MILK TESTING.

New Zealand Mail, Issue 1196, 1 February 1895, Page 7

MILK TESTING.

THE OPINION OF A SCIENTIST. There has been a good deal of discussion of late in the columns of the newspapers in various parts of the colony on the value of the milk tests in use, and the opinion has been expressed by several suppliers and factory managers that the methods employed cannot be depended upon to give, an accurate estimate of the amount of butter fat contained in the milk. In order to throw some light on this important question a representative of the New Zealand Mail called on Mr C. Hulke, head master of the Newtown State school, a few days ago, and invited that gentleman to give his views on the matter. Mr Hulke’s reputation as an experimental scientist is a guarantee that his opinions are worthy of very great consideration, both by suppliers and factory proprietors, especially as he has taken a very deep interest in milk testing for some time ; in fact samples from miles outside Wellington are brought daily to his laboratory to be tested. He possesses a large stock of literature on the subject, and has two testing machines—a Babcock and a Leffmann-Bean—inconstant use, so that his views are not theoretical, but the result of practical experiments. Before giving his views Mr Hulke made several tests by both machines and lucidly explained how the results were obtained, and why discrepancies occurred between the percentage of butter fat by the testing machine and the output afterwards from the factory.

In answer to a question by our representative a 3 to whether he had seen any correspondence in the press on the subject of milk testing, Mr Hulke replied that he had, and that according to a letter published in the Auckland papers by the manager of the New Zealand Dairy Association, Aucklaud, to the Pukekohe farmers, the Babcock testing appliances did not indicate properly the value of the milk in butter fat. This writer had complained that the Babcock machine gave the suppliers a greater average of butter fat than he had been able to collect,, Having paid the suppliers on the Babcock test, he had during the month of October sustained a serious loss. He said that his test indicated 3*32 per cent of butter fat, whilst the butter returns accounted onlv for 3-16 per cent. Before criticising, said Mr Hulke, the statements made in the letter, the best thing to do is to understand what the Babcock test is. In 1888, Werner Schmidt, a German chemist, introduced a process for extracting butter fat from milk by means of ether, after having dissolved the caseine by hot hydro-chloric acid. In the followin o year Dr Babcock, an American, followed suit, by using sulphuric acid, and instead of extracting the fat by ether, employed a centrifugal apparatus for its separation. This is the principle of what is known as the Babcock process. The milk is placed in a bootle furnished with a graduated neck, sulphuric acid is then added, and this dissolves the caseine, the fat being set free. In order that the process may be used quantitatively a definite measure of milk is placed in the bottle. In this case it is 17£ c.c. (cubic centimetres). The normal specific gravity of milk being on an average I*o3o, the 17+ c.c. will just weigh 18 grammes. If we take one tenth of 18 grammes we have I*B, and as the resulting butter fat will be measured in the neck of the bottle at a temperature of about 55 deg. centigrade, the specific gravity of the fat will be about *9. Now if 1”8 be divided by *9 it will give 2, which is the measure in cubic centimetres on 10 per cent by weight of fat in the quantity of milk taken. The length in the neck of the bottle of the volume of the 2 c.c. is therefore marked off and graduated in ten main divisions, each of these representing one per cent of butter fat. These divisions are again sub-

divided into fifths, and therefore each subdivision represents 2-10th per cent. Let it be then supposed that the sulphuric acid has been put in the milk, and the whole subjected to the action of the centrifugal machine, we have the bottle, with the butter fat collected in the neck ready for reading. The base of the column of butter fat will be flat, while the top of the column will be concave. If we take some salad oil and put it into a narrow tube, closed at the bottom, and then pour water into the tube, the oil will rise to the top of the water, and it will be found that the base of the oil is not flat, but curved similarly to the top of the column. To calculate the length of the column of oil, we would simply read from the lowest point at the base to the lowest of the concavity. But in the case of the butter fat, the bottom of the column being flat, in order to arrive at the correct quantity we must read from that flat line, not to the bottom of the concavity at the top of the column, nor to the top edge of the flat, but half way between the two point. It is here, Mr Hulke thinks, where the reading of the Babcock test is improperly done. The proper reading is, Mr Hulke considers, a matter of vital importance. He pointed out that a circular which had been sent out with some of these apparatus, directed that the higher reading be taken at the very top of the column of fat, and where such is done the calculations are very misleading and the returns would show a loss of 2|lb of butter on every 10001 b of milk. This wrong reading would doubtless account in many cases for the discrepancies between the indications and the real returns. Mr Hulke said he would assume that the printed directions used as a guide in the Auckland district were diffeient from those in his possession, and that the concavity at the top of the column of butter fat had been properly allowed for. If such were the case the facts as given in the letter quoted indicated a great laxity in the management, for surely no factory manager would allow a whole month to elapse without having taken steps to satisfy himself that the separators were doing their work properly. There are three things, said Mr Hulke, to be attended to in the management of the separator. In the first place there is what may be termed the “ constant" of the machine, that is the proper speed at which each is to be driven so as to insure the best results ; secondly, the temperature at which the milk should be put through; thirdly, the rate at which the machine should be fed, and it is here that a very great mistake is often made. There is a great tendency to feed too fast. The machine may be guaranteed to put through 300 gallons of milk per hour, but because the machine is guaranteed to do this it does not follow that in every case the best results will be obtained. The milk that is passed through the separator requires to be tested just as much as the milk that was put in, and this should be done daily. By taking tests of the milk after passing through the machine it is easy to see whether the separator is doing its work efficiently or otherwise. Where it is found that the separation is not being properly effected the feeding would have to be slower, or have to be regulated, as incomplete separation denotes that the milk is being put through too rapidly. Another point which tends to produce discrepancies in the returns is the fact that in many instances the milk supplied is not tested daily, for when the tests are made not oftener than once a week, ic is impossible to get anything like a fair average. Where the factory does its own creaming the testing should be done usually every afternoon ; but where, as in the case of large factories, the milk is brought from distant creameries, a far better plan is for the person in charge of the separator to have a bottle, say an eight ounce one, for each supplier, into which he puts a sample of that supplier’s milk daily, and at the end of the week the bottle would contain an average sample of the milk supplied by that person. To prevent the milk from curdling, and to keep it so fluid that there would be no tendency to form cream all that is necessary is to put into the bottle before the first sample of milk, a small quantity say eight or ten grains, of bichromate of potash, and shake the bottle when putting in a fresh sample. This plan of keeping milk for analysis was introduced in 1892 by Dr Alen, of Gothenburg. It is only in this way that a fair average can be struck. Before we can speak of discrepancies in the returns it is necessary that the column of fat be correctly estimated, the separator milk daily tested, and the average weekly sample from each supplier analysed. Mr Hulke considers that no better machine could be devised for the purpose of testing separated milk than the centrifugal apparatus and process invented by Drs Leffmann and Bean, of Philadelphia, which is a perfect jewel. He has had a sample four bottle machine at work in his laboratory for some time, and the testing of four samples doss not take longer than four minutes. No hot water, such as is necessary with the Babcock, is used a mixture of amylic alcohol and hydrochloric acid being employed to assist the sulphuric acid in setting the butter fat free. In this machine the whole of the butter fat is set free, where as in the Babcock, at least those in general use here 3-10th per cent is retained. It is possible to obtain the whole of the fat by mean* of the Babcock process, but then very strong acid must be employed, and the machine should be driven at a speed of

not less than 1700 revolutions per minute, a rate which would require a much more strongly constructed apparatus than that in use in this Colony. Mr Hulke contended that the Leffmann-Bean machine is such a compact, clean, and efficiently working apparatus that no. one having once used it would be satisfied with the results obtained from the Babcock.

In conclusion he pointed out that the only way to remove the distrust which the milk suppliers appear to entertain with reference to the returns furnished by the factory proprietors, is to carry out each and every process connected with the milk testing as accurately as possible, not trusting to what is said in circulars or books, but to examine for themselves and follow out the plan pursued by one large factory in Auckland where the suppliers are invited to see for themselves how the testing is conducted. >