EDIBLE FATS AND OILS. 145 



the oil under examination, and heat in a bath of boiling saturated brine for 

 from 1 to 2 hours. In the presence of as little as 1 per cent of cotton-seed oil, 

 ;i characteristic red or orange-red color is produced. 



Lard and lard oil from animals fed on cotton-seed meal will give a faint 

 reaction ; the fatty acids also give this reaction. 



The depth of color is proportional, to a certain extent, to the amount of oil 

 present, and by making comparative tests with cotton-seed oil some idea as to 

 the amount present can be obtained, but it must be remembered that different 

 oils react with different intensities, and oils which have been heated from 200 

 to 210 C.o react with greatly diminished intensity. Heating 10 minutes at 

 250 C. renders cotton-seed oil incapable of giving a reaction.** 



(c) BECHI OB SILVEB NITRATE TEST FOB COTTON-SEED OIL. PROVISIONAL. 



(1) PBEPABATION OF REAGENT.^ 



Dissolve 2 grams of silver nitrate in 200 cc of 95 per cent alcohol and 40 cc 

 of ether, adding 1 drop of nitric acid. 



(2) DETERMINATION. 



Mix 10 cc of oil or melted fat, 5 cc of reagent, and 10 cc of amyl alcohol <* in 

 a test tube. Divide, heat one half in a boiling water bath for ten mintues and 

 compare with portion not heated. Any blackening due to reduced silver shows 

 presence of cotton-seed oil. 



Other oils which have become rancid and lards which have been steamed or 

 heated at high temperature contain decomposition products which have a 

 reducing action on silver nitrate. Some salad oils which contain no cotton-seed 

 oil, according to the Halphen test, give a brown coloration with the Bechi 

 reagent and in some cases reduce silver. Such oils when purified give no 

 reaction, and therefore the oils or fats should always be purified before testing. 

 To effect this, heat from 20 to 30 grams of the sample on a water bath for a few 

 minutes together with 25 cc of 95 per cent alcohol, shake thoroughly, decant as 

 much of the alcohol as possible, wash with 2 per cent nitric acid and finally 

 with water. Heating the oils or fats to 100 C. or simply washing with 2 per 

 cent nitric acid is not sufficient except in a few cases. 



(d) RENARD'S TEST f FOB PEANUT OIL AS MODIFIED BY TOLMAN. PROVISIONAL. 



Weigh 20 grams of oil into an Erlenmeyer flask. Saponify with alcoholic pot- 

 ash, neutralize exactly with dilute acetic acid, using phenolphthalein as indicator, 

 and wash into a 500 cc flask containing a boiling mixture of 100 cc of water and 

 120 cc of a 20 per cent lead acetate solution. Boil for a minute and then cool 

 the precipitated soap by immersing the flask in water, occasionally giving It a 

 whirling motion to cause the soap to stick to the sides of the flask. After the 



Allen, Commercial Organic Analysis, 3d ed., 2 (1) : 143. 



6 Holde and Pelgry, J. Soc. Chem. Ind., 1899, 18: 711. 



e Pearmain and Moor, Allen's Commercial Organic Analysis, 3d ed., 2(1): 143 ; Wesson, 

 J. Amer. Chem. Soc., 1895, 17: 724. 



d The addition of amyl alcohol is not necessary, but is very convenient, as it dissolves 

 the oils or fats, which then mix with the reagent much better. 



'Wesson, J. Amer. Chem. Soc., 1895, 17: 724; Winton, Conn. Agr. Exper. Stat. Kept., 

 1900, Part 2, p. 143. 



'Comp. rend., 1871, 78: 1330; Benedikt and Lewkowitsch, Oils, Fats, and Waxes, 

 p. 365. 



