THE TECHNIQUE OF SERUM REACTIONS 303 



this purpose is freed from red blood corpuscles by centrifugation. 

 From the 1 to 10 diluton any number of higher dilutions may be made, 

 by mixing given parts of the 1 to 10 dilution with normal salt 

 solution ; thus one part of a 1 to 10 dilution plus an equal quantity 

 of salt solution gives a dilution of 1 to 20. One part of one to ten 

 dilution plus two parts of normal salt solution gives one to thirty, 

 etc. It must not be forgotten that, when equal parts of the serum 

 and bacillary emulsion have been mixed, each one of these dilutions 

 is doubled. 



In making the microscopic agglutination test, equal quantities of 

 serum dilution and bacterial emulsion are mixed upon a cover-slip. 

 The mixture may be made either by measuring out a drop of each 

 substance with a standard platinum loop, depositing them close 

 together on the cover-slip, and mixing; or equal quantities may be 

 sucked up, each to a given mark, in a capillary pipette, mixed by 

 suction in and out, and deposited upon the cover-slip. The cover- 

 slip is inverted over a hollow glass slide, the rim of which has been 

 greased with vaseline. The drop is then observed through a (Leitz) 

 No. 7 lens, ocular No. 3. 



Macroscopic agglutination, preferable for exact laboratory re- 

 search, is made in narrow test tubes measuring about 0.5 cm. in 

 diameter and about 5 cm. in length. 



Equal quantities, usually 1 c.c. each, of serum dilution and 

 emulsion are mixed. A series of tubes is prepared, in each subse- 

 quent one of which the dilution is higher. These mixtures may be 

 placed in the incubator for a few hours and then kept at room 

 temperature. After removal from the incubator agglutination is in 

 some instances hastened by transference to the ice chest. When 

 agglutination takes place in these tubes, clumps of bacteria may be 

 seen to form, which settle to the bottom of the tube, very much 

 like snow-flakes. The surface of the sediment is heaped up and 

 irregular. The supernatant fluid becomes entirely clear. When the 

 reaction does not occur the sediment is an even, granular one with, 

 a flat surface, and the emulsion remains turbid. 



Instead of using test tubes as described above, Wright has sug- 

 gested the use of throttle pipettes of comparatively large diameter 

 into each of which at least throe or four different dilutions can be 

 sucked with a nipple, a small air bubble being left between, the 

 mixtures. By sealing the distal end of these pipettes in a flame 

 the various dilutions are kept at a distance from each other, and 



