556 The Philippine Journal of Science 1923 



mined as were the rotations of the oils themselves, but they serve 

 to show that the variation of optical rotation in the different oils 

 is really due to differences in the composition of the fatty acid 

 fraction and not to differences in the manner of its combination 

 with glycerol or to the presence of some other optically active 

 constituent. This is especially interesting in the case of Pan- 

 gium edule, which will be discussed in the next section. 



FRACTIONATION OF THE ETHYL ESTERS 



In the preceding section certain tests have been shown to be 

 valuable in distinguishing the chaulmoogra group of oils, in 

 differentiating to some extent between the members of this group, 

 and, in connection with clinical data, in judging roughly the 

 therapeutic value of a given oil. These tests are not strictly 

 analytical, however, and taken alone do not determine, quanti- 

 tatively or qualitatively, any one constituent. The next step, 

 reported in this section, was to effect a partial separation of 

 the mixed fatty acids. A complete separation is such a difficult 

 matter that it has not been attempted in this study, but the 

 work on Taraktogenos kurzii and Hydnocarpus wightiana is 

 being extended with the expectation of reporting the approximate 

 quantitative composition of these in a later paper. 



The most successful general method of separating fatty acids 

 uses the difference in boiling point of the esters to separate 

 acids of different numbers of carbon atoms, and the difference 

 in solubility of certain salts or addition products to separate 

 acids with differing degrees of unsaturation. In the case of 

 commercial chaulmoogra oils Dean and Wrenshall 9 have modified 

 this method in that they prefer to distill the acids instead of 

 the esters, and crystallize the acids instead of their salts. Their 

 immediate object was to obtain pure chaulmoogric and hydno- 

 carpic acids rather than to find the percentage composition of 

 the oils used. 



For the present study direct crystallization of the acids was 

 adopted, as being more rapid and nearly as effective as crystal- 

 lization of the salts. The ethyl esters were distilled, rather 

 than the acids, being as easy to prepare, much easier to distill, 

 and not subject to as much decomposition during distillation. 

 In order to cover the ten samples in a reasonable time only one 



2fi J, Dean > A - L " and Wrenshall, R., Jpum. Am. Chem. Soc. 42 (1920) 



