THE PNEUMOCOCCUS 325 
of Cole and his associates suggests that a second attack of the disease 
may be caused by a different type of pneumococcus; their experiments 
indicate that antibodies specific for one type are not protective against 
infection with the other types. This view, however, does not always 
seem to be borne out by the facts. Numerous cases are known in which 
two or more attacks of acute lobar pneumonia have occurred in the 
same person and the same type of organism has been found each time. 
The serum of convalescent pneumonia patients exliibits relatively 
feeble bactericidal acti\'ity, even upon the homologous strain of the 
pneumococcus, and the mechanism which leads to recovery is not 
definitely known. Xeufeld^ and others have advanced the hypothesis, 
based upon careful observation, that the crisis in pneumonia, which 
usually marks the end of the prominent clinical symptoms, is asso- 
ciated with a somewhat abrupt increase in the amount of specific 
opsonin of the blood— an increase in bacteriotropins in Neufeld's 
terminology. This theory assumes that leukocytes play a prominent 
part in the healing process, and that phagocytic activity becomes 
efficient at or about the time of the crisis. 
Neufeld and Handel,' and Cole and his associates^ have produced a 
serum by injecting gradually increasing doses of very virulent pneu- 
mococci into horses, which protects susceptible animals, as mice, 
against many times the fatal dose of the homologous strain of organism. 
Cole has used these sera clinically in the treatment of pneumonia with 
promising results in infections caused by Types I and II of the pneu- 
mococcus. The serum appears to destroy or greatly reduce the number 
of pneumococci in the blood, and to be of material benefit in reducing 
the severity of the infection. At present a satisfactory serum for infec- 
tion with Type III, Pneumococcus mucosus, has not been prepared. 
Cole specifically directs attention to the necessity of identifying the 
type of infecting organism (by agglutination reactions) before admin- 
istering the serum. It is imperative that the homologous serum be 
used. The use of vaccines has been suggested as a method for inducing 
immunity. Experiments upon animals have been promising, but 
evidence from human immunization experiments is as yet inconclusive. 
Cecil and Steften,^ however, have reported some important experiments 
in which monkeys were protected against Type I pneumococci by sub- 
cutaneous injections of Type I pneumococcus vaccine. 
Bacteriological Diagnosis. — Pneumococci are found in the throats 
of a \ery considerable percentage of healthy adults, consequently 
the identification of pneumococci in the sputum is probably of com- 
paratively little clinical significance unless the type of the organism 
is determined. Type III pneumococcus, however, frequently shows 
a distinct capsule when stained by Gram's method. Dochez and 
1 Ztschr. f. Immunitiitsforsch., orig., 1909, 3, 159. 
- Arb. a. d. kais, Gesamte, 1910, 34, 293. 
' Jour. Am. Med. Assn., 191.3, 41, 663; New York Med. Jour., .January 2 and 9, 1915. 
* Jour. Exp. Med., 1921, 34, 245; 1923, 38, 149. 
