BACILLUS TYPHOSUS. 427 
occurs—nothing more is needed; the result is negative 
as far as this specimen is concerned. If marked 
clumping and immobilization of the bacilli immedi- 
ately begin and become complete within five minutes, 
this is denominated a marked immediate typhoid re- 
action, and no further test is considered necessary, 
though it is always advisable to confirm the reaction 
with higher dilutions up to 1 to 20 and 1 to 50. If, 
however, upon examination of the mixture there is no 
marked immediate reaction, but the bacilli only show 
in the first few minutes an inhibition in their motility 
and a tendency to clump, which becomes more marked 
but not complete within five minutes, this must be 
tested with the higher dilution of 1 to 20, so as to 
measure the exact strength of the reaction. If in the 
1 to 20 dilution a complete reaction takes place within 
thirty minutes, the blood is considered to have come 
from a case of typhoid infection, while if a less com- 
plete reaction occurs it is considered that a probability 
only of typhoid infection has been established. The 
time allowed for the development of the reaction with 
the high dilutions is by many from one to two hours, 
but to us thirty minutes seem safer. Positive results 
obtained in this way may be taken to be conclusive 
unless there be grounds for suspecting that the reaction 
may be owing to a previous fairly recent attack. The 
absence of reaction in one examination is considered 
by us to in no way exclude typhoid infection. If the 
case remains clinically doubtful, the examination should 
be repeated within a few days. 
Proportion of Cases of Typhoid Fever in which a Defi- 
nite Reaction Occurs and the Time of its Appearance. 
As the result of a large number of cases examined in 
