VI. ESTIMATION 287 



method of analysis for the essential fatty acids. Mitchell et at.- reduced the 

 phenomenon to a quantitative method using 7.3 % KOH in ethylene glycol 

 as reagent and a treatment of 30 minutes at 180°. This has been modified 

 and improved by Brice and Swain.^ They used 11% KOH in glycerol to 

 reduce the variable blank absorption, and they have introduced l)ackground 

 corrections in the calculations to improve the accuracy of the method. 

 Protection of the sample during isomerization by an atmosphere of nitrogen 

 has also been recommended. The method, even in its present highly 

 developed form, is still subject to some limitations. The 11 % KOH reagent 

 is optimum for linoleic and linolenic acids, but gives only a low degree of 

 the specific conjugated isomers of the more highly unsaturated acids.* 



The isomerization treatment of a hexaenoic acid results in production of 

 a variety of hexaene isomers in which six, five, four, three, or two of the 

 double bonds are conjugated. Thus the spectrum of such an isomerized 

 sample shows principal maxima at 3750 A., 3475 A., 3000 A., 2700 A., 

 and 2340 A. Treatment of a pentaenoio acid will give rise to similar maxima 

 showing pentaene, tetraene, triene, and diene conjugation. Arachidonic 

 acid will give rise to three types of conjugation, linolenic acid two, and lino- 

 leic acid only one. Thus it is apparent that, if linoleic acid is to be deter- 

 mined in the presence of the more highly unsaturated acids, appropriate 

 correction must be made for each of the others. 



The ratios of the various kinds of conjugated isomers derived from a 

 single polyunsaturated acid can be varied by alkali concentration and time 

 of treatment. For example, the conjugated tetraene yield from arachidonate 

 treated with 23 % KOH for 8 minutes is twice that from arachidonate 

 treated with 11 % KOH for 30 minutes.* Thus, use of 11 % KOH reagent 

 for "determination" of linoleic and linolenic acids in animal fats can lead 

 to error, because the tetraene, pentaene, and hexene acid content of the 

 sample is ordinarily overlooked and no correction is made for it. The polyim- 

 saturated acids of all types are thus unknowingly reported as linoleic and 

 linolenic acids. The confusion is compounded by the present lack of em- 

 pirical analytical standards for the five and six douljle-bond values. Thus 

 any analysis which pretends to give the contents of various polyunsaturated, 

 oleic, and saturated acids of animal fats by use of the 11 % KOH alkaline 

 isomerization method and iodine value should be regarded only as an 

 approximation. 



By use of 23 % KOH reagent the sensitivity of the method for the four, 

 five, and six double-bond acids is doubled. The author has used this reagent 

 in estimating the relative contents of these acids in animal fatty acuds. To 



2 J. H. Mitchell, Jr., II. R. Kraybill, and F. P. Zscheile, Ind. Eng. Chan. Anal. 

 Ed 15, 1 (1943). 



3 B. A. Brice and M. L. Swain, ,/. Opt. Soc. Amer. 35, 532 (1945). 

 ^ R. T. Holman and G. O. Burr, Arch. Biochem. 19, 474 (1948). 



