If we glance again at the classification proposed by Andrewes and Horder we are struck at once by the fact that no hard and fast differences in fermentative properties characterize the various groups, and that the authors prefer to have recourse to the most frequent habitat of the organisms and to chemical tests, rather than to confine themselves to the latter method alone. From what has been said earlier in this paper, it becomes clear at once that the results gained by these authors are not convincing, for in their work they have failed to use the most favorable medium for growth of the organisms. And thus it can hardly be doubted that different reactions might have been obtained, under the conditions adopted by us. How otherwise could the fact be explained that of thirty-four of their pneumococci only eight fermented inulin. We had a similar experience while using the plain broth which led us to reject it as an unfavorable medium. Further, all our pneumococci fermented lactose with great rapidity. Andrewes and Horder report that eight of their series left lactose unchanged.
Taking our own tests into consideration we find that working with but a small number of carbohydrates we were able to find six different varieties of streptococci among only 33 pathogenic strains. Which of the various substances is to decide us in the grouping of the organisms? Would we not find even greater variations from the most common type (see No. 4 in table) if we were to extend our tests over a larger series of chemical agents? These questions are difficult to answer and only extended experimentation with a great many streptococci and many media will clear up the doubtful points.
In concluding it may be stated:
1. Streptococci vary considerably in their ability to produce acid from various carbohydrates.
2. Chemical tests of this kind should be made only in the media which are most favorable for the growth of the organisms.
3. Our results gave us six groups of streptococci, when tested upon dextrose, levulose, galactose, maltose, saccharose, lactose, inulin, dextrin and mannite, viz.: Those fermenting (1) all; (2) all but mannite; (3) all but inulin; (4) all but inulin and mannite; (5) all but inulin and lactose; and (6) all but inulin, mannite and saccharose.
4. In view of the comparatively small number of streptococci employed we are hardly warranted in making a definite classification. Perhaps a larger series of tests upon the media employed will enable us to divide streptococci into distinct classes characterized by certain fixed fermentative properties.