186 THE ANALYST. NOTES ON MILK ANALYSIS. BY M. DECHAN AND T, MABEN, Analytical Chemists, Eawick. IN undertaking the series of milk analyses, the results of which are appended to this paper, we had two objects in view, viz. : to demonstrate the practical utility of the system of fat extraction which we have for some time past employed in preference to the methods in common use, and also to escertain whether the limit adopted by the Society of Public Analysts was or was not too high. The experience of the working of the Adulteration Act has proved how difficult it is to obtain convictions, even in very glaring cases, owing to the different results obtained by dieerent analysts, as well as the lack of a uniform staudard. It would be absurd to expect the occupants of the judicial bench to be acquainted with all the intricacies of the various methods of work- ing, and hence we must not think them unreasonable if they look solely to results; accordingly, when these fail to agree, a judge has little difficulty in dismissing a case.It is of the utmost importance, therefore, that a uniform method of analysis should be adopted, no less than a uniform standard. On both of these particulars, the report, of the ‘ 4 Milk Committee ” of the Society is eagerly awaited, and it is to be hoped it will be the means of settling this vexed question. Pending its appearance, we submit for consideration our method of working and the results we have obtained. Before proceeding further, however, we may briefly indicate our objections to the methods in more general use.Regarding the somerset House process as described by Dr. Bell, one of its most objectionable features, in our opinion, is the filtration necessary to remove the fat from the solids not fat, as there is always a difficulty in being certain. that the last trace of fat has been remoxed from the 6Iter. We have also found the method approved by the Society to be unsatisfactory in so far that the residue obtained after evaporation is covered by a thin glossy film, and is thus in the best possible con- dition to resist complete fat extraction ; indeed, it is admitted that perfect separation is not obtained, and we are inclined to believe that this part of the process is to blame forTEE ANALYST. 187 certain incongruities in the results obtained from the same milk. The limitation of time for drying is also objectionable unless the size and the shape of the vessel be distinctly specified.We have obtained very different results from the same milk on weigking at the end of three hours, by simply substituting a round-bottom for a fiat-bottom vessel ; with the latter alighter weighing is invariably obtained. After a series of test experiments which led us to conclusions similar to those arrived at by Mr. Hehner with regard to the insolubility of milk sugar in ether, we adopted the following process with very satisftwtory results. The total solids are determined separately in a shallow vessel with a flat bottom, the quantity of mjlk taken being 5 grms, The vessel is large enough to allow the residue to form in a thin film, and there is no difficulty in obtaining weighings which correspond with the weight of the fat and non-fat solids.This therefore, forms a good check, being €or all practical purposes a duplioate analysis. For the fat, and aolids not fat, ten grams are weighed into a capsule capable of holding double that quantity. This is placed on an open water-bath and the milk stirred repeatedly during evaporation. By this means a granular residue is obtained, which, when reduced to powder, is in the best possible condition for the extraction of the fat. This we accomplishwith ether by means of one of the many forms of extraction apparatus modelled on the principle of Soxhlet’s tube. The fat and solids not fat are determined separately, both being dried till they lose not more than 0*001 gram in an hour.We have been in the habit of using two extraction tubes, both of which are peculiarly well suited for the purpose in hand. The &st is that devised by Messra. Dunstan and Short, of the Pharmaceutical Society’s Laboratory, and figured in the Phamnaoczctica;lJouma;l (vol. xiii, p. 664), and the other, that of Mr. West-Knights, as described by him in the ANALYST (vol. viii, p. 65). On the whole, we prefer Nr. West-Enights’ apparatus, which is less complicwted than the other, more easily managed, and not so liable to accident. In actual practice, however, we h d that rather better results aro obtained by tapering the lower end of the tube and inserting a small plug of cotton wool in the neck thus formed. Those who have already used this apparatua for other purposes can readily imagine its great value for the extraction of milk fat, and we now bring it forward with the view of inducing its more extended use in this direction.We find that from one and a-hali? to two hoursis quite sa%cient for the perfect separation of the fat, and when ether is used as the solvent there is no risk of any of the solids not fat being dissolved. The advantages claimed for this process over those which consist in msceratisn and filtration are very considerable. The solvent irJ kept at the boiling point, which cannot be obtained by any other method, the form of the tube gives the maximum of extracting power with the minimum of loss of ether ; the apparatus is simplicity itself, and is SO little liable to accident that whoh once set in working order it needs no further attention till the extraction is cowplete.Any number of oxtractions can thua be carried on simultaneously, and this of itself ipI a very great advantage,______-___ -- ------ 188 THE ANALSST. The second object of our experiments was to ascertain for our own guidance, whether the limit adopted by the Society is or is not too high. For this pui'pose, we selected a small dairy of ten cows, and took samples from each of these morning and evening. These on analysis gave UB the results as expressed in the following. table, the average of which very nearly corresponds with two analyses made of the mixed milk obtained from the same dairy in the regular course, which are also appended. It is a well-known fact that the first portions of milk drawn from a cow, at any given milking, contain much less fat than the last portions. It is easy to understand why this should be so.The fat naturally floats more or less on the surface of the milk, and it is only when the udder becomes partially emptied that milk rich in fat begins to flow. We had samples drawn toillustrate this fact, and the analyses of these are also given. From the very low figure8 which are aometimes given as a standard by certain well-known analysts, as wellas by the authorities at Somerset House, it is perfectly possible to con- ceive that samples from only one portion of the milking had in some cases been analysed as representative milks. If, for instance, the sample were obtained at the beginning of the milking, it would be poor in fat, and if at the end, it would be poor in solids not fat.Granted, therefore, that the lowest of a number of fats, irrespective of its complementary non-fat solids, and similarly, that the lowest solids not fat irrespective of its fat, be taken at3 the limit, me might easily have a standard, which for poverty could never even be approached by a natural milk ; but, obviously, this would be altogether nnfair. It is unnecessary to refer further to the tables than to point out that alike in fat and non-fat, the average is far above the limit adopted by the Society. Taking the single corns, we find that in only three cases are the non-fat solids under the limit, but these aro more than made up by the fat. On calculating their value by the factors suggested by Mr.Estcourt in the ANALYST (vol. viii, p. 245), they are found to be all above the limit. Cows No. 1 and 5 were said to be poor milkers, but even their milk is higher than what would pass for genuine. In no case is the percentage of fat lover than 2.7, while in mixed milk it is as high as 3.25. It is quite true that the first portion of the milk of No. 10 cow--an exceptionally rich milker-gives as low as 2-05, but it would be utterly wrong, as we have already pointed out, to assume that this represented the true amount of fat in the milk. In a recent number of the ANALYST (vol. viii, p. 248), Dr. Dupr6 submitted a table for ascertaining the relative proportions of milk mgar and proteids in the solids not fat, by calculation from the specifio gravity.We have compared the results so obtained with the actual analysis, and we find that while in Borne cases the figures come very near, in others they are somewhat vide of the mark. Bor example, No. 5 morning milk gives by calculation 5.54 of milk sugar and 2.54 of proteids, whereas the actual analysis gave 4.18 and 3.9 respectively. On calculating the specsc gravity from the factors suggested by Dr. Dupr6, we get the following result : Constituents, Iiifluenoe on Gravity. 4*18x3'7 .. ,. . , .. 15.466 $ k%%i 3-9 x 2.55 b , , , , , , . 9.945 0 ' 1 2 ~ 7 ' 5 . b b b 6'4 3'0 X0.725 b .* 2.175 +A& - Fat Calculated specific gravity 1038w63Q But in reality the specific gravity was found to bo 1030*3.THE ANALYST. 189 From this it is evident that the factors are not quite correct, or that the gravity is not so wholly dependent on the principal constituents of the milk as it issupposed to be.Of course, it must be remembered that Dr. Dupr6 makes no claim for perfect accuracy for the figures suggested by him. If carefully followed up, this line of inquiry is certain to have valuable results, as it is possible that the composition of the proteids themselves and also of the adh, may yet be ascertained by calculation. In concluding this paper WQ would submit for the consideration of the 6rbdilk committee '' the following suggestions :- I. That the total solids be determined separately ; and if time be specified, that the quantity in weight of milk, and the shape and size of the evaporating vessel, be stated.2. That the solids for fat extraction be obtained in a fine granular condition by ropeated stirring during evaparation. Using ten grams of milk the time necessaryfor this need not exceed I-&, or at the very utmost, 2 hours. 3. That the fat be extracted insome such apparatus as WQ have recom- mended; time allowed for extraction to be from I+ to 2 hours, 4. That all the constituents be weighed, viz. : Total solids, fat, solids not fat, and ash. This gives practically duplicate analyses and forms a valuable check on the accuracy of the results. 5. That the fat and solids not fat, be both considered by the analyst in estimating the genuineness or otherwise of the milk. 6. That in the case of mixed dairy milks the limit should not be lower than that at present adopted by the Society, viz.fat 2.5 per cent., and solids not fat, 9.0 per cent., or their equivalent, as calculated by some such method as suggested by Mr. Estcourt. 7. That in the case of singlo cows the limit might belowered to fat 2.5, and solids not fat, 8.5, or their equivalent. The fixing of two limits would of course require it to be stated when the milk is sent for analysis, whether it is from a single cow or from the mixed milk of a dairy. NO. 1 2 3 4 5 G 7 8 9 10 1 2 3 4 6 G 7 8 0 10 Cows, atall fed. Avr~bire.. .. Ciossbrod , . Do. .. .. Do. .. .. Ayrshire ., Do, .. ,. Shorthorn ,, Do. .. .. Shorthorn .. zvcllt92y. Ayrshire.. .. Crossbred .. Do. .. .. Do. .. .. Ayrahhe*. .I Do. .. .. Shorlhorn ,, Do. .. .. Ayrahire .. Shorthorn .. XomiHg. Ayrahiro ,, Average L a .. .. .. *. .. .. .. .. 4 . .. . I .. .. .. .. .. . * .. .. .. Mixed milks from Bame dairy Erst portions of milking, No. 10 cow 1 . *, Do. do, Total Solids. 11.8 12-26 12-75 12.17 11'8 12-3 13'0 13'94 12.7 14-61 11*66 12-29 12.17 13-36 11.88 12.9 12-9 14-08 12-82 14'35 12'784 12'7 12.63 12,5 - Last do, do, 15% Fat. 2.7 2-75 4.4 3.17 3-0 3.0 3.2 4.3 3-6 3.99 2.85 2.89 3.07 4-16 3-0 3*7 3@4 4-36 3.7 3'0 3461 3'3 3'26 2-0 6'2 - aolids not fat. 9.1 9.5 8-35 9.0 8.8 9.3 9.8 9.64 9.2 10.62 8.81 9.4 9.1 9.2 8.88 9#2 9.6 9.7 9.12 10*45 9'334 9-4 9.38 10*45 9*68 Ash 0.7 0% 0'72 0.74 0.72 0-73 0.76 0.7 0.75 0-7 0.71 0.7 0.7 0.76 0-72 0.71 0.74 0.74 0.7 0-7 0.721 0.716 0.71 0.7- 0.70 - Spedfio Gravity. 1030.7 1032.5 1028 0 1031-4 1030-2 1032% 1032-0 1032% 1031.9 1036.0 1030*2 1032.6 1030.0 1031-5 1030-2 1031-6 1033.2 1032-6 1032.0 1035.2 1031.84 1032.5 1032.5 1036.0 1031.0