1. In the course of experiments for obtaining pneumococci from the oral secretions of human individuals, the "plate method" described in this paper was found to be the most reliable method for isolating pneumococci.
2. An alkaline two-per-cent, glucose-serum-agar was regarded as the most favorable medium for the rapid development of colonies of the pneumococcus.
3. The morphological identification of the pneumococcus was made by means of a special capsule-stain. The recognition of the organism was accomplished with precision and reliability when present in the blood and exudate of test animals, and upon the various culture media employed.
4. Pure cultures of pneumococci could by the methods described be obtained within forty-eight hours for inoculation into animals for tests of virulence.
5. Normal persons often harbor the pneumococcus in their mouths. Of the seventy-eight cases examined, thirty-nine showed the presence of the pneumococcus.
6. In a second and larger series of studies on so-called normal cases by means of cultures made from the throat on Loeffler's medium, the organism was detected seventy-one times in 204 examinations, or in 348%. In 145 cases the Streptococcus mucosus capsulatus occurred eight times (5.5%), and the Friedländer bacillus twenty-one times (14.5%).
7. Certain individuals may acquire the pneumococcus in their mouths in the hospital wards, and may continue to harbor it for a considerable period of time.
8. The pneumococcus may persist for days or weeks in the mouths of patients who have recovered from pneumonia.
9. From a study of patients in the hospital wards, certain conclusionsin regard to communicability were drawn. "Normal" individuals, in whose mouths the pneumococcus is repeatedly found to be absent, may acquire the organism by association with pneumonia or "positive normal" cases. The handker-chiefsand dishes of pneumonia and "positive normal" cases are to be regarded as some of the means of transference of the organismfrom person to person. The lips of drinking-cups and the sputum or saliva in sputum cups were found to contain living and virulent pneumococci.
10. Certain characteristic morphological types of the pneumococcuswere regularly met with in this study. These include the following: (1) the typical, (2) the small, (3) the large, (4) the bacillary, and (5) the streptococcus type of the pneumococcus.
11. Pneumococci in the mouths of normal individuals possess the morphological and cultural properties which are characteristic of the same organisms when isolated from other sources.
12. All pneumococci possess the power of fermenting inulin with the formation of acid, even if not in all generations. But this holds true only when the inulin serum-water medium as modified by the writer is used.
13. A certain configuration of the colonies, designated by the term "ring-form," when present is diagnostic of the pneumococcus.
14. The percentage of virulent pneumococci present in the mouths of normal persons was 79; in cases of pneumonia, 77.
15. Pneumococci of the large mucoid type, and giving large mucoid colonies, should be distinguished from the Streptococcus mucosus capsulatus.
16. There are doubtful diplococci of atypical morphology which can be grouped neither with the pneumococci nor with the streptococci, although they possess many features in common with both.
17. The Streptococcus mucosus capsulatus belongs to a group which is related to, but distinct from, the pneumococcus.
18. Morphological characters and animal tests are of greatest value in the identification of the pneumococcus. By the employment of suitable culture media it is possible, in most instances,to recognize the pneumococcus by its morphology alone. The chief cultural tests are the following: the manner of growth upon the serum-glucose-agar, and the modified inulin-serum water, and the appearance of the "ring-type" of colony.
19. All pneumococci, irrespective of their source, were agglutinatedby means of pneumococcus immune serum.
20. An immune pneumococcus serum was found capable of agglutinating various pyogenic streptococci, certain atypical organisms, and several strains of the Streptococcus mucosus capsulatus.
21. The serum of pneumonia patients varied in its power to agglutinate different pneumococci. Some strains were agglutinated, others not.
22. The sera of normal individuals and of normal rabbits possess no agglutinating power for pneumococci, the atypical organisms, certain streptococci, and the Streptococcus mucosus capsulatus.