i n I'll > m ml ! Ill' ^m fork HuU fflolljgf of Agrtrulturp At (dnrttPU UnittPrattg Ktljata. N. ^. Htbrarg GlfT OF .BJ5..Q.E.. :KH..mM....,.,_ Reprinted from The Brooklyn Medical Joubnal, March, 1889. CONDENSED MILK, WITH REFERENCE TO PREVIOUS SKIMMING.^ BY WALTER H, KENT, PH.D., Chemist of the M Brooklyn Department of Health. During the summer months the attention of the Department of Health was called to the fact that a certain brand of condensed milk sold in the market was made from skimmed milk, and in accordance with directions the particular brand complained of, and also the various other brands here sold, have been obtained and analyzed with refer- ence to fraud of this kind. Before presenting the results of this examination, it may be well to consider briefly the methods of determining by an analytical process, whether a given brand of condensed milk is really made from milk wholly, or in part, deprived of cream. With ordinary milk this question may generallj#be answered by knowing the specific gravity and the amount of fat which the milk con- tains. But, when milk is condensed by evaporation, afeig the case with that sold in bulk from the wagons, or condensed and preserved by the addi- tion of 30 to 40 per cent, sugar, as is the case with most of that sold in cans, the question becomes somewhat more difficult. As the den- sity, and also the amount of fat, may vary with the amount of condensa- tion, as well as with the amount of sugar added, these factors either alone or together, would not be sufficient to determine whether the milk had at any time been robbed of one of its most valuable con- stituents. The method here used is that given by Dr. J. Koenig in his cele- brated work on the analysis of foods (Die menschlichen Nahrungs- und Genussmittel, Vol. II., p. 272). Dr. Koenig makes the statement, that for every 100 parts of albu- minous substance in cow's milk there are from 100 to no parts of fat, and that this relation must obtain in condensed milk. Or, in other words, the amount of fat must equal or exceed the amount of albu- minous substances. It will be remembered that the albuminous substances of cow's milk, or albuminoids, as they are frequently termed, are substances > Report made in November, 1888, to Dr. John Griffin, Commissioner of Health of the City of Brooklyn. 2 WALTER H. KEi\T, PH.D. containing uniformly about 15.8 per cent, of nitrogen, and that they are of a sufficiently stable character to withstand the process of the evaporation of the milk without loss. As by the evaporation of milk there is no loss of fat or albuminous substance, and as by the addition of pure sugar neither fat nor albu- minous or other nitrogenous substance is added, the ratio of fat to the albuminoids must be the same in condensed as in normal cow's milk, and on these facts is based the method which Koenig gives. Before accepting the statement, however, that the amount of fat in cow's milk does really equal or exceed the amount of albuminoids, and basing conclusions as to whether a given condensed milk is made from skimmed milk, it may be well to examine the statement as to whether it is really true This can only be shown from the analysis of nor- mal cow's milk with reference to the amount of fat and albuminoids such milk contains. Fortunately, such analyses have been made so many times by chemists of good repute that little more is necessary than to examine the published results. (See Koenig's Nahrungs-und Genussmittel, Vol. L, p. 30.) Below are given abstracts from the liter- ature bearing on this point. In January and February of 1883 Dr. S. M. Babcock made forty analyses of mix^ morning and evening milk, in every case of which the fat exceeded the albuminoids (Second Annual Report of the N. Y. State Agricultural Experiment Station, 1883, p. 103.) Dr. H. W. Wiley gives one hundred and forty-nine daily analyses of milk from Maythorpe dairy. College Station, Md., and eight of milk from various other sources, and in all cases the fat exceeded the albuminoids. The average of the 149 analyses was : Fat 4.08 per cent. Albuminoids 2.78 per cent and of the 8 analyses : Fat 3.91 per cent. Albuminoids 2.47 per cent. — (U. S. Dept. of Agriculture, Division of Chemistry, Bulletin, No. 13, p. 115.) Alexander, Mueler and Eisenstuch made fifteen analyses of morn- ing and evening milk at various times during the summer of 1864 ; of these 1 2 contain more fat than albuminoids, while the three exceptions are found to contain less than .2 per cent, more albuminoids than fat (Die landwirthschaftlichen Versuchs-Stationen, Vol. V., p. 161, and Vol. VI., p. 3.) CONDENSED MILK— PREVIOUS SKIMMING 3 P. Pertersen made five analyses from the Oldenburger breed and full blood shorthorns, in which the fat of four was in excess. In the case of the single exception, the excess of albuminoids was less than i percent. (Milchzeitung, 1876, pp. 2179, 2 191.) Vernois and Becquerel made analyses of the milk of eleven differ- ent breeds at the Agricultural Exhibition in Paris in 1856, and in ten of these the amount of fat exceeded the amount of albuminoids, the exception being with a Norman breed (Die Naturgesetze der Fuetter- ung, von Gohren, 1872, p. 466.) J. Moser and G. Belleville made fifteen analyses of the milk of different breeds exhibited at the Vienna Exhibition in 1873, and in every case the fat exceeded the albuminoids. Ten analyses were also made of other milks in which the average fat was 4.02 per cent, and the average albuminoids, 2.67 per cent. (Milchzeitung, 1874, p. 915). W. Eugling analyzed the milk of an Italian cow, and found 4.54 per cent, fat and 3.49 per cent, albuminoids (Milchzeitung, 1877). E. Marchand made eighteen analyses of milk from eighteen differ- ent breeds at the World's Exhibition at Paris in 1878, in all of which the fat exceeded the albuminoids (L'industrie laitiere, 1878, No. 46.) K. Portele analyzed twenty samples of milk from three different breeds, morning and evening, while experimenting \Mith diff'erent kinds of feed, in January, 1880. In all cases the fat exceeded the albumi- noids (Die landwirthschaftlichen Versuchs-Stationen, Vol. 27, p. 133.) A.Voelcker made eleven analyses of cow's milk, of diff'erent breeds, at pasture alone and at pasture with varying amounts of oil-cake, in nine of which the fat was in excess (Journal of the Royal Agricultural Society of England, 1861, p. 33, and 1863, pp. 302, 309 ) J. Lehmann analyzed fourteen samples of milk of shorthorn and Holland cows while feeding various kinds pf foods. In thirteen cases the fat was in ■ excess ; the exception being with a Holland cow on winter food, in which the fat was 3. 11 per cent, and albuminoids 3.27 percent. (Der Land wirth, 1869, pp. i, 9). At another time Lehmann made four analyses, and in every case the fat was in excess (Zeitschrift des landwirthschaftlichen Vereins in Bayern, 1869, p. 261.) M. Fleischer made ten analyses of the milk of two cows under varying conditions of feed, and in all cases the fat exceeded the albu- minoids (Journal f. Landwirthschaft, 1871, p. 371 ; 1872, p. 395). Kuehn, Gerver, Wackwarth, and Kisielinski give eight analyses of the milk of two cows fed with variations of food, in all of which the fat exceeded the albuminoids (Saechsische landwirthschaftliche Zeitung, 1875, p. 153.) Kuehn, Aarland, Baesecke, Dietzell, Haas, and Schmidt publish forty-two analyses (Journal f Landwirtschaft, 1874, pp. 168, 295 ; 4 WALTER H. KENT, PH.D. 1875, p. 481; 1876, p. 173; 1878, p. 168) of the milk of eight differ- ent cows, also made while experimenting on the effects of different kinds of feed. In thirty-nine cases more fat was found than albumin- oids. The three cases 'in which the albnminoids are in excess are all with the same cow, and may be counted as one exception. The figures in this case are as follows : 2.73 per cent. fat. 2.78 per cent, albuminoids. 2.82 " " 2.85 " 2.69 " " 2.72 " " In the table below, two analyses are given of bottled Orange County milk obtained from large dealers in the market of this city, both of which show more fat than albuminoids. As the milk in such cases is received into large tanks from which the bottles are filled, such sam- ples may be considered as the average milk of a number of cows. In the great number of analyses mentioned in the foregoing, it will be seen that the cases in which the albuminoids exceed the fat are exceedingly rare. In the exceptions which do occur the differences are slight, and in all cases they are with foreign milk and not with milk of 'this country. It is impossible to say to what extent the exceptions are due to typographical errors, or errors in analysis, as might occur, by the use of Wanklyn's method for the determination of fat, which is known to give uniformly less fat than the milk contains. We do know that the methods of milk analysis have been much improved in recent times, while it will be noticed that six of the nine exceptions occurred as *'ar back as 1864. Some of the recorded results are excluded from con- sideration here, either for the reason that they are known to have been made by Wanklyn's method or that they were made too long ago to be thought reliable. One of the exceptions occurred in March, and possibly it is due to the presence of colostrum which contains on an average four or five times as much albuminoids as fat. It is also suspected that other exceptions may be due to a diseased condition of the cows, in which case an excess of albumen in the milk would be somewhat analogous to albumen in the urine in certain dis- eases. In this connection may be given the following interesting fig- ures taken from some analyses of the milk of diseased cows made by A. W. Blyth (Journal of the Chemical Society, 1879, ?• 534)- CONDENSED MILK - PREVIOUS SKIMMING. 5 1st. Milk of cow suifering from pneumonia ; Pulse 82 Temperature 102.4" Fat 2. 965 per cent. Albuminoids 4.29 per cent. 2d. A case of phthisis with a cow five years old, where there were extensive tubercular deposits in right lung : Dec. 7, 1878. Feb., 1879. Fat 2.77 per cent. 3.83 per cent. Albuminoids. . . .4. 517 per cent. 5-765 per cent. 3d. A cow two years old in an advanced stage of phthisis : Jan. 29th. Feb. 17th. Fat 2. 599 per cent. 3. 28 per cent. Albuminoids ....3.00 percent. 3.98 per cent. 4th. "A sample of milk drawn from an udder actually infiltrated with tubercular deposits :'' Fat o. 49 per cent. Albuminoids 3 597 per cent. The rule which Konig gives is, therefore, good, and with the normal milk of a number of healthy cows taken together it is true that the amount of fat exceeds the amount of albuminoids. If market milk, or condensed milk, or milk condensed and pre- served with pure sugar, is found to contain less fat than albuminoids, it may, therefore, be concluded that it is skimmed milk or made from milk deprived of cream to an extent at least proportional to the excess of albuminoids. There may also be the suspicion that it contains colostrum or the milk of diseased cows, though, with condensed milk this would not be probable, and it would hardly be just to the producer of the article to think this to be the case without other proof or circumstances which would support such a conclusion. METHODS OF ANALYSIS. Taking the Sample. — In each case below given, the sample was bought in the market as any consumer would buy. Condensed milk preserved with sugar in cans, as is known, is in a viscid, semi-solid condition. From the manner of its preparation it was at first not thought necessary to take the precaution of mixing the sample before weighing out the portion for analysis, but this was soon 6 WALTER H. KENT, PH.D. found to be an error. It occurred to me that possibly during summer weather, or otherwise, the cans might become heated and sufficiently fluid to permit the fat to rise in the can, so as to make a difference in the composition of the two ends. A couple of experiments were there- fore made to settle this point. I St. With Borden's Eagle Brand. At one end the fat was found to be 6. 1 6 per cent., and at the other end 4.94 per cent. 2d. With the Anglo- Swiss Condensed Milk. At one end the fat was found to be 5.72 per cent., and at the other 6 26 per cent. Accordingly, in the work below given, the samples were thoroughly mixed before weighing out the portion for analysis. The weighing is conveniently done in a small bulbular pipette, into which the milk is drawn by the use of a Sprengle pump. The various portions for analysis are removed by attaching a short piece of rubber tubing and blowing out the viscid material and drawing back the adhering drop before again weighing the pipette. ALBUMINOIDS. As above stated, the albuminoids of milk contain 15.8 per cent, nitrogen. A determination of the amount of albuminoids may there- fore consist in first accurately determining the amount of nitrogen and then in multiplying the weight of nitrogen by the factor 6.33. The nitrogen is determined by Kjeldahl's method, as described by the Association of Official Agricultural Chemists (U. S. Department of Agriculture, Division of Chemistry, Bulletin No. 16, p, 64.) One or two grammes of condensed milk is treated in the pear- shaped Kjeldahl flask with about 20 c. c. strong sulphuric acid and a small amount of mercuric oxide, at a temperature near the boiling- point of the acid, until nearly colorless. While still hot the oxidation is completed by the addition of minute particles of potassium perman- ganate. After cooling it is somewhat diluted, with water, about 25 c. c. potassium sulphide added, and transferred to a retort with bent neck containing a few granules of zinc ; 50 c. c. strong sodium hydrate solution is then added through a safety funnel, and the ammonia dis- tilled off through a tin-lined condenser and collected in deci-normal hydrochloric acid. It is then tritrated back with deci-normal sodium hydrate solution, using cochineal as an indicator. Every cubic centi- metre of deci-normal hydrochloric acid neutralized by the ammonia corresponds to 1.4 milligrammes of nitrogen or to 8.862 milligrammes of albuminoids. FAT. About one gramme of the sample isdischarged from the, weighing bulb into a plenty of sand which has been freed from organic matter CONDENSED MILK— PREVIOUS SKIMMING. 7 by ignition. The milk is then saturated with sand by the aid of a small glass rod and allowed to dry at ioo° C. It is then pulverized and transferred, together with the rod, to the Soxhlet extractor (Journal of the American Chemical Society, Vol. IX., p. 182), where the fat is thoroughly extracted with ether. WATER AND ASH. One or two grammes of the sample is transferred to a small dish which has been previously weighed with a small glass rod. It is thoroughly dried at 100° C, with occasional stirring. The water is determined from the loss in weight. About the same quantity is transferred to a platinum dish, dried and ignited at a dull red heat, and the ash weighed. The samples analyzed are included in the following table. From the rule which Konig gives, and which for healthy milk abundant evidence shows to be true, it must be concluded that sam- ples 3 to 19 inch were made from milk the cream of which had been more or less removed. In some cases, where more than one analysis is made of the same brand, a considerable difference is noticed in the results. As con- densed milk is made in batches, such differences may of course arise from unequal evaporation or the addition of varying amounts of sugar, or, where skimming is done, by removing unequal amounts of cream. Wer? the milk not skimmed, the latter chance for inequality would be removed and a compound of more uniform composition would be ex- pected. With regard to the sanitary effects of condensed milk made from skimmed milk, and especially its action on young children, I refer to the Report of the Brooklyn Health Eepartment for the year 1886, p. 60, and also to some remarks by Dr. E. H. Bartley in Vol. I,, p. 180 of The Brooklyn Medical Journal. It will be remembered that milk sold in the market must contain, according to legislative act (Laws of 1887, Chap. 430), at least 3 per cent, of fat; while with several brands of condensed milk, if diluted according to the directions given for making milk, they would contain much less than that amount. The 25 or 30 per cent, of water of the condensed milk is, of course,' a remaining portion of the 87 or 88 per cent, which ordinary milk con- tains, 'i/sy 'lupy .If J PQ •^3)Vj^ '^USJ ,iPJ spouiiunqiy ^O CO . ro , 00 00 00 ■* ■ -"f rf- CT^ d> (^ ror--oo 00 CO 00 20 r^ t^ 1^00 00 00 00 00 a\ 0^ OiOO o o "fVJ •}U?3 A?^ r^hHOO^OyD i-i N O ■^OO I- 00 m O 00 t^OO w ro -^ i>- ro t^ OCOOO O ■-■ 0>N r-~0 "^00 M^HH ^ l-l^^t^lr^O^'^^->.0^•~' -/ •y-; ■H tc -K «i 42 yi4^ tj - u >- u . (_) - o - o : Th O f^ r^ Ln N O (^ sO o HH t-l c c G G c c c rt = ft C rt rt ri rt rt <:i a rt C!J d o u U o CJ o N - £ :S jd £ " ■£ £, " 1^ l-< i <;» i= s. I c o 2 « ; u o ' ■J^ 6 « P3 , o- .SP Pi' -OA^ W ro -^ "-)^ t^OO o^ o ^_ Cornell University Library SF 259.K37 ^ Condensed milk, with reference to previou 3 1924 003 007 196