In order to avoid the disadvantages of the continuous machines, viz., lack of control, an apparatus has recently been devised which can handle large quantities of milk, heating the same to any temperature for any desired time. In such a machine the milk is first heated in a continuous heater, and is then passed into large tanks in which it is allowed to remain for the desired time, and from which it flows over the coolers. Such an apparatus is called a "holding" machine, and is probably the most feasible type of pasteurizer now on the market, when all factors are considered. In some of the continuous machines, an attempt is made to accomplish the same result, by building the machine so that the milk requires fifteen to twenty minutes for passage through the machine, but in all such cases the same disadvantage of variation in rate of flow, as in other continuous flow type of machines obtains. =Tests of pasteurizing machines.= It is possible for the operator to test the rate of flow in a machine, so as to determine whether all of the milk is heated for a uniform time. This is done most easily in the following manner: The machine is first filled with water, heating the same to the desired temperature, and regulating the rate of flow as it would be if milk was used. The flow of water is then turned off, and a stream of milk containing a known per cent of fat admitted to the machine. The time elapsing between the admission of milk to the machine, and that at which the first sign of turbidity is noted at the outlet, will be the minimum period necessary for any portion of the milk to flow through the machine. At frequent intervals thereafter, samples of the outflowing liquid may be collected, noting the time at which each sample is taken. The percentage of fat in the various samples is determined by the Babcock test; at the moment when all of the water has been removed, the sample taken will show the same fat content as the milk used. The samples taken previous to this will show a lower fat test, dependent upon the relative amount of water and milk. In this manner, the minimum, the maximum, and the average period of exposure of milk in the machine tested, can be determined with exactness.