The determination of conjugated diene formation revealed that the tool % conversions of all cis-6,9,12-oetadecatrienoic acid [y-linolenic, 18:3(n-6)], all cis-5,8,11,14eicosatetraenoic acid [arachidonic, 20:4(n-6)], all cis-5,8, ll,14,17-eicosapentaenoic acid [20:5(n-3)], and all cis. 4,7,10,13,16,19-docosahexaenoic acid [22:6(n-3)] into conjugated diene products by soybean lipoxygenase-1 at pH 9.0 were 84, 86, 60 and 40% of that of all cis-9,12-octadecadienoic acid [linoleic, 18:2(n-6)], respectively. On the other hand, the conversions of all cis.9,12,15-octadecatrienoic acid [a-linolenic, 18:3(n-3)], all cis-5,9,12-octadecatrienoic acid (c5,c9,c12-18:3), and trans.5,cis-9,cis-12-octadecatrienoic acid {t5,c9,c12-18:3) were equal to that of 18:2(n-6). The lowering of the conjugated diene formation in the oxidation of 18:3(n-6), 20:4(n-6), 20:5(n-3), and 22:6(n-3) by the lipoxygenase was thought to be caused by the further oxidation of conjugated diene monohydroperoxides to yield conjugated triene products. For this reason, the conventional lipoxygenase method gave erroneous values for cis, cis-methylene interrupted polyunsaturated fatty acids (PUFA) in oils containing a large amount of 20:5(n-3) and 22:6(n-3) such as fish oils. However, by changing the pH of reaction mixtures from 9.0 to 11.0, the secondary oxidation of conjugated diene monohydroperoxides was completely inhibited, and the PUFA values in fish oils obtained by this improved method were in good agreement with those obtained by a GLC method.