THE ANALYST. 267 ON THE COMPOSITION OF HONEY. Read 6efore the Xociety of Padlic Anakysts OP 1 s t May, 1878. TBEBE are so few analyses of honey on record that it seems desirable to bring before this Society a selection of analyses of authentic examples of genuine honey with which commercial samples may be compared. The only detailed analyses of honey with which 1 am acquainted are those published in Hassall’s But as they do not give the proportions of the different kinds of glucose, but only the total glucose, and are accompanied by the extraordinary statement (p.274) that starch sugar is the adulteration, which it is scarcely possible in many cases to detect, they will not be of much practical value to the members of this Society. The analytical determinations which are useful in the analysis of honey are the following :- BY J.CAMPBELL BROWN, D.SC. Food and its Adulterations,” 1876, p. 266. 1. 2. 3. 4. 5. 6. 7. 8. 9. Estimation of the water of solution expelled at a temperature slightly over looo c. Water of combination and other volatile matter expelled only at a higher temperature ; this may be sometimes safely estimated by difference.Insoluble matter ; pollen, wax, &c. Xicroscopical examination of the honey, and especially of the pollen. Estimation of the ash, if necessary. Qualitative examination of the ash when the quantity is great. Estimation by the polariscope saccharimeter of the action of a solution of known The same after inversion. Estimation of the total glucose by standard solution of copper salt.strength on the polarised ray. 10. A similar estimation after inversion is often useful as a check. The result of 7, 8, and 9 give, by an obvious calculation, the proportions of dextro- and laevo-glucose. The proportion of cane sugar is calculated from 9 and 10, or may be deduced by means of Clerget’s tables from 7 and 8. I have never been able t o satisfy myself of the occurrence of cane sugar in honey ; and I am by no means sure that the figures representing cane sugar in the following analyses, and in those of Dr.Hassall, do not really represent experimental error; my figures for cane sugar are calculated from observed results in which a very small observational error would give a difference of 1 or 2 per cent. RESULTS OF ANALYSES OF AUTHENTIC SPECIMENS OF GENUINE HONEY.English. Welsh. mz:&a German. Greek. Lisbon. Jamaica. fz:l. Mexicall. Water expelled at looo ...... 19.1 16.4 15.5 19.11 19.8 18.8 19-46 17.9 18.47 Water expelled at a much higher temperatureand loss] 7-6 6.56 4.95 11’ 7-8 6.66 7.58 8.13 10.03 Laevulose ........................ 36.6 37.2 36.88 33.14 40. 37.26 33.19 37.85 35-96 Dextrose .......................36.65 39.7 42.5 36.68 32.2 34.94 35-21 36.01 35.47 Cane sugar (?) .................. } d~::t-none. none. none. none. 2.2 none. ‘;$t- *05 ne$,, 2.1 f;:, trace. Wax, pollen, and insoluble) good trace, slight trace, Mineral matter .................. .15 ‘14 *17 *17 -15 -14 *26 *11 907 1.2 matter ........................ j trace. trace.268 THE ANALYST. The specific gravity of honey is about 1-41, but varies slightly with the proportion of water.The proportions of water are higher than might have been expected, but I have confirmed some of the above results by a combustion with oxide of copper. The rotation of a polarised ray, produced by a solution of 16-26 grammes crude honey in 100 C.C. water, is generally from - 3 O - 2 to -FjQ at 60° Fahr. The only one of the above samples which gare a higher rotation was the Greek honey, which gave nearly -54O. The rotation produced by a solution of the same weight of dried honey is generally not far from --4O-8 ; but some latitude must be allowed until a larger number of observations have been made.