

EPIDEMIC INFLUENZA 109 



the earlier generations was mixed with protein precipi- 

 tate that interfered with the reactions and had proper- 

 ties that precluded its use. It was therefore necessary 

 to devise special methods of cultivation, and before 

 these methods became available the first opportunity 

 was lost to test for antibodies in the blood of influenza 

 patients and of affected rabbits. 



It was found later that if the Smith-Noguchi medium 

 was enclosed in a collodion sac, surrounded by distilled 

 water or physiological salt solution, anaerobic conditions 

 were shortly established throughout the system and 

 the nutritive and growth-promoting substances of the 

 medium diffused through the membrane in sufficient 

 quantities to support a luxuriant growth in the surround- 

 ing liquid. The protein precipitate that collected 

 around the tissue fragment was retained within the sac. 



When it was possible to cultivate Bacterium pneumo- 

 sintes by this method in quantities sufficient for use, 

 rabbits were repeatedly injected with small doses of 

 live cultures, or of heat-killed organisms. After a suit- 

 able interval, their blood serum was found to possess 

 specific antibodies against Bacterium pneumosintes. 



A significant feature of the immunity experiments 

 and also of these serum tests was the fact that all the 

 strains tested had similar properties and reacted iden- 

 tically with the specific antibodies produced by any 

 one of them. This is what would be expected if they 

 were all derived from a common source. 



The experiments described above based on the 

 epidemic of 1918-1919 and the recurrence of 1920, had 

 extended over three years and the facts had been care- 



