THE ANALYST. 283 ABSTRACTS OF PAPERS PUBLISHED IN OTHER JOURNALS. FOODS AND DRUGS ANALYSIS. Determination of Cane-Sugar, Lactose, etc., in Milks, etc. F. W. Richard- son and Adolf Jaff6. (Journ. SOC. Chenz. I?zd., 1'304, xxiii., 309-311.)-The following polarimetric method is described : If the temperature of a solution of invert sugar be raised from 20" C. to 86" C., the polarimetric reading becomes nZZ, or the minus reading of the lzvulose becomes at 86" C. equal to the plus reading of the dextrose. For the analysis of a solution containing sucrose and invert sugar readings of the solution are taken at 20" C. and at 86" C., and a reading of the inverted solution at 20" C. Sucrose has [.ID = 63.05 a t 86" C., instead of 66.5, as at 20°C. ; therefore any pZus reading of the original solution at 86' C.divided by 0.6305 gives the amount of sucrose. Moreover, a 10 per cent. solution of invert sugar has a rotatory power of -20 at 20' C. One gramme of sucrose furnishes 1.053 gramme of invert sugar, with a reading of - 0.20 x 1.053 = - 0.2106. The 1 gramme of sucrose disappears on inversion, and284 THE ANALYST. with it a plus reading of 0.665; therefore every gramme of sucrose when inverted causes a reduction in the rotatory power of -0*2106+(-0-665)= -0,8756. Any diminution of the plus reading effected by Herzfeld's process divided by 0,8756 gives the sucrose. Knowing this and the rotatory value of the original solution, it is easy to calculate the invert sugar. The same process may be applied to solutions con- taining sucrose and lactose. I n the case of condensed milk, the acid mercury nitrate solution added to pre- cipitate the proteids completely inverts the sucrose when heated to 86" C.Readings of the filtrate, therefore, at 20" C. and 86' C. give the amount of lactose and sucrose. The authors are also investigating a method for mixtures of sucrose, lactose, and dextrose, depending on the very different changes which these three sugars undergo when heated in solutions containing 10 per cent. of hydrochloric acid. w. P. s. On the Solubility of Glycerides in Acetic Acid. I;. Hoton. (Bzdl. XOC- Chim. BeZg., 1904, xviii., 147-165.)-0n cooling a solution of acetic acid in a fat, two distinct layers are formed, the amounts of which vary with the nature of the fat. The glycerides with low critical temperature of solution, low refractive index, and low rnelting-points are the most soluble in the case of butter, whilst in the case of other fats arid oils the glycerides that dissolve first have high refractive indices.The following method of analysing butter is based on these facts : Five grammes of the melted and filtered butter and 10 C.C. of acetic acid (specific gravity 1.057) are heated to 60" C., with occasional shaking, in a tube 16 millimetres wide graduaked in millimetres. The liquid is then cooled to 40" C., and the amount of the upper layer measured. The cooling is continued, and at 35" C. the lower layer is decanted into a, flat dish, and evaporated at about 70" C. until the residue becomes constant in weight (extract A). The portion left in the tube is treated in the same way with 10 C.C.of acetic acid, yielding extract B and a residue C, each of which is dried and weighed and used for a determination of the critical temperature of solution, Valenta figure, and refractive index. The following are typical results thus obtained : 1 BUTTER. ARACHIS OIL. SESAME OIL. Mixtures of the above butter with margarine gave the following results :THE ANALYST. 285 Fraction. A ... ... B ... ... c ... .,. BUTTER + 10 PER CENT. MARGARINE. 40 45 65 f 20 PER CENT. MARGARINE. -l--i- 1 MARGARINE. +50 PER CENT. MARGARINE. -1- -I - 56 35 62 57 83 1 413 l The butter-fat had the following constants : Reichert-Meissl value, 25.3 ; critical temperature (neutralized butter), 51 ; Valenta figure, 47 ; Burstyn degree, 2 ; and refraction at 40" C., 42-8.The solubility of butter does not depend on the amount of volatile acids it contains. When a mixture of butter and margarine is treated in this way, there is a concentration of the latter in the fractions separating, the amounts dissolved by the acetic acid not being proportional to the respective coefficients of solubility of the butter and other fats. The author regards a figure obtained by subtracting the critical temperature of solution of fraction A from that of fraction C as the most important factor. This figure was 11-5 on the average in the case of eleven samples of pure butter, while butters containing 10 to 15 per cent. gave an average figure of 15.5. Similarly, the mean differences for the Valenta figures C - A were 22 for pure butters and 27 for the adulterated samples.Although in some cases pure butters gave results as high as those containing 10 to 15 per cent. of margarine, the author considers the method may still be of service in the case of butter with a Reichert value of 23 to 27; for in his opinion such samples, giving a value of less than 11 for the critical temperature C - A, are to be regarded as pure. C. A. M. The Amount of Cocoa Butter contained in the Cocoa Bean. S. H. Davies and B. G. McLellan. (Jozim. SOC. Chenz. Incl., xxiii., 480.)-The cells of the cocoa nib are not permeable to fat solvents, and must therefore be broken down mechanically before the fat is extracted. Neglect of this precaution is the probable cause of the low results obtained by earlier investigators.The determinations were made by breaking up the fully roasted cocoa beans, and removing the husks by winnowing. The nibs were next finely ground on stones in a small mill. The pasty mass was cooled quickly to prevent separation of " butter," and the mass of cocoa sampled. Two grammes of this sample were then extracted in a Soxhlet apparatus with the fraction of petroleum ether boiling between 40" and 50" C. The extraction was allowed to proceed for several hours, the solvent standing on the cocoa overnight. The extracted fat was dried at 95' C. for four hours and then weighed. The following results were obtained :286 THE ANALYST. Origin of Bean, Ecuador ... ... ... Venezuela ... ... ... Dutch Guiana ... ... Brazil ... ... ... Africa ...... ... Ceylon ... ... ... Mean Percentage of Fat. 53.76 52.19 56.36 54 a 6 5 54.18 53.36 Origin of Bean. Mean Percentage of Fat. I Trinidad ... ... ... Grenada ... ... ... Dominica .. I ... ... Santo Doming0 ... Unfermented ... Highly fermented Jamaica ... 54.57 55-30 55.03 55.38 54.68 58-23 General mean, 54.44 per cent. fat. Prolonged fermentation of the cocoa bean causes a loss of carbohydrate and albuminous material, yielding a product richer in fat. The variation in the amount of fat in cocoa beans of different origin is not great, and the determination yields no clue as to the geographical source of the bean. A. R. T. The Chemical Composition of the Inner Shell of the Coffee Fruit. Bela von Bit& (Jounz. Landw., 1904, hi., 93 ; through Chem. Zeit. Rep., 1904, xiii., 158.) -The following is the composition of the inner shell of the fruit of the coffee-tree (Cafea Arabica) : Calculated on On the the Dry Sample.Per Cent. Per Cent. Moisture ... ... .. - . 11.18 ... Ash ... ... ... 2.63 ... 2 *96 Soluble in ether ... ... ... 1.15 ... 1-29 Nitrogenous matter ... ... ... 5-50 ... 6.19 Extractives free from nitrogen ... 20.66 ... 23.26 Fibrin ... ... ... ... 58.87 ... 66-28 99.99 ... 99.98 Proteids (Stutzer) ... ... ... 3-94 ... 4.43 -- -- The fat is of a greenish colour, and has the following constants : Koettstorfer number, 141.2 ; acid number, 82.7 ; ester number, 58.5 ; free fatty acids (as palmitic), 37.84 per cent. ; glycerides (as tripalmitin), 28.03 per cent. The shell contained 0,022 per cent. phosphorus (on the dry sample), equal to 0.58 per cent.lecithin ; and caffeine, 0.35 per cent. The extractives contain appreciable quantities of pentosan ( C5H,0,), Chalmot and Flint's method giving 21.50 per cent. (on dry sample). 48.59 per cent. of the ash is soluble in water at ordinary temperatures. H. A. T. The Detection of Alkali Fluorides in Flesh Products. J. Froideveaux. (Journ. Pharm. Chim., 1904, xx., 11, 12.)-The following method is used in the Paris Municipal Laboratory : About 30 grammes of the finely-divided substance are mixed with 1 to 2 C.C. of a 50 per cent. solution of sodium carbonate and incinerated at a dull red heat in a platinum crucible. The charred mass is powdered, and boiled forTHE ANALYST. 287 some minutes with 5 to 6 C.C. of water, and the extract filtered, cooled, and mixed with an excess (2 to 3 c.c.) of hydrochloric acid and a few drops of helianthin solu- tion.A saturated solution of ammonium acetate is now added until the colour changes to yellow, and the liquid tested for fluorides by adding 1 to 2 C.C. of a 20 per cent. solution of calcium chloride. If a turbidity or precipitate be obtained, the presence of a fluoride is confirmed by the usual test with sand and sulphuric acid applied to the washed and dried precipitate. The method is capable of detecting 0.5 gramme of an alkali fluoride in a kilogramme of meat, The presence of acetic acid prevents the precipitation of phosphates by the calcium chloride, while it pro- motes the precipitation of fluorides. C. A. M. The Essential Oil of Artemisia Herba Alba. E. Grimal. (Bull. Xoc. Chim., 1904, xxxi., 694-697.)-This plant is very abundant in Algeria, where it is greatly esteemed for its medicinal qualities, By distillation of the fresh plant with water, the author obtained a yield of 0.3 per cent. of a greenish-yellow oil, with a strong aromatic odour, and a somewhat bitter taste resembling that of camphor. I t was soluble in 25 parts of 60 per cent. alcohol, in 2 to 2-5 parts of 70 per cent. alcohol, and was very soluble in 80 per cent. alcohol and the ordinary solvents for essential oils. I t gave the following analytical values: Specific gravity at 15" C., 0.9456; refractive index at 20" C., n, = 1.47274 ; [a]L, = - 15" 30 ; acid value, 6.46 ; saponifica- tion value, 89.23. The esters, calculated as CH,.COOCloH17, amounted to 31.15 per cent., and the free alcohols (as C,,H1,O) to 12.65 per cent. There were also identified among the constituents-Z-camphene, cineol, and esters of caprylic and capric acids. C. A. M.