HOME SEPARATION.

New Zealand Journal of Agriculture, Volume VI, Issue 5, 15 May 1913, Page 510

HOME SEPARATION.

CREAM-TESTING

G. M. VALENTINE,

Sampling

Unless a. representative sample of the cream delivered to the factory is taken, it follows that the result of the testing cannot be correct. Where cream is received daily it is an easy matter to get a correct sample by means of a sampling-tube. Provided a supplier’s cans are all of the same diameter, a. small brass tube dropped into the cream and closed at the top end with the thumb will take a proportionate sample representing the whole depth of the can. With thick sour cream, however, so much adheres to the sampling-tube that it is no improvement on an ounce dipper. Under factory conditions, as we find them in New Zealand to-day, I have seen nothing better than a vigorous use of the perforated metal plunger. A gentle stir with the sampling-dipper or a circular stir with a stick ■is quite useless, but a thorough mixing up and down will get as close to a representative sample as it is possible to obtain with that class of cream. There is a very big difference in the fat-content of the cream on top bf the can and that at the bottom, especially with thin sour cream, hence the necessity for thorough mixing. Neglect of this matter has led to the idea among suppliers that sweet cream tests lower than sour cream. It certainly does, unless the greatest care is taken to mix the sour cream so as to thoroughly break up the thick .milk at the bottom of the can and thereby get a correct sample. . A fairly proportionate sample can be taken with the dipper if a little judgment is used and any variation in the weights noted. The only possible means of getting the proportion exact would be to have, say, 5 c.c. graduations marked on the side of the sample-bottle, and take 1 c.c. for every 10 lb. of cream.

The sampling-dipper should be rinsed after handling rich cream, or the supplier following will benefit. In some exceptional cases of very rich cream or thin sour cream it may be necessary to warm or strain it in order to get a correct sample. Thin. cream requires most care, however, as there is not the difference in the top and bottom of the can with very rich cream that there is with thin cream.

Daily Tests.

If properly done, there is no doubt that daily tests are more accurate than, a composite test. An aliquot sample is just as necessary when dealing with "two- cans of cream from the same supplier as it is between a two days’ supply. Against the risk of carrying a . composite sample is to be placed the seven chances of mistake in actual testing, for there is only one with a composite sample. It is better to test once a week and .do it carefully than to test seven times and hurry over it. The cost has been the principal reason why the daily test has not been taken up more generally in New Zealand, and it is still a.disputed point whether the advantages compensate for the extra ■expense.

Care of Samples

■ Where a d&ily test is not made, samples must be kept in tightly corked battles, or the evaporation of the moisture will cause the results to be too high. The condensation of moisture round the necks of the sample - bottles on a hot day should prove to any manager the necessity for this precaution. Tin-topped bottles .are not. to be depended on, as the lids soon get. knocked about, and fit badly. . One manager claims to have increased his overrun by 2 per cent, by adopting corked bottles last year, though his methods of manufacture were the same.

Preservatives.

Formalin, bichromate of potash, and a mixture of three parts of bichromate to. one -of corrosive sublimate are . all good, preservatives, but care must be. taken not to use too much of either, which, will ■cause discoloration of the “fat” column when testing. . .

The plan of keeping the sample-bottles in crates,, and' removing them to. ah artificially cooled room between deliveries, is being adopted by one factory next season. . This should help towards keeping the ■samples in good condition. It is necessary to shake the bottles with a rotary motion after adding each sample, in order to mix the preservative. . .. •

Testing-periods.

Testing should be done weekly, but where conditions are such that this is not possible it may be done every ten days or fortnightly, but extra care will be necessary to keep the samples in good condition for the longer period.

Mixing.

Before taking the sample from the composite bottles they, should be placed in a bath and heated to. 100° to 110° and thoroughly warmed through. . Overheating will melt the fat and . harden. the curd. Pour from one bottle to another, or .. mix with a wire brush, until the -cream is smooth and free from lumps.

The Sample.

The use of a pipette for measuring cream samples into the testbottles is absolutely unreliable. The Babcock test is: based on a sample weighing 18 grammes. An 18 c.c. pipette will deliver 18 gr. of water, but cream being lighter it will deliver proportionately less. . Further, the richer the cream to be sampled the less weight will • the pipette deliver, and the test will be lower in proportion’ . ■ 'i

• To get over this; a table is sometimes used to add a percentage ;to the results obtained when testing rich cream. This might be possible if all the samples were taken at the same temperature and were ;in perfect condition, but with sour or gassy cream this is also quite unreliable. To illustrate, four samples of fresh morning’s cream were tested by weighing and by 9 c.c. pipette, the average by the scales being 0-75 of 1 per'cent, higher than by pipette, the scales used being a sensitive chemical balance. . . ■ , Scales: 51, 45, 31, 27 .. 38-50) 0-75 higher average by 9 c.c. pipette : 50, 44, 31, 26 .. 37-75 J scales. Four samples of sour cream, on the other hand, -averaged 1-5 per cent, higher by the scales than by the pipette. Scales: 39, 39, 44, 38 1 r ... , . „ -xx oo oo An on fl' s higher average by scales. . 9 c.c. pipette : 38, 38, 42, 36 ■ & J These figures serve to show that instead of sour cream testing higher, as some claim, it is considerably lower if the sample is measured with a pipette. •Another point is that the more vigorously the sample is shaken when mixing the lower will be the resulting test. This is due to the aeration of the cream. One has only to note the weight of a quantity' of whipped cream to see what effect this will have. The 18 c.c. Pipette. It naturally follows that if the 9 c.c. pipette is wrong the 18 c.c. will be further out. When using the 9 an equal quantity, of water can be used for rinsing, but with the 18 the size of the cream-bottle will not allow of this. I have found the 18 c.c. pipette to be about 1 per cent, lower than the 9 c.c. 9 c.c. : 36, 43, 31) A • A ~ , , n 18 cc • 35 42 go | Verae ’ 1 P er cent - higher by 9 c.c. pipette.

Weighing the Sample.

In justice to both, factory and supplier the sample should be weighed, and the scales used must be tin perfect order. A chemical balance is the most reliable, but is usually regarded as too slow for factory , use. The twelve-bottle scales more commonly used are accurate

if kepi; in good order, but they must be kept in a dry case when not in use, and be discarded when showing signs of sluggishness. Check, tests should be frequently made with the chemical balance. Tests which I have made by this means showed a new set to be exact, while another set two years in use, under, poor conditions, were giving results 1 per cent, too high, which would mean a reduction of about 2| per cent, in overrun. Chemical balance : 39, 47, 33, 33, 35, 45, 33, 34. Average, 37-37. 12-bottle scales: 40,49,33,35,36,45,33,36. „ 38-37. It need hardly be mentioned that whatever scale is used the greatest care must be taken to see that the weighing is done correctly. Weight of Sample. While 18 grammes is the full charge used in testing cream, equally correct results ■ are possible if 9 grammes are used and the result doubled. Extra care is necessary, however, as any error in weighing or reading the “fat” column is doubled when a 9-gramme sample is used. When an 18-gramme sample is used it is also necessary to use a 50-per-cent. testing-bottle, as the average fat-content of cream received in New Zealand factories is over 30 per cent. The 50-per-cent. bottle cannot be regarded as the best for several reasons. Tests of the same cream which I have made by both weights showed an average of 0-5 of 1 per cent, lower reading by using the heavier sample. . ■ 9 gramme: 40, 41, 39, 35 .. 38-75) Average 0-5 lower by 18 „ : 39, 40-5, 38-5, 35 .. 38-25 ) 18-gramme sample. Both tests were made in 50-per-cent. bottles, so it is probable that had the 9-gramme samples been done in 30-per-cent. bottles there would not have been even this difference. Testing-bottles. The bottles most commonly in use in New Zealand factories are the 6 in., 30 and 50 per cent. The former is much to be preferred. ■ It is graduated to 0-5 of 1 per cent., which makes closer reading possible than with the 50-per-cent., which is graduated to 1 per cent. The 30-per-cent. bottle has the further advantage that, having a narrower neck, the graduations are much farther apartsay, 6 per cent, on a 30-per-cent. bottle occupying the same space as 10 per cent, on a 50-per-cent. There is ’ also a. difficulty at times in getting a clear straight line at the bottom of the “ fat ” column when using 50-per-cent. bottles. Methods of Sampling.

The ' ordinary pipette cannot be regarded as the best means of putting the sample into the testing-bottle. The curd in the composite

sample will block up the small opening, and only the fat will be drawn into, the pipette. This can be avoided .by breaking the end off the pipette, or, when small sample-bottles are used, the cream may be poured directly into the test-bottle. ... Adding the Acid. When using a 9-gramme sample 9 c.c. of water is added, and a full'charge of acid is used. With a little experience the added water may be used to great advantage in getting a good clear reading. By its use ■ the temperature can be regulated to suit the ■ strength of the acid. If an 18-gramme sample is used, less acid is required, as there is less material for it to work upon than with milk. Personally, I have always been able to get a much clearer “fat” column when using a 9-gramme sample than with an 18-gramme. This; is no doubt due to the fact that in an 18-gramme sample of 50-per-cent, cream there are 9 grammes of fat and only 9 of milk serum, while in a 9-gramme sample of the same cream with 9 c.c. of water added there are 4| grammes of fat and 13| of water and serum. Should the mixture appear too dark after shaking; the acid and cream together, a little chilled water added will sometimes check the action of the acid. . . I have found it a good plan to shake the bottles again after the first whirling, as it helps to clear the fat from any dark specks. Bor. the. same reason I prefer to whirl three times instead of twice, as is ; sometimes done. Reading the Tests. ' If the water added to the test-bottles has been up- to 200, the samples will be about 140° in a turbine machine when finished. They must be at once removed. to a water bath at this temperature, the water reaching to about the top of the “ fat ” column. , There .is perhaps no point in cream-testing about which there is a greater difference of opinion than the reading of the “ fat ” column. Some managers read the extreme, points, the same as in milk-testing, some to the middle of the meniscus or curve, and others again to the bottom of the meniscus, or, rather, to where they think it is. Others read to extreme points and deduct 1 per cent, when using 30-per-cent. bottles, and 2 per cent, when using 50-per-cent. bottles, with a 9gramme sample. This last plan comes very close to being correct, but is not reliable.

: r The depth of the meniscus depends on the diameter of-the neck of the test-bottle, and ranges from 0-5 in a small 30-per-cent. bottle to 1 and even 2 per cent, in a very wide 50-per-cent. bottle'."

Fat-saturated Alcohol.

By the use of fat-saturated alcohol the meniscus can be done away with. -Alcohol, being lighter .than fat, floats on the top of the column, leaving a clearly defined flat line between the two.

As alcohol has a solvent action on butter-fat, however, it is necessary to saturate it with fat before using. This is done by dissolving .about a small teaspoonful of unsalted butter in an 8 oz. bottle of alcohol. After adding the butter, warm and shake the bottle until no more fat will dissolve, and leave it to settle before using. Drop enough to fill a space of about 2 per cent, into the neck of each test-bottle as it is taken from the bath, and read off immediately. The alcohol will at once rise to the surface of the fat, and the meniscus will appear to float up to the surface of the alcohol. Should the line be irregular, give the bottle a shake, or add another drop of alcohol.

The best results seem to be obtained when the alcohol is prepared fresh each time it is required, and it should be warm when used. If cold, it has an inclination to form a whitish layer of cold fat on top of the column. Care must be taken not to get any undissolved fat when taking the alcohol, from the bottle.

Result of using Alcohol.

The difference in reading the “ fat ” column to extreme points and reading with the alcohol added varies according to the diameter of the neck of the test-bottle, but it is always lower when alcohol is used. At the same, time, it does not seem to vary with any regularity, in all 30- or -all 50-per-cent. bottles, so ; • that- it is not possible to make any rule for reading tests which will avoid, the necessity of using the alcohol. Thirty-per-cent, bottles will vary from 0-5 to 1-5 per cent., using a 9-gramme sample, the average difference of those samples of which I have kept a record being 0-94 of 1 per cent, lower.

Using 50-per-cent. bottles and a 9-gramme sample, the result with alcohol added has been from 1 to 4 per cent, lower than when reading

to extreme points, the average being 2-11 per cent, lower. Had an 18-gramme sample been used, the difference would be only half. '

A study of these figures does not show any apparent relation

between the variation and the fat-content, a high test in some cases having the same small difference as a low one, and vice versa. The whole of these samples were held in a. warm bath until read off, and were carefully checked, by the managers ,of the respective factories at which they were done. Measurements were taken by the dividers.

It is absolutely necessary to maintain the temperature of the testbottles between 130° ' and 140°. If they are allowed to cool, the alcohol and fat will be inclined to mix, which will defeat '.the object for which the alcohol is used. It will also cause a shrinkage of the “fat.” column, and a consequent • low reading.

All the tests quoted in this paper were weighed on a sensitive chemical balance, and fat-saturated alcohol used in reading, except where otherwise mentioned. . The greatest care has been taken -to ensure accuracy. \ ■ ■ .

During the month of April the Moumahaki Experimental Farm was inspected by 230 farmers, and 180 letters of advice were despatched. Waerenga Experimental Farm was visited by 108 fruitgrowers and others. At Ruakura Faim of Instruction there were 500 visitors, 398 letters were received, and 337 letters were despatched. This farm was inspected by 704 people in March. ■

The calves of the dairy herds at Weraroa Experimental Farm are looking exceptionally well, and have exhibited no indications of scouring, a trouble experienced in former seasons. The improved condition is attributed to the use of chaff and pollard, the feeding of which was commenced when the calves were from two to three months old. This mixture was taken greedily by the calves, which were quite indifferent as to their milk ration at weaning-time. Shelter-sheds are being erected in the fields at this farm for the heifer calves during winter. 1 ' • •

■ Read, from - ■ - With Fat-saturated Extreme Points. Alcohol. •... 37 . .. ... 36 ' 43-5 .. ■ .. . ' .. - 42-5 ■ 38-5 .. .. ' .. .. . . 37-5 - 45-5 .... .. /: 44 28 .. 27-5 41 • ... .. 40 '23-5 .. .. ;. .. 23 34-5 .. .. .. .. .. 34 ’ 36-5 ... . ... ... .. ..35

Read from Extreme Points. With Alcohol. Read from WHh Extreme Alcohol. Points. ■ Read from With Extreme Alcohol. Read from Extreme Points.' With Alcohol. 52 50 ' . 43 42 . 37 35 46 44 32 30 47 45 ■ 32 31 53 51 ' 35 ' 33 28 26 33 31 ■ 42 40 i 53 51 29 27 . 50 49 '■ 47 45 47 - 45 . 36 33 33 31 43 39 37 35 . 29 27 50 47 39 36 36-5 34. 36 33 47 45 ' 38-5 36 35 33 35 33 36 35 ~ . 37 34 39 . 36