VI. MANAGEMENT OF CREAMERIES AND CHEESE FACTORIES. DIRECTIONS FOR TAKING AND PRESERVING COMPOSITE SAMPLES OF MILK IN CREAMERIES AND CHEESE-FACTORIES. (FARRington). The modern creamery and cheese-factory uses the Babcock test for determining the richness of the milk delivered by each patron. The most common and satisfactory method of paying for the milk according to its test is to take a small sample of each lot of milk each day, pour this into a covered glass jar containing a small amount of some preservative, and at the end of a week or ten days test this composite sample. The essential features of the process are given in the following directions: 1. Provide a pint or quart jar or bottle for each patron. 2. Label each bottle with a number, giving the same number to a patron on the milk-recording sheet. 3. Composite-test sample-bottles made for this purpose with a tin cover and numbered brass tag wired to the neck of each bottle can be obtained of creamery supply-firms. 4. These sample-bottles should be placed on shelves within easy reach of the man at the weigh-can, and protected from the light. 5. A small quantity of powdered potassium bichromate, corrosive sublimate, formaline, borax, or peservaline is put into each clean bottle, to keep the milk from souring until testing-day. Some of these preservatives are put up in tablet form, each tablet containing the necessary amount to use in one sample. 6. After each lot of milk is poured into the factory weighcan and weighed, a small amount of it is dipped from the can and poured into the proper sample-bottle. 7. These samples are usually taken with a small (1-oz.) tin dipper, a Scovell sampling-tube, or from a drip in the conductor-spout. 8. Each lot of milk sampled must be sweet, containing no clots, lumps of curdled milk, or small butter-granules. The sample should be taken just as soon as the milk is weighed, and while it is evenly mixed. 9. The use of a small (1-oz.) tin dipper for taking the composite sample has been proved to be practically correct. As the quantities of milk delivered from day to day by each patron vary but little, the error introduced by taking the same amount of milk for each sample is too small to be worth considering in factory work, and this method of composite sampling is usually adopted in separator creameries and in cheese-factories, where the payment of the milk is based on its quality. 10. When it is desired to vary the size of the samples according to the quantity of milk delivered each day by a patron, it is necessary to use a "milk-thief" or a Scovell sampling-tube. In using either of these tubes, the size of the sample is regulated by the amount of milk in the weighcan. In all cascs cylindrical sampling-cans must be used. II. Continue adding a sample of each patron's milk to his particular jar every time he delivers milk, for a week or ten days; then test this composite sample. 12. The composite sample-jars should be kept covered, to prevent loss by evaporation, and in a cool, dark place. Every time a new portion of milk is added to the jar it should be given a horizontal rotary motion to mix the cream already formed in the jar with the milk, and to rinse off the cream sticking to its side. Unless this is done every time fresh portions of milk are added to the jar the cream on the milk becomes lumpy and sticks in patches to the side of the jar, thus making it nearly impossible to evenly distribute this cream through the entire sample. 13. Composite samples having patches of dried cream on the inside of the jar are the result of carelessness or ignorance on the part of the operator. 14. A test of the composite sample takes the place of the daily tests of each lot of milk and gives accurate informa |