The Estimation of the Total Lipids and the Lipid Partition in Feces*

For the estimation of total lipids in the feces a number of excellent methods are available, which depend upon primary saponification of the material. When the fat partition is to be determined, however, it becomes necessary to use extraction procedures, a number of which are described in the literature. A method widely used is that of Holt, Courtney, and Fales (l), which is essentially an adaptation of the Roese-Gottlieb method for the analysis of milk powders. Although subsequent workers (2-5) have altered the procedure in one way or another with a view to improving its accuracy, it remains troublesome and time-consuming. It is the purpose of this publication to present a simplification of this procedure which avoids some of its difficulties, and also to call attention to an error made by previous workers; almost without exception those who have worked with human material have failed to appreciate that the “neutral fat” fraction invariably contains a large proportion of unsaponifiable material; hence the deductions made in regard to fat splitting under various conditions have frequently been unsound. In order to obtain a true picture of fat splitting it is necessary to determine the unsaponifiable matter of the stool fat as well. The technical difficulties in the procedure of Holt, Courtney, and Fales and its modifications are (1) the tendency for emulsions to be formed during extraction, which often separate out with extreme slowness, and (2) the fact that the determination of “neutral fat,” fatty acid, and soap is not made on a single sample,