This is the temperature at which contact between the germicide and the test germ takes place, and is of importance, since some germicides (_e. g._, Phenol) appear to be more powerful at high temperatures. 18° C.--practically the ordinary room temperature--is a temperature at which the multiplication of B. coli is a comparatively slow process, but variation of a degree above this temperature or of two or three degrees below is of no moment. If the room temperature is below 15° C. when the experiments are in progress, arrange a water-bath regulated at 18° C. for the reception of the tubes containing the mixture of germ and germicide; if above 19° C. immerse the tubes in cold water, to which small pieces of ice are added from time to time to prevent the temperature rising above 18° C. _c. Relative Proportional Bulk of Test Germ and Germicide_, ~50:1.~-- Five cubic centimetres is a convenient amount of germicidal solution to employ, and to this 0.1 c.c. of the emulsion of test germ should be added. _d. Bulk of Sample Removed from Germ + Germicide Mixture at Each of the Time Periods_, ~0.1 c.c.~-- This is sufficient to afford a fair sample of the germ content of the mixture, and at the same time is insufficient to exert any inhibitory action when transferred to the subculture medium. _e. Subculture Medium._ ~Bile Salt Broth.~-- A _fluid_ medium is essential in order to obtain immediate dilution of the germicide carried over; at the same time it is advantageous to employ a selective medium which favours the growth of the test germ to the exclusion of organisms likely to contaminate the preparation, and if possible one which affords characteristic cultural appearances. Bile Salt Broth (page 180) combines these desiderata; it permits only the growth of intestinal bacteria, whilst the formation of an acid reaction and the production of gas in subcultures prepared from the germ-germicide mixture is fairly complete evidence of the presence of living B. coli. The amount of medium present in each test-tube is a matter of importance, since the medium not only provides pabulum for the test germ, but also acts as a diluent to the germicide, to reduce its strength below its inhibition coefficient. For routine work each subculture tube contains 10 c.c. of medium, but it is obvious that if germicide-x possesses an inhibition coefficient of 0.1 per cent. the addition of 0.1 c.c. of a 10 per cent. solution to 10 c.c. of medium would effectually prevent the subsequent growth of the test germ after a contact period insufficient to destroy its vitality. Hence the preliminary tests may in some instances indicate the necessity for the presence of 12 c.c., 15 c.c. or more of the fluid medium in the culture tubes. _f. Incubation Temperature_, ~37° C.~-- _g. Observation Period of the Subcultivations_, ~Seven Days.~--