THE BACILLUS OF TUBERCULOSIS 307 



"The range of the patient's temperature is ascertained by taking it 

 at 8 A.M., 3 P.M., and 8 P.M. for three or four days before making the 

 test. The first injection should not exceed 0.5 nig., and if any fever is 

 habitually present should be even less, and is best given early in the 

 morning or late at night, as the typical reaction usually begins, in my 

 experience, within six or twelve hours. Such a small dose, while it will 

 often be sufficient to produce the looked-for rise of temperature, has, 

 under my observation, never produced unpleasant or violent symp- 

 toms. An interval of two or three days should be allowed between each 

 of the two or three subsequent injections it may be necessary to give, 

 as reaction in very rare cases may be delayed for twenty-four or even 

 thirty-six hours. On the third day a second dose of 1 mg. is given, and 

 if no effect is produced a third, of 2 mg., three days later. In the great 

 majority of cases of latent tuberculosis an appreciable reaction will 

 be produced by the time a dose of 2 mg. has been reached. If no effect 

 has been caused by the tests applied as above I have usually gone no 

 farther, and concluded that no tuberculous process was present, or at 

 least not to a degree which need be taken into account in advising 

 the patient, or which would warrant insisting on a radical change in 

 his surroundings and mode of life. If some slight symptoms, how- 

 ever, have been produced by a dose of 2 mg., it may be necessary 

 to give a fourth injection of 3 mg. in order to reach a positive conclu- 

 sion. Nevertheless, it should be borne in mind that in a few cases the 

 exhibition of even larger doses may cause reaction and indicate the 

 existence of some slight latent tuberculous lesion, and the test should 

 not, when applied within the moderate doses described, be considered 

 absolutely infallible. 



"No evidence in connection with the tuberculin test as applied to 

 man and animals has been forthcoming thus far from those who have 

 made use of it, which would tend to sustain the general impression 

 that this method is necessarily dangerous and tends invariably to aggra- 

 vate the disease, and my own experience has developed nothing which 

 would seem to confirm this impression. It is evident that the size of 

 the doses given has much to do with the limitations of this method 

 for usefulness, and the correctness of the conclusions reached by its 

 application. The tuberculin used is also a matter of some importance 

 in determining the dosage, as different samples vary considerably in 

 their efficiency. If the test be pushed to the injection of such large 

 amounts as 10 mg. or more, as advocated by Maragliano, such doses 

 are by no means free from the objection of occasionally causing un- 

 pleasant and sometimes dangerous symptoms; and even if the amount 

 given be not carried to the dose of 10 mg., which is known to produce 

 fever in healthy subjects, it is likely that on account of individual sus- 

 ceptibility or the presence of some other morbid process in the body, 

 reaction will be found to occur with the larger doses when no tubercu- 

 lous process exists. The adoption of an initial dose so small as to 

 guard against the absolute possibility of producing violent reactionary 

 symptoms, and the graded increase of the subsequent doses within 



