are separated as described under Milter's 

 method/' The liquid fatty acids are, treated in a test tube with a mix- 

 ture of 1 cc of a 5 per cent aqueous solution of silver nitrate and 10 cc 

 of ( .M) per cent alcohol, and heated to from 70 to 80 for several hours 

 if necessary, in order to get a reduction of silver. By this method 

 small quantities of heated cotton-seed oil in olive oil nmy be detected. 

 Other reactions which will show the presence of heated cotton-seed oil 

 are the nitric-acid test, using acid of specific gravity l.:-57.~>. and the 

 Brulle '' test, w r ith nitric acid and jegg albumen. The former was tried 

 on some heated cotton-seed oil which would not react with the Hal- 

 phen reagent, and a strong reaction was obtained with a mixture of 

 In per cent of heated cotton-seed oil and 90 per cent of olive oil. 

 Smaller amounts could easily be detected. The test is not as sensitive 

 as the Halphen or Bechi test on untreated oils, but it is of great value 

 foruse with heated oils. It may also be of value in determining whether 

 a weak test with the Halphen reagent is due to a small amount of 

 indicated cotton-seed oil or a larger amount which has been heated 

 enough to weaken the Halphen reaction. Some of the other oils 

 react, however, with nitric acid. Sunflower oil, treated by thi- 

 method, gives a deep brown color which can not be distinguished from 

 the color obtained with cotton-seed oil. Peanut oil gives only a very 

 slight test. Maize oil reacts very strongly, giving a peculiar reddish 

 color that might be mistaken at lir-t for the cotton-seed oil 



U. S. Dept. A;:r., Ilun-un of Clu-inistry P.nl. M. }>. 28. 

 &R. Brullr, romp, ppnd.,1893, IfMJ: 1017. 



