In thus declaring the action of bacteria to be purely chemical and analogous to that of well-known inorganic poisons, I would particularly point out that this very justifiable statement is a pure hypothesis; it is an excellent illustration of the fact that we cannot get on in the explanation of the most important natural phenomena without hypotheses. We can see nothing whatever of the chemical molecular structure of the plasm, even under the highest power of the microscope; it lies far below the limit of microscopic perception. Nevertheless, no expert scientist has the slightest doubt of its existence, or that the complicated movements of the sensitive atoms and the molecules and groups of molecules they make up are the causes of the vast changes which these tiny organisms effect in the tissues of the human and the higher animal body. Moreover, the distinction of the many species of bacteria is of interest in connection with the general question of the nature and constancy of a species. Whereas formerly in biological classification only definite morphological characters, or definable differences in outer form or inner structure, were regarded as of any moment in the distinction of species, here, in view of the vagueness or total lack of these characters, we have to look mainly to the physiological properties, and these are based on the chemical differences in their hypothetical molecular structure. But even these are not absolutely constant; on the contrary, many bacteria lose their specific qualities by progressive culture under changed food-conditions. By a change in the temperature and the nutritive field in which a number of poisonous bacteria have been reared, or by the action of certain chemicals, not only the growth and multiplication are altered, but also the injurious effect they have on other organisms by the generation of poisons. This poisonous effect is weakened, and--what is most important--the weakening is transmitted by heredity to the following generations. On this is based the familiar process of inoculation, an admirable example of the inheritance of acquired characteristics.