THE BACILLUS OF WHOOPING-COUGH. 473 



teria and numerous other extraneous organisms. This latter form 

 of expectoration should be avoided in making cultures. If the 

 expectoration is examined on successive days, it is found that the 

 leucocytic exudate contains fewer and fewer specific bacilli and 

 that phagocytosis occurs more frequently than at first. Although 

 the accesses of whooping-cough continue numerous and characteris- 

 tic, the richness of the culture becomes less. It may be that the 

 organism still persists in the tissue of the bronchi, but, at any rate, 

 it is certain that the number of organisms eliminated by expecto- 

 ration decreases notably. This later exudate is less suited for 

 growing the organisms, and, what is more important still, micro- 

 scopical preparations of it evidence this fact clearly; to one who 

 has studied the previous earlier expectorations, the subsequent less 

 favorable ones show clearly the importance of this bacillus and 

 its function. The relatively short duration of the favorable period 

 for obtaining cultures of the whooping-cough bacillus, the unequal 

 distribution of the bacillus in the sputum, and, furthermore, the 

 difficulty in obtaining the secretions at the proper moment from 

 infants, all render the study difficult. Although the research is 

 difficult even when dealing with pure cases of whooping-cough, 

 it becomes even more so when the disease is complicated by non- 

 specific colds, bronchitis, bronchopneumonia, etc.; in other words, 

 when other bacilli of respiratory affections are mixed with those 

 that cause the disease. The sputum obtained from children at 

 home is therefore evidently much more suitable for study than 

 materials obtained in the hospital. 



The necessary conditions for success in such a research, as well as 

 the causes of error that may throw an investigator off the track, will 

 appear more clearly if we give a short resume of our own attempts. 

 Incidentally we shall mention certain non-specific bacteria that we 

 have frequently met with which may be confounded with the true 

 bacillus and to which other bacteriologists have drawn attention. 



Our first researches on whooping-cough began in 1900. At 

 this time a child, B, aged 5 months, subsequent to contact with 

 children that were beginning to cough and in whom the diagnosis 

 of whooping-cough was later determined, suffered an attack of 

 typical whooping-cough. The health of this child had hitherto 

 been perfect. She had never had even the slightest disease of the 



