190 THE ANALYST. ABSTRACTS OF PAPERS PUBLISHED IN OTHER JOURNALS. FOODS AND DRUGS ANALYSIS. A Source of Error in Optical Sugar Analysis. F. G. Weichmann. (School of Mines Quarterly, 1904, xxv., 183-193.)-The presence of the lead precipitate formed in clarifying cane-sugar solutions is a source of error which has been known for years, but which is usually ignored. From the results of experiments described in this paper, it is seen that the volume of the precipitate varies from 0.05 C.C. to 1.1 c.c., no two samples of sugar giving equal amounts of precipitate. The most voluminous precipitates are not always obtained from low-grade sugars, and the composition of the impurities in the different kinds of cane-sugar must vary, for the specific gravities of the lead precipitates differ widely, from 1-65 to 4.38.An average of + 0.25O Ventzke may be taken as the average plus factor of error from this cause. w. P. s. Estimation of Fat in Cheese. B. Sjollema. (Chemisch Weekblnd, 1904, No. 29.)-The author considers that the difficulty of extracting the whole of the fat from cheese dried at 100' C. is due to its becoming incorporated with the casein, which latter coagulates at a high temperature and becomes horny. He therefore concludes that it may be possible to secure good results if the heating be avoided. He tried the three following methods of estimating the fat. ( u ) The cheese was ground both with and without sand, dried for an hour at about 100' C. in a current of gas, and subsequently extracted in a Soxhlet apparatus with ether.After evaporating the ether off, the fat was dried for an hour and a half and weighed. ( b ) This is similar to method a ; the cheese was, however, not dried at 100" C . , but in a vacuum exsiccator over sulphuric acid. ( c ) The Method of Bondzynski.::: The purity of the fat extracted by all these methods was ascertained from its Taking into account that there is always a little decomposition The agreement between the methods b and c is fairly satisfactory, but method The sample No. 2 wa8 not ground up with sand, and this serves refractometer figure. caused by the drying at looo C., the agreement is satisfactory (see following t.able). a gives low results. t o explain the difference of 0.75 per cent. between methods c and b. * For description of this niethud see f'~1lowi11g abstract, p.192.THE ANALYST. 191 Number of Sample. 1 2 3 4 5a .~ _ _ Method a. -- 12.6 13-4 17.0 3.2 12.7 12.1 15.0 Method b. 16.55 16.6 18.4 5.8 15.4 15.6 17.2 I Method c. _ _ 16.0 17.35 18.35 5.4 15.6 17.6 - Refractometer Figure at 25" C. Method a. __. _ _ _ _ 53.2 52.8 53.8 54.3 Method b. 53.8 53.0 54.3 56.0 - -____- Method c. 52.8 53.5 55.0 54.3 Sjollema found that the addition of a little 96 per cent. alcohol (5 C.C. for 3 grammes cheese), very much facilitated the grinding of the cheese. He tried three different methods : The mixture was washed with a little ether, dried in a vacuum exsiccator, and the fat extracted in a Soxhlet apparatus. (p) Similar to a, but filter-paper instead of sand was used to absorb the moisture. (6) The cheese was ground with a little alcohol, and brought at once into a flask After standing some hours with repeated shaking, the (a) The cheese was ground up with a little alcohol and mixed with sand.with about 50 C.C. ether. ether was filtered through a very dose filter, and the fat determinated. The results were as follows : Number. Method (Bondzynski), 17.6 per cent. 16.8 ,, 13.0 ,, 28-0 ,, Method a. 17.2 per cent. - - Method 8. 17.3 per cent. - - Method 6. - 16.8 per cent. 13.1 ,, 28.0 ,, To obtain the result with method p it was found necessary to extract twice with ether. More fat is also extracted when the mass is ground up repeatedly with sand. These results sbow that by using method 6 the fat is totally dissolved, and that it is not advisable to dry the mass before extraction, not even in an exsiccator.When using 50 C.C. ether for 3 grammes of cheese the whole of the fat is dissolved in the ether in a few minutes. I t was found that the presence of the small quantity of alcohol used did not influence in any way the purity of the fat extracted. J. J. L. v. R. Estimation of Fat in Cheese and Foodstuff'. H. L. Visser. (Hoorn 1. CHEEsE.-The author hw compared three methods for the estimation of fat in Chemisch Weekblad, 1904, No. 29.) Dutch cheese.192 THE ANALYST. (a) The cheesezwas ground up with sand, and dried at 100' C., the fat being subsequently extracted with ether. ( b ) The Method of Bodqnski.-From 3 to 5 grammes of cheese are weighed and placed in a dry flask, 10 C.C. hydrochloric acid (20 to 25 per cent.) added, and boiled for about ten minutes. After cooling, a known quantity of petroleum spirit (boiling-point 50" to 60" C.) is then added, the flask closed, shaken repeatedly, and put aside for a few hours to allow the petroleum spirit solution to separate, the fat being subsequently determined in an aliquot portion of this fat.( c ) The Gerber Method.-In this series of determinations Edam cheese, made at the Experimental Farm connected with the Government Agricultural Experimental Station, was used. The quantity of milk uaed was known, also the percentage of the fat in the milk and in the whey, as well as the weight of the cheese when fresh, and after being ripened ready for the market. Eence it was possible to calculate the percentage of the fat in the cheese with considerable accuracy, and to control the reliability of the different methods of analysis.The cheese was inade-as is usual in the provinces of North Holland-with a mixture of skimmed evening milk and full- cream morning milk. Nu. 1408 1409 1410 141 1 1413 1413 1414 Ether Extraction Met hod. { E) { G} { ;;:;] { g::) { ;;:: ;- 16.9 18.1 Bondzynski Met 11 od . 21.44 19.7 18.3 19.3 19.2 19.3 19.5 Gerlwr Met hod. j 20.9 '\21-5/ 19.5 18.5 19.0 20.0 19.0 19.0 Calcnlated. 21.8 20.6 18-4 19.4 19.6 19.4 From these figures it may be seen that there is a satisfactory agreement between the calculated percentage of fat and that obtained by the Gerber and Bondzynski methods, but that the ether extraction method gives low results, the differences being t 3 per cent. Dr. Visser 8160 analysed a, sample of full-cream American Cheddar cheese, and here the results show 8 satisfactory agreement in all the three methods.Ether Extraction. Hondzynski. Gerber. Per Cent. Per Cent, Per Cent. 36-5 ... ... 36.80 ... ... 36.8 31.0 ... ... 36.88 ... ... 36.9THE ANALYST. 193 A skim-milk Gouda cheese gave : Ether Extraction. Per Cent. 2.2 ... Bondzy iiski. Gcrber. Per Cent. Per Cent. ... 5.5 (3.83 \ 144s j * * - ... Several analyses were made in order to ascertain if the composition of the fat was altered by the boiling with hydrochloric acid, and it was found that no decomposi- tion took place. The author therefore recommends the methods of Gerber and of Bondzynski. 2. FOoDsTuFFs.-The method of Bondzynski cannot be used for foodstuffs containing cellulose, since, on boiling with HCl, these yield a thick magma from which the fat cannot be extracted by shaking out with petroleum spirit.He there- fore uses the method proposed by Berntrop for bread: 5 to 10 grammes of the sample are gently boiled for half an hour with 100 C.C. of HC1 (10 per cent.) in a beaker, covered with a clock-glass. After being cooled and distilled with water it is filtered, and the residue is washed with water until it has a neutral reaction. The wash water is then removed as completely as possible, the filter placed in a thimble, and dried at 100" C. in a current of gas. Subsequently the fat is estimated in the usual way by extraction with ether. The iodine value and the refractometer figure of the fat are also taken : Maize ...... Linseed cake ... i ) 1 ) ... Linseed meal ... 9 ) i i .. 9 1 1 1 ... 1 1 1 1 ... Rice meal ... . . . Meat meal ... Ground-nut cake? Gluten meal ... Gluten food ... Rape-seed cake . . . Cotton-seed meal.. . (Arachis) . . . 4.57 8.2 5.08 9.2 9.5 8.9 2.27 15.0 11.2 8.2 6.73 2.5 11-06 9.34 I 91.2 91.3 87.3 86.3 - 48.9 at 40" C. 63.5 at 40" C. - 60.0 at 40°C. 5-21 8-96 5.56 9.8 10.12 9.41 (3-20 15.52 9.7 -\ 3.37 (1;:: 8.1 { E;) 4.82 12-10 10.86 108.4 189-1 17 9.9 - - 86.6 132.3 109.7 108.9 92.8 90 -3 87-0 86.3 47.5 at 40" C. 65.5 at 40" C. 60.5 at 40' C.194 THE ANALYST. It will be seen from these figures that the HC1 method always gives the higher results, and especially so in the case of samples containing gluten. The fat from the meat meal was also found to contain kreatin.The fat of the ground-nut cake consisted almost entirely of free fatty acids. These facts may serve as an explanation of the high results given by the extraction methods. The fat obtained from linseed cake after boiling with HC1 was shown to be pure from its iodine and saponification values and its refractometer figure. For those foodstuffs containing molasses, which are composed of linseed cake meal, the only reliable method is the Berntrop one. J. J. L. v. R. The Composition of Meat Extracts and Yeast Extracts. J. GraE (zeit.fiir Untersuch. der Nahr. uitd Genussmittel, 1904, vii., 389-392.)-Frorn the results given in the following table, it will be seen that the chemical composition of yeast extract does not greatly differ from that of meat extract.For the determination of the xanthine bases Micko's method (ANALYST, this vol., p. 158) was employed : Make. MEAT KXTKACTS. ' I Praerie " Extract .. , Flagge " Extract .. Armour's Extract .. . " Terton " E x t r a c t , fluid ... "Terton" E x t r a c t , solid,, . . . . . . " Bolero Extract, fluid,, . . . .. . ' I Bolero E x t r a c t, solid ... ... Cihil's Extract ... Rio Boiiillon ... ... Tsssen Bouillon . . . Armour's ' I Vigoral " Armour's ' I Beef Juice " A r ni o u r's " Soluble I3cef" ... ... YEAST EXTRACTS. Siris . . . . . . . . Pana . . . . . . . . . Bcdu in ... ... 0l)ron . . . . . . . . , Ovos, fluid .. . .. , ,, solid ... ... Sitogen, fluid ... ... ,, solid ... .., Biou, h i d ... ... ) ) solid ... ... % 16.12 21 *37 21-56 61 *22 16-93 64.77 16.91 63.86 66-11 ti5 '62 55.12 57.79 20'65 28.45 60.52 55.81 66.50 71.09 25'99 61 -5 1 35.43 31 '73 27-92 _I % 9.74 10.01 9 -32 3.64 i.91 2-49 9'94 3.85 1.85 1'89 5.21 4.97 9-56 7.22 1.53 2-67 2.05 2.97 5.71 1 '98 5 -33 4 '54 6.68 z i 22 qz i E - % 2.26 2-78 2-58 1-42 0.82 0'60 1 *98 1.91 0.29 0.92 0.28 0.63 1 *58 0.50 0.19 0.33 0 -29 1-18 0.33 1-41 0.71 1 *02 0.28 5i 23 :; Q hd 3% - % 5-65 5-21 2-97 1 *37 1 *76 0.92 1-98 2-47 0 **52 0'42 0'49 0-53 5.88 2-68 0'58 1'19 0.83 3-03 0.85 1 -79 2.20 3.84 1 -38 - kLt 55 %$ 2r.G u K % 1-38 1.35 1-58 1.20 1-24 3.36 3-48 3.35 1-76 0'34 0-47 D -36 0.39 0'53 0'41 0.27 0.3c 0.2: 0'3( 0'31 0'3: 0'4: 0.26 x 13'90 19'73 20 -25 15'37 23.66 16'69 17-49 13'90 15.43 25 '1 4 16.27 13-67 12.75 15.15 21 '5C 19'4.E 17'4% 25'54 17*3( 22'9: 20.6: 21'5f 2 o a 9 .- 05 2 32 8-7 a- 0 s % 3'65 4 '64 5'01 1 -89 4'75 - - 3'58 0'84 ' L ' i l 2.58 3.09 6'9:3 1.49 2'68 1'99 3 '29 5-67 2-35 5.49 3'65 5-07 - - w. P.s.THE ANALYST. 195 On the Valuation of Quinine by Andre’s Reaction. E. L6ger. (Journ. Phurnc. Chim., 1904, xix., 434-435.)-The author describes a series of experiments which show that up to a certain limit the poorer the solution is in quinine the brighter is the coloration given by AndrB’s reagent (ANALYST, this vol., p. 160). He concludes from thie that the colorimetrical method given by the Swiss and Italian Phsrmacopaeias ought to be abandoned. C. A. M. Estimation of Lead in Citric and Tartaric Acids, and in Cream of Tartar. C. T. Bennett. (Chemist and Druggist, 1904, lxiv., 633.)-The following method is proposed, being based upon the coloration produced by adding sodium sulphide to ammoniacal solutions of the acids or cream of tartar. Ten grammes are dissolved in 15 C.C. of water, 25 C.C. of a 10 per cent. ammonia solution are added and the volume diluted to 50 C.C. One drop of a 10 per cent. sodium sulphide solution is now added, and the coloration produced is matched in Neesler glasses by adding from a burette a standard lead acetate solution (containing 0.0001 gramme of lead in 1 c.c.) to 50 C.C. of water to which a drop of sodium sulphide solution has previously been added. If iron or copper be present, the addition of 1 C.C. of a 10 per cent. potassium cyanide solution is necessary. A yellow coloration sometimes caused by the potassium cyanide may be matched before adding the sodium sulphide, and the amount of standard lead solution so used deducted from the total quantity required. w. P. s.