SOURCES OF HUMAN INFECTION. 259 



ner by which tuberculosis is naturally transmitted can be readily 

 understood. Though the experiments of Sander show that 

 tubercle bacilli can multiply on vegetable media to a certain 

 extent at warm summer temperature, it is doubtful whether all 

 the conditions necessary for growth are provided to any extent 

 in nature. At any rate, the great multiplying ground of tubercle 

 bacilli is the animal body, and tubercular tissues and secretions 

 containing the bacilli are the chief, if not the only, means by 

 which the disease is spread. The tubercle bacilli leave the 

 body in large numbers in the sputum of phthisical patients, and 

 when the sputum becomes dried and pulverised they are set 

 free in the air. Their powers of resistance in this condition 

 have already been stated. As examples of the extent to which 

 this takes place, it may be said that their presence in the air of 

 rooms containing phthisical patients has been repeatedly dem- 

 onstrated. Williams placed glass plates covered with glycerin 

 in the ventilating shaft of the Brompton Hospital, and after 

 five days found, by microscopic examination, tubercle bacilli on 

 the surface, whilst Klein found that guinea-pigs kept in the 

 ventilating shaft became tubercular. Cornet produced tuber- 

 culosis in rabbits by inoculating them with dust collected from 

 the walls of a consumptive ward. Tubercle bacilli are also dis- 

 charged in considerable quantities in the urine in tubercular 

 disease of the urinary tract, and also by the bowel when there 

 is tubercular ulceration; but, so far as the human subject is 

 concerned, the great means of disseminating the bacilli in the 

 outer world is dried phthisical sputum, and the source of danger 

 from this means can scarcely be overestimated. Every phthisi- 

 cal patient ought to be looked upon as a fruitful source of 

 infection to those around, and should only expectorate on to 

 pieces of rag which are afterwards to be burnt, or into special 

 receptacles which are to be then sterilised either by boiling or 

 by the addition of a 5 per cent solution of carbolic acid. 



Another great source of infection is in all probability the milk 

 of cows affected with tuberculosis of the udder. In such cases 

 the presence of tubercle bacilli in the milk can usually be 

 readily detected by centrifugalising it, and then examining the 

 deposit microscopically, or by inoculating an animal with it. As 

 pointed out by Woodhead and others, the milk from cows thus 

 affected is probably the great source of tabes mesenterica, which 



