EXPERIMENTAL INFECTIONS 805 



The intranasal instillation of H. pertussis into anaesthetized mice produces a patchy 

 or diffuse interstitial pneumonia, leucocytic infiltration round vessels and bronchioles, 

 proUferation of the bronchiolar epithelium, and mucous secretion in the bronchioles contain- 

 ing masses of bacteria (Burnet and Timmins 1937, Bradford 1938). The histological 

 picture in many respects resembles that of the lung in human pertussis, and clearly offers 

 a near approach to the natural disease for immunological study. For intraperitoneal 

 infection, the normally low virulence of H. pertussis by the intraperitoneal route may 

 be enhanced by starch (Powell and Jamieson 1937) or mucin (Silverthorne 1938). Witebsky 

 and Salm (1937), usmg rabbits injected intradermally, produced inflammatory lesions 

 followed in 2-3 days by necrosis. 



After intratracheal inoculation, Culotta, Harvey and Gordon (1935) produced in 

 three monkeys a disease with a 10-day incubation period, a catarrhal stage, and a febrile 

 coughing stage not unlike human pertussis. By similar means Sprunt, Martin and McDear- 

 man (1938) produced an interstitial pneumonia in the monkey, characterized by a mono- 

 nuclear cell reaction, and accompanied by a lymphocytosis ; and North and his colleagues 

 (1940) induced in Macacus monkeys an uifection which by the seventh day resulted in 

 a sticky tracheal and bronchiolar exudate full of H. pertussis, and pulmonary congestion 

 with conspicuous fibrinous and cellular infiltration, both interstitially and in the alveoH. 

 None of the monkeys developed a cough. 



H. influenzce-snis. — In association with a filtrable virus (see Chapter 74) this organism 

 is an important natural pathogen of swine, and the disease can be experimentally produced 

 in these animals. In relation to the small animals of the laboratory this organism appears 

 to behave much in the same way as H. irijluenzce. Large intravenous injections may be 

 fatal for rabbits, and large intraperitoneal injections for guinea-pigs or mice ; but the 

 results are very irregular, and there appear to be great differences in the virulence, or toxicity, 

 of different strains (see Lewis and Shope 1931, Kirchenbauer 1934). 



H. cants. — The data with regard to the pathogenicity of this species for laboratorv 

 animals are extremely scanty. Rivers (19226) notes that the intraperitoneal injection of 

 1 ml. of a 24-hours' culture in blood broth failed to kill a mouse ; 2 ml. intraperitoneally 

 did not kUl a small guinea-pig ; 1 ml. intravenously did not kill a small rabbit. 



H. ducreyi. — Tomasczewski (1903) was successful in reproducing the disease in human 

 subjects with pure cultures. In man, progressive purulent lesions follow the intradermal 

 injection of cultures ; and it is apparently a common practice to separate H. ducreyi from 

 contaminating saprophytes in genital material by injecting it intradermally into the 

 patient (see Cunha 1939). Ulcerative lesions have followed the inoculation of monkeys 

 and rabbits with cultures several generations removed from primary isolation. (Relen- 

 stierna 1921, Nicolle 1923). 



H. para-influenzce. — This organism appears to be non-pathogenic for laboratorj' animals. 



Variation. 



The available data with regard to variation in the genus Hcemophilus 

 have already been referred to in the discussion of antigenic structure and of patho- 

 genicity. 



Both H. influenzcB and H. pertussis give rise in artificial culture to variants that 

 are essentially of the rough type. It would, indeed, seem that these species are 

 peculiarly liable to undergo this change. The evidence suggests that rough strains 

 of H. inflmnzcB occur very commonly in the normal nasopharynx, so that the 

 smooth — >• rough variation must be supposed in this case to be of frequent 

 occurrence when the organism is living in its normal habitat. In the case of 

 H. pertussis we have, at present, no evidence that rough variations occur among 

 recently isolated parasitic strains ; but there is much evidence to suggest that the 

 production of rough variants is readily induced by growing the organism on a 

 relatively unfavourable medium. 



