456 BACTERIOLOGY. 
phology may vary considerably, even at times when 
grown on the same culture media, and the motility is 
not always equally active; the flagella formation may 
vary; the rapidity of growth may differ, especially 
between freshly made and old cultures; the grape-leaf 
appearance of the surface colonies on gelatin, which is 
usually characteristic, may vary with the composition 
of the gelatin, at times no typical colonies at all being 
presented; the threads in the Hiss media may be lack- 
ing; the indol test requires great care in its perform- 
ance, and in rare instances the typhoid bacillus produces 
it; the growth on potato is not to he depended on, being 
often visible and not characteristic; the virulence of 
both the bacilli is so little characteristic that it can 
hardly be used for diagnostic purposes; and, finally, the 
serum test is not absolutely infallible in all cases, for 
once we met with a bacillus in feces which grew ina 
manner utterly at variance with the typhoid bacillus, 
yet still gave the Widal reaction perfectly with the 
serum of an immunized horse. It is also stated by 
Abbott that all typhoid bacilli do not give the Widal 
reaction with the serum derived from a typhoid in- 
fection with a single variety of a typhoid bacillus. 
This is an experience that as yet we have not met 
with. The Pfeiffer reaction in guinea-pigs is a matter 
of extreme delicacy, aud varying results are sometimes 
obtained. 
In spite, however, of these difficulties it is very easy 
to sufficiently identify the typhoid bacillus for all prac- 
tical purposes. A bacillus which grows typically in 
the Hiss tube media and shows the Widal reaction with 
a high dilution of the serum of an animal immunized 
