I n 1938, Paschke (23) introduced a method f o r detection of horsemeat in admixture with beef, mutton, o r pork, based on the observation that horse fat is richer in linolenic acid than are the depot fats of other large domestic animals. According to this procedure, the hexabromide number of the mixed fatty acids is used as a measure of the percentage of linolenic acid in the fat extracted from the meat sample under investigation. Variations of the Paschke procedure have been used by Fisher (14), Crowell (9), Dalley ( l o ) , and by Hynds (19,20). These workers report hexabromide values ranging from 4.1 to 21.5 for horsefat, and from 0.1 to 1.0 for beef, mutton, and pork fats. The chief difficulty mentioned by these authors is the poor reproducibility of the hexabromide method. Hynds (20) found that hexabromide values from 1.4 t o 4.0 were obtained when identical mixtures of horsemeat and beef were analyzed by several independent workers, using the same procedure.This paper describes a modified Paschke procedure, whereby ultraviolet spectrophotometry is used t o determine the percentage of linolenic acid in the fat extracted from meats and sausages. I n recent years, this method has generally supplanted other procedures for analysis of fats for polyunsaturated acids, particularly in instances where a high degree of accuracy is required. EXPERIMENTAL METHODSThe experimental work forming the basis of this paper included a study of extraction procedures in order to arrive at a standard analytical procedure which gives precise results for the percentage of linolenic acid in the f a t obtained from undried meat, and the application of this procedure to meats and sausages.Preliminary studies. Moderately lean chuck meat from U. S. inspected beef (sample 4), and lean horse flank meat (sample 10) were used t o determine the effects of cooking and of extraction conditions on the percentage of linolenic acid found in the extracted fat. About 500 g. of each material was passed 3 times through a handoperated food chopper, which is equipped with a cutter blade, and which has an extrusion plate with %-inch perforations. The meat was quartered and kneaded after each passage through the chopper.Approximately 25 g. of the chopped beef or horsemeat was used for each extraction study. Single-thickness thimbles were used f o r the Soxhlet extractions; the samples to be cooked were weighed in the thimbles, and the Sled thimbles were placed for one hour in an air oven maintained at 90°C. Thus, the percentages of extracted f a t given in Table 1 are based on the weights of the original uncooked and undried samples. The extractions were carried out in all-glass equipment, using diethyl ether or petroleum ether (b. 30-60°C.) for either 3 or 5 hours. The heating rate was adjusted so that siphoning occurred 1 2 t o 15 times per hour.Extractions were also carried out by maceration of the chopped meat samples with ice and solvent in a Waring blendor. The blendor jar was charged with about 50 ml. 'Sponsored in part by the Cudahy Packing Compa...