154 
F. B. HADLEY. 
tions were obtained with serum from many of the horses upon 
which it was tried. At first we were inclined to believe that the 
method of preparing the agent was defective or possibly the cul¬ 
ture from which it was made was attenuated and thus influenced 
the test. Upon using at least four different preparations of the 
mallease, manufactured from two different strains of the organ¬ 
ism, we were unable to get satisfactory results. 
Our experiments have been limited in number, so we are not 
in a position to draw definite conclusions. However, if the reac¬ 
tion is so delicate as appears, it seems entirely too sensitive for 
ordinary clinical use. 
The only inference we can draw from otir work with mallease 
in conjunction with our other tests on glandered, suspected and 
healthy horses is that it is unreliable. In fact, we attempted to 
base a diagnosis upon this test alone, and pronounced certain 
horses free from the infection on the strength of a negative re¬ 
action. Later other tests were used on the same animals and posi¬ 
tive reactions of an undoubted nature were obtained, necessitating 
a change of diagnosis, as well as placing us in a compromising 
position. 
2. The Agglutination Test or Gruber-Widal Reaction .—This 
test is also based upon the formation of certain specific antibodies 
which are present in the blood in the early stages of certain in¬ 
fectious diseases. These antibodies are known as agglutinins. 
Their formation is stimulated in the case of glanders by the 
bacilli which cause the disease. The particular part which these 
substances play in the organism is not fully known. 
The phenomenon of agglutination can be demonstrated in a 
test tube or microscopically. It consists of a clumping of sus¬ 
pended bacilli specific for the infection. It is especially applicable 
in glanders, as many cases in the first stages of development, im¬ 
possible to be determined by the usual tests, are readily diag¬ 
nosed. 
The test fluid necessary for the experiment is made of a sus¬ 
pension of a known agglutinating strain of the glanders bacilli 
killed by heating to 60 C. for two hours. To insure safety, as 
