86 CLINICAL BACTERIOLOGY AND H^MATOLOGY 



preferably of carbolic lotion) to the required point, insert the 

 pipettes, and place it in the incubator. If this is not at hand, 

 the test-tube may be prepared as before, and occasionally 

 warmed over the flame as it cools. The reaction is often 

 complete in ten minutes, usually in half an hour, and never 

 need be watched for more than one hour, provided it is incu- 

 bated as described. The reaction can be carried out at the 

 room temperature, but is much slower and not so satisfactory. 



Examine the tubes from time to time. In a negative 

 reaction the bacilli will very gradually settle to the bottom 

 of the column in all the pipettes, including the control, and 

 after twelve to twenty-four hours or more form a compact 

 white mass. In a positive reaction the appearances are very 

 different. They are first seen in the stronger dilution, then 

 the weaker ones in order. The fluid in the pipette showing 

 them first becomes slightly granular : this is best seen by 

 looking at it by transmitted light, holding the tube in front 

 of a dark object over which the light passes. After a time 

 the granularity becomes easily visible, and the bacilli have 

 collected into flocculi most obvious to the naked eye. These 

 then settle to the bottom, forming a loose white or greyish 

 mass, which appears much more voluminous than the com- 

 pact mass formed by the settling- of unclumped bacilli. If a 

 powerful serum is used the whole process may take place in 

 a few minutes, and as a matter of fact it is unusual to meet 

 with a positive reaction in which there is not marked clumping 

 in the I in 10 pipette in ten minutes or a quarter of an hour. 

 If it is not practicable to watch the actual clumping, it is pos- 

 sible to distinguish between the agglutinated bacilli and those 

 that have merely settled without being clumped, the deposit 

 in the former case being so much looser and more voluminous 

 than in the latter. If the tubes are inverted agglutinated 

 bacilli will fall in a coherent mass, whilst those which are 

 deposited will settle in a loose shower. 



Sera are often met with which clump in much higher dilu- 

 tions than are here described, but it is not necessary to go 

 higher in practical diagnosis. A positive reaction at i in 50 

 is conclusive, at i in 30 practically so, and it is extremely rare 

 to find a serum which clumps in half an hour at i in 10 which 

 does not indicate a positive reaction. 



