I02 Journal of Agricultural Research voi. vi.no. 3 



EARLIER INVESTIGATIONS 



For the separation of stearic from other fatty acids, David (i) recom- 

 mended a special alcohol and dilute acetic-acid solution saturated with 

 stearic acid at 15° C, in which solution oleic acid was shown to be 

 soluble. 



The Hehner and Mitchell (3) method for isolating stearic from other 

 fatty acids was based on the hypothesis that a mixture of fatty acids 

 heated with a solvent saturated at a given temperature with the acid 

 under determination might be expected on cooling to that temperature 

 to crystallize the whole of the acid sought, provided the other constitu- 

 ents did not increase the solubility. The solvent employed was methy- 

 lated alcohol (94.4 per cent) saturated with stearic acid at 0.2° C, pre- 

 pared by chilling a solution of 3 gm. to i liter overnight in ice water and 

 siphoning off the saturated mother liquor through a small thistle tube 

 covered with fine calico, using suction. The tests were conducted in a 

 similar manner, taking from 0.5 to 5 gm. of insoluble acids (according to 

 content) to 100 c. c. of alcohol-stearic-acid solution. Shaking was found 

 to increase precipitation. Supersaturation and esterification were recog- 

 nized as possible sources of error. The method gave concordant results 

 with solid fats containing considerable stearic acid, but slight, if any, pre- 

 cipitate from the acids of butter fat and from mixtures of the acids of 

 Japan wax and pure stearic acid. 



Emerson (2) noted considerable variation in the content of different 

 saturated solutions and found that supersaturation seemed to occur when 

 less than 0.7 gm. to 100 c. c. was employed in preparing the solution. 

 The formation of ethyl ester appeared to be a source of error and to have 

 increased the apparent solubility of the stearic acid. 



Kreis and Hafner (5) showed that small amounts of stearic acid below 

 0.1 gm. to 100 c. c. of a saturated solution formed supersaturated solutions, 

 and that less than 0.05 gm. gave low and extremely variable results, even 

 upon the addition of crystals of stearic acid. 



Lewkowitsch (6, p. 556-559) claimed that the method yielded capri- 

 cious results with mixtures of stearic, palmitic, and oleic acids, and that 

 in many cases the results were entirely unreliable when other acids were 

 present. He stated that a considerable proportion of lauric acid would 

 prevent the complete precipitation of stearic acid, even when super- 

 saturated alcohol-stearic-acid solutions were used, and that acids of 

 higher melting point, when present, such as arachic, behenic, etc., would 

 appear in the separated acids. He reported a precipitate of 0.49 per 

 cent from butter fat, of which a portion might be arachic and myristic 

 acids. 



The results obtained by various investigators indicate that the solu- 

 bility of stearic acid increases with the strength of the alcohol, but 

 the figures reported are too variable to warrant further deductions 

 (Table I). 



