No. 6. DEPARTMENT OP AGRICULTURE. 135 



lu the intestine ouly oue point of injury can be found, and most of 

 the mesenteric glands are normal in size, and show a few nodules 1 

 mm. ill diameter. The glands associated with the upper iotestiue 

 are much enlarged and show caseation, though not more advanced 

 than what is seen in the bronchial glands. 



The specimens from the monkey illustrate another important point, 

 namely, the occurrence of coincideiit infection through the upper part 

 of the digestive tract and the intestine. If this animal had been 

 examined without the knowledge of its history it would almost cer- 

 tainly have been put down as one of those cases in which the avenue 

 of infection could not be positively determined. The kmgs show 

 much more extensive disease than any of the abdominal organs, and 

 the bronchial glands show caseation as well as some of the mesen- 

 teric, and to an equal degree. This monkey was tested with tuber- 

 culin before the experiment began, and was in perfect health when 

 the feeding with tubercle bacilli commenced. The key to the situa- 

 tion lies in the enlargement and caseation of the cervical lymphatic 

 glands, indicating an infection through the tonsils or some neighbor- 

 ing part of the upper digestive tract, though no lesion could be de- 

 tected. We have here an undoubted food tuberculosis, yet the res- 

 piratory tract shows the chief involvement. Does not the same 

 thing occur also in children, and more often than we suspect? 



With these facts in view it seems not impossible that an infec- 

 tion gaining entrance through the intestines may appear lirst in the 

 lung, and thus be erroneously attributed to the respiratory tract. 



On one point all pathologists as well as clinicians seem to agree, 

 that in children tuberculosis shows a marked tendency to become 

 generalized, and this occurs with such rapidity that it is often impos- 

 sible to decide where the infection gained entrance. 



The postmortem statistics are gathered mainly from hospitals and 

 asylums for children, and it is in children that we have most reason 

 to fear infection by food, their diet consisting as it does so largely 

 of cow's milk. The tremendous mortality of children from tuber- 

 culosis has forced us to pay particular attention to the study of the 

 factors at work in early life, and a number of interesting and valu- 

 able analysis of autopsy records have been published. From Eng- 

 land we get the most positive evidence of infection through the in- 

 testine, and the most striking proofs of its frequency. English 

 pathologists are practically unanimous in regarding the intestine as 

 a frequent path of infection, the primary lesion being associated 

 with this tract in about 25 per cent, of all cases. We may examine 

 the figures given by some of the more recent writers on this point. 



Carr, examined 120 cases at the Victoria Hospital for Children, 

 London. Of these in eighty-two the disease was generalized, in the 



