merican
pasteurizer, the milk is forced in a thin sheet between two heated
surfaces, thereby facilitating the heating process. In the Farrington
machine heated discs rotate in a reservoir through which the milk flows
in a continuous stream.

One of the most economical types of apparatus is the regenerator type (a
German machine), in which the milk passes over the heating surface in a
thin stream and then is carried back over the incoming cold milk so
that the heated liquid is partially cooled by the inflowing fresh milk.
In machines of this class it requires very much less steam to heat up
the milk than in those in which the cold milk is heated wholly by the
hot water.

A number of machines have been constructed on the principle of a
reservoir which is fed by a constantly flowing stream. In some kinds of
apparatus of this type no attempt is made to prevent the mixing of the
recently introduced milk with that which has been partially heated. The
pattern for this reservoir type is Fjord's heater, in which the milk is
stirred by a stirrer. This apparatus was originally designed as a heater
for milk before separation, but it has since been materially modified so
that it is better adapted to the purposes of pasteurization. Reid was
the first to introduce this type of machine into America.

~Objections to continuous flow pasteurizers.~ In all continuous flow
pasteurizers certain defects are more or less evident. While they
fulfill the important requirement of large capacity, an absolute
essential where large volumes of milk are being handled, it does not of
necessity follow that they conform to all the hygienic and physical
requirements that should be kept in mind. The greatest difficulty is the
shortened period of exposure. The period which the milk is actually
heated is often not more than a minute or so. Another serious defect is
the inability to heat _all_ of the milk for a uniform period of time. At
best, the milk is exposed for an extremely short time, but even then
portions pass through the machine much more quickly than do the
remainder. Those portions in contact with the walls of the apparatus are
retarded by friction and are materially delayed in their passage, while
the particles in the center of the stream, however thin, flow through
in the least possible time.

The following simple method enables the factory operator to test the
period of exposure in the machine: Start the machine full of water, and
after th