162 THE ANALYST. A SUGGEBTION RESPECTING THE EXPR,ESSION O F THE RESULTS OF BUTTER ASSAYS BY DR. KOETTSTORFER’S METHOD. BY A. H. ALLEN, F.C.S. Read before the Society of Public Analysts, at Shefield, o n 22nd A 4 q i ~ s t , 1879. I HAVE been much interested in perusing the description of Dr. Koettstorfer’g method of detecting foreign fats in butter, and have no doubt that experience will prove it to be extremely useful.It appears to me, however, that his mode of stating the neutrdizing power of the fats is very cumbrons, and has the disadvantage that if sodaTHE ANALYST. 163 be used for saponification instead of potash a different amount of it will be necesmry, and hence the relation between the results obtained will not be directly evident. As the simplicity and accuracy of Dr.Koettstorfer’s process are likely to bring it into very general use, I propose, that instead of stating the results in milligrammes of K H O used per gramme of fat, they slioulcl be expressed by figures representing the combining equivalent of the fats present. These will in each case be one-third of the molecular weights of the fat, owing to the triad character of the glycyl radical.Thus, 56.1 parts of K H 0 will theoretically saponify 296.7 parts of tri-stearin ; 292-0 parts of tri-olein ; 296.0 parts of tri-yalmitin ; or, 100-67 parts of tri-butyrin, Dr. Koettstorfer’s results are capable of translation into equivalents, by dividing the milligrammes of K H 0 employed into the number 56-1. By 80 doing, we find that the average equivalent weights of the animal fats tested by Dr.Koettstorfer were as follows :- Beef dripping . . . . . . . . . . . . . . . . . . . . 285.5 Commercial tallow . . . . . . . . . . . . . . . . . . 285.1 Lard from kidneys . . . . . . . . . . . . . . . . . . 286.5 Lard from unsmoked bacon . . . . . . . . . . . . . . . . 2867 Commercial lard . . . . . . . . . . . . . . . . . . 287.1 Mutton dripping .. . . . . . . . . . . . . . . . . 284.8 Average = 286 - - Calculating in the same manner, his results shorn 292-5 as the equivalent of olive oil, and 313.9 as that of colza oil.‘:: On the other hand, the equivalents of the thirteen samples of butter varied from 253.5 to 241.4, the average being 247*1. Hence, average butter fat has an equivalent lower by 80 than the averwe of animal fats.Dividing this difference by 100, we find that every rise of a 8 8 in the equivalent of the sample represents 1 per cent. of probable adulteration; but, of course, though these figures may be properly used for calculating the probable proportion of admixture, a butter should not be condemned unless its equivalent be found to exceed 255. By expressing the results of analysis in equivalents, it is evident that it becomes a matter of indifference whether soda or potash be used for saponification, It will be noticed that the expression in equivalent weight is at owe obtained by dividing 1*000 by the number of c a b of norrnd alkali required to saponify one grrnslme of the fat under examination.