Sidebilder
PDF
ePub

unnecessary, as one of the riders was found to differ very seriously from its supposed weight. Very recently I was struck with certain differences in the determinations of densities of butter, lard, and other fats at the boiling point of water when made by the plummet and by the specific gravity bottle or Sprengel-tube, and for some time was wholly at a loss to account for the discrepancy, which, however, has now been definitely traced to the faulty division of the beam of the Westphal balance. The instrument I used was one purchased of a London firm that has supplied a number of instruments in this country. Mr. W. Chattaway, who discovered the nature of the error, has examined all those in stock at present, and finds the error common to all, and therefore very probably it exists in all or most similar instruments in use in England. The fault consists in the notch at the eighth division of the beam being sensibly nearer the fulcrum than it should be. The effect of this is evident. All weights placed on the eighth division do not exert their full leverage, and additional weights have to be added to effect the counterpoise. In the case of my instrument, the error can be readily detected with a fine pair of compasses, and is even visible to the unassisted eye of some observers when their attention is directed to the point. The consequence is that the greater number of the determinations of gravity, commencing with 8, including all determinations of specific gravities of butter, lard, and other fats at 100° made and published by me up to a very recent date, have been in excess of the truth by about 0015. Hence the statement in my book that the average density of butter at 99° is 0.868, should be modified to 0.8665 = 0.8658 at 100°. The discovery of the error in question would have been made long ago had the balance been checked against the specific gravity-bottle with rectified spirit instead of proof spirit and milk. The history of it I have given will serve as an illustration of the ease with which an error may creep into absolute determinations, and hence how difficult it is for chemists to compare strictly their own results with those obtained in other laboratories. Of course all my results are strictly comparable among themselves, so that the error has not affected any conclusion arrived at; but I think it right to place the fact on record, as I believe other chemists are likely to be obtaining erroneous figures from exactly the same cause. One other practical conclusion may be deduced from my experience, and that is, that the rider system is open to grave objections when applied to the larger weights of a balance.

With regard to the determination of the saponification equivalent and of the soluble and insoluble fatty acids, I have nothing to say, except that the highest estimations of insoluble acids were obtained in experiments in which the washing was known to be carried out less perfectly than in the other cases.

The turbidity-temperatures by the acetic acid test were made by heating 3 c.c. of the melted butter-fat with its own measure of glacial acetic acid until the mixture was clear, and then stirring with a thermometer until the liquid became turbid throughout. The test-tube in which the experiment is conducted should be immersed in an empty flask, so as to avoid too sudden cooling and the effects of currents of air. I have now used this test for several years with very satisfactory results. It enables butter to be distinguished from other fats with the utmost facility, and in the case of samples of genuine butter, the temperature at which turbidity occurs closely follows the other characters of the butter; that is to say, a butter rich in lower fatty acids, and therefore giving a high result by the Reichert-Wollny test, will give a low turbidity-temperature by the acetic acid test. Repeated experiments on the same samples give results which are concordant within 1o C. The strength of the acetic acid used affects the test in an enormous degree; in fact, I fancy the solubility of a definite fat in glacial acetic acid would be the best possible test for the strength of such acid. A sample of acid which when used for testing genuine butter will show a turbidity-temperature of 60° C. will show with butterine a turbidity-temperature of 100°, but by putting a very small

proportion of water to such acid the temperature of turbidity is so raised that the results of the butter-test will be like those given in the table. Hence, in order that the results may be comparable, it is necessary to define in the strictest manner the strength of acid used, and this is somewhat difficult to do. The acid I employed for the experiments quoted above had a specific gravity of 10563 at 15.5o C., and when titrated with standard baryta water and phenol-phthalein, was found to contain 95.14 per cent. of real acetic acid (C2H,O2).

From a careful consideration of the foregoing experiments, it appears that the butter in question marked "B" had a density quite outside previous experience of genuine butter-fat, and by analysts relying much or exclusively on this indication would be very liable to be condemned as adulterated. The volume of alkali required by the butter when examined by the Reichert-Wollny process was sensibly below the limit hitherto recognised. The saponification-equivalent was at or beyond the extreme limit hitherto observed (253). The proportion of soluble fatty acids was lower than has been observed in any previous genuine butter, except some analyses of doubtful accuracy. The insoluble fatty acids are sensibly in excess of those previously observed; the highest figure recorded by Dr. James Bell in a butter not altered by keeping being 89.9 per cent. The foregoing remarks apply, though somewhat less strongly, to the sample marked "O." Before leaving this part of the subject, I may say that I have on the table a further quantity of the butters in question contained in the original sealed packets. I have permission to place them at the disposal of any Public Analyst who will undertake to examine them fully and communicate his results to me.

Although the Danish butters which yielded me the abnormal results already described are of exceptional character, we must remember that they really represent the butter produced from the mixed milk of a large number of cows, which were yielding much the same kind of butter a year previously, and in the face of these facts, and of the Swedish experience already recorded, it cannot be said that such butter may not be met with in practice.

My attention has been called by Mr. Faber to some experiments made by Adolph. Meyer, published in the "Milch-Zeitung" for October 17. These relate to the butter produced by a particular cow which, according to Meyer, yielded milk the butter-fat of which had the following characters.-

[blocks in formation]

Meyer states, however, in a foot-note, that the specific gravity figures are merely comparative, not absolute; so that the results have less value than would otherwise have been the case.

Nilson, of Stockholm, again, in experiments made in 1896 by the unmodified Reichert process, found the volatile acids to vary materially with the length of time since calving. While very low at first, requiring only about 10 c.c. (+2=20 for 5 grms.) of alkali for their neutralization, they rose to a maximum in about a week, and maintained this high figure for about four weeks, after which they fell gradually to about 12 (24), but only very rarely below that figure.

The produce from single cows may, of course, be expected to vary still more than that from a number of animals, but in practice there must be very little butter sold

which is the produce of only one or two animals, and the milk of a normal cow being usually much larger in quantity than that of an animal yielding an abnormal product, the variations from the mean composition in the case of united milk is not likely to be great.

A very great object would be gained, if we could trace a sample of butter back to the farm and cows which produced the milk, and I think we should look forward to the day when an alteration of the law will render this possible.

Although the results I have recorded undoubtedly show that butter-fat varies in composition more than was supposed, and will render the detection of adulteration decidedly more difficult in cases where a moderate percentage of foreign fat has been added, I do not think we have yet come to the end of our powers in butter analysis. The detection of cocoanut oil, which I met with in one sample of butter in considerable amount, can be effected with certainty, by the large yield of solid volatile acids and their peculiar odour. Besides, the specific gravity of the butter is rather increased than diminished by the presence of cocoa-nut oil. Cotton-seed oil will indicate its presence in more than one way. Seeing that all margarine now-a-days contains cotton-seed or other vegetable oil, it is worth while to try and detect their presence. The isolation of phytostearin, or vegetable cholesterin, might possibly afford a test for vegetable oils, as suggested by Mr. Hehner with respect to lard. Some modification of the chloride of sulphur test may be found available, while cotton-seed and most vegetable oils might be indicated by the increased iodine-absorption of the sample or of the oil separated from it by pressure. In such expressed oil, earthnut oil might possibly be recognised by isolating arachidic acid, while sesame oil can be detected by the sugar reaction even in the original sample. The chemistry of the vegetable oils is still imperfectly understood, and researches in this direction would almost certainly lead to the discovery of new methods of detecting them. Hence, we ought to aim at a further advance in our methods of examining butter, rather than remain content with processes which, however good and ingenious in their way, do not give us all the information we require.

Professor Wollny, of Kiel, has requested nearly thirty chemists to assist him in ascertaining the natural variations in analytical characters of butter-fat, in order to render the results more strictly comparable. He has published and circulated a printed description of the exact method to be employed in the analysis, together with forms on which the results are to be tabulated. To give an idea of the field proposed to be covered by the scheme, I may quote the following data with which Dr. Wollny requests each analyst to furnish him :-The strength of the sulphuric acid with which the baryta solution is set. The number of c.c. of baryta used for neutralisation. The strength deduced from this result, and the strength of the baryta solution deduced from its titration by acid oxalate of potassium. In estimating the volatile acids, the result of a saponification with and without alcohol. The time of each distillation, and blank analyses in each case. The mean molecular weight of the volatile fatty acids (distilled from their magnesium and copper salts). The volume of baryta solution neutralised by the acids soluble in boiling water and in 10 per cent. alcohol. The percentage of insoluble acids, by washing on the filter and by washing in a current of steam; and their mean molecular weight. The percentage of free fatty acids in the butter-fat. The saponification-equivalent of the fat. The molecular weight of the total fatty acids. The iodine absorption. The refraction index of the butter-fat; and, for comparison, of olive oil, nitrobenzene and alpha-monobrom-naphthalene. The special gravity at 0° or 15° and at the temperature of boiling water, by the specific gravity bottle, plummet method, and hydrometer. These determinations, with the melting point and solidifying point of the fat, seem a tolerably complete examination. It would be impossible, even were it desirable, for our Society to institute any series of experiments of the same wide

reaching character; but it occurs to me that we might with advantage appoint a small committee to receive and tabulate results which might be communicated to them by members of the Society, and to devise, if possible, some method of examining butter which should be less dependent on the natural variations in its composition than the methods at present in us?.

In conclusion, I desire to express my indebtedness to Mr. Faber for his good offices as interpreter and travelling-companion, and to Mr. Wm. Chattaway for the zeal and care with which he has executed many of the determinations, the results of which I have recorded.

DISCUSSION.

DR. VIETH (who had temporarily taken the chair) said they would all agree with him that the question Mr. Allen had brought forward was of the greatest importance to public analysts. Personally, he was very pleased to see that Mr. Allen had not only taken the trouble to see the butter made, but had seen the cows milked and the milk locked up, so that no hitch could occur similar to that which happened in the case of the Swedish dairy in which the butter was produced, which was the cause of the investigations dealt with in Mr. Allen's paper. In this case the Government's officers only saw the cream churned, and had taken no precaution to put the purity of the cream beyond doubt. Next to milk, butter had given them perhaps the greatest amount of trouble lately.

As regards the requirements of the German Commission which had been mentioned, he should like to add that, not only one, but five determinations of each kind were to be given, and this, of course, caused a great sacrifice of time. Had Mr. Allen made any experiments as far as regarded the distilling off of the volatile fatty acids? Wollny and Stein laid much stress on the fact that distillation should not be proceeded with too quickly, but that the fatty acids should be melted first, and the aqueous liquids allowed to become clear. What difference would it make if they proceeded a little more quickly, and did not wait for the clearance of the aqueous liquid? As far as his experience went he did not think there was much difference. He had found that the butter sent to the members of the German Commission, which was produced at Kiel in July, required about 30 c.c. alkali-Reichert-Wollny-while a number of samples of French and Danish butters, examined during the last weeks, required from 25 to 26 c.c.

N

10

Mr. HARALD FABER said-It is not to be denied that the proved existence of genuine butter of such an abnormal composition as those mentioned by our president interferes very seriously with the detection of mixtures of butter with foreign fat, and it might therefore seem peculiar at first that an exertion has been made on behalf of the butter producers in Denmark, being very far from interested in mixtures, to put such a check on the control of butter. But I think the position of the Royal Agricul tural Society of Denmark to this matter is very clear. A recent case, which Mr. Allen mentioned, proved what risk we run by leaving matters as they were. I wish to point out that we have in no way questioned the correctness of the English analyses of the butter sampled in Wigan and condemned as adulterated; but having had the opportunity of examining samples from the dairy from which the butter in question originated, and some of the butter fat rendered from the official duplicate sample, and having learned already last year that genuine butter can be produced of still more abnormal composition than the butter in question, we expressed the opinion that the fact of this butter having such a composition was not sufficient proof of its being adulterated. But our assertion that genuine butter might be of such a quality that English analysts would call it adulterated was, perhaps, not likely to be taken notice of without further proof. Knowing that some day genuine Danish

butter might be condemned here as adulterated if no step was taken, two courses were open to us - either to leave matters as they were, which meant running the risk of losing the well-merited name as producers of first-class and genuine butter, or to prove the existence of the abnormal butter, and then, perhaps, have to compete still more with mixtures from other countries, which would then be more difficult to detect. I urged the adoption of the latter course, which was finally agreed to, and we have all reason to be well satisfied that Mr. Allen, at our invitation, undertook the long journey to Denmark, and also with the excellent work he has done. The way in which these abnormal butters were first found is as follows:-As it has been said, Prof. Stein is the Public Analyst for the whole of Denmark, which is equivalent to about half of London as regards population. He received in 1887 some samples of butter of which he at first felt convinced that they were mixtures. He therefore ascertained the names of the farms from where they originated, and he had samples taken with all guarantee of purity, and in that way found that in September, 1887, some large farms in Denmark, with large herds of milch cows, gave butter which yielded below 25 c.c., by Reichert-Wollny, one even as low as 217, another 21.9. His experience in butter analysis is now, as mentioned by Mr. Allen, very extended, having three samples a day from our inspectors under the Margarine Act. Whenever a sample of doubtful character has been taken at a dairy, a warehouse, or a shop, the inspector is sent at once to the dairy where the butter was produced, to obtain an authentic sample. In this way it has been found that butter of this abnormal character may occur at different farms, but, as it seems, for a very limited period only. The lowest figure by the Reichert-Wollny test yet found in butter-fat from a herd is, as far as I know, 21 c.c. It is only this extended experience of Prof. Stein's that enabled him to pick out a farm where we could invite Mr. Allen to go and take samples, and it was only done with great difficulty. Some of the farms where the butter had been abnormal in September, 1887, gave October of this year butter in every respect normal, others gave abnormal butter when first tested, but when the test was repeated a week later, the butter was good again. It was very much of a chance that we hit on a farm yielding butter of a description like the sample "B."

I have been informed by my friend Dr. O. Hehner, that Prof. Besana, in Italy, has proved the existence in Italy of genuine butter of the same abnormal condition. He found amongst 114 samples 2 yielding between 21.8 and 22 c.c. by the Reichert-Wollny process, 1 yielding between 23 and 24, 2 between 24 and 25. It may, therefore, reasonably be expected that such abnormal butter may occasionally be produced even in other countries. It seems happily to be rare exceptions, and this in a double meaning. Not only are such butters met with only on a few farms, but they are found there only at certain times. The reason of this peculiar quality of the butter cannot be given yet; but two circumstances at least seem to have some effect on the percentage of volatile acids. One is, the condition of the field or meadow, another, the distance from the time of calving. Very poor grass land may have the effect, but grass on reclaimed land, although it could by no means be called poor, neither the land nor the grass, and although the grass was plentiful, have in some cases been considered the cause of the low contents of volatile acids in the butter. As far back as 1886, Prof. Nilson, of Stockholm, experimenting on a large number of single cows, found that the contents of volatile acids varied with the distance from the time of calving; on the first days the contents was very low indeed10 c.c. by the old Reichert test; in about a week it reached a maximum, where it remained for a few weeks, when it gradually fell towards the end of the time of lactation. In this connection it will be of some interest to learn that the cows on farm B which gave the most abnormal butter had, until some days before the samples were taken, been on grass on reclaimed land of good quality. They were most of them in calf, but with a variation

« ForrigeFortsett »