178 PATHOGENIC BACTERIA. planted. It is always well to use a considerable number of tubes. Bouillon is generally employed. If the anti- septic is nonvolatile, it may be added before sterilization, which is to be preferred; but if it is volatile, it must be added by means of a sterile pipette, with the greatest precaution as regards asepsis, immediately before the test is to be made. Control-experimentsi. e. without the addition of the antisepticshould always be made. The results of antiseptic action are two: retardation of growth and complete inhibition of growth. As the tubes used for the study of the antiseptic are watched in their development, it will usually be noticed that those con- taining very small quantities develop almost as rapidly as the control-tubes; those containing more, a little more slowly; those containing still more, very slowly, until at last there comes at time when the growth is not deferred, but prevented. Sternberg points out that certain circumstances may .modify the results obtained. They are: 1. -The composition of the nutrient media, with which the antiseptic may be incompatible. 2. The nature of the test-organism, no two organisms being exactly alike in their susceptibility. 3. The temperature at which the experiment is con- ducted, a relatively greater amount of the antiseptic being necessary at temperatures favorable to the organ- ism than at temperatures unfavorable. 4. The presence of spores which are always more resistant than the asporogenous forms. II. The Germicidal Vahte. Koch's original method of doing this was to dry the micro-organisms upon sterile shreds of linen or silk, and then soak them for varying lengths of time in the germicidal solution. After the bath in the reagent the threads were washed in clean, sterile water and then transferred to fresh culture- media, and their growth or failure to grow observed. It will be observed that this method is aimed at the deter- mination of the time in which a certain solution will kill.