386 PATHOGENIC BACTERIA. column to which the blood rises can be taken as a proof of actual variation. It may be true that no two of the tubes have exactly—absolutely—the same contents, but when the given precautions are taken the variation will be so small as to make no significant error in the results obtained. The use of the tubes is extremely simple. The ordi- nary puncture is made in the lobule of the ear or the fin- ger-tip of the patient, and one end of one of the tubes touched to the surface of the oozing drop and held there until the blood ceases to rise in the tube. So little blood is required that a number of tubes may be filled with the blood from a single puncture if desired. The blood in the tube coagulates in a few minutes, and can be allowed to dry, or be drawn to the central portion of the tube and sealed in by fusing the ends in a flame if it be desired to keep it moist. When the agglutination reaction is to be made the blood should not be blown out of the tube, as the total quantity contained is small and a large relative quantity will remain in the tube. A better method is to crush the lube in a small crucible or other diminutive vessel and dissolve its contents directly in the culture. The proper proportionate amount of culture is meas- ured with a finely graduated pipette (graduated to thous- andths of a cubic centimeter), the calculation according to the standard tube of the writer's experiments being: dihitioii i : 10 3= 0.153 c.cm. of the culture; dilution x : 100 = 1.53 c.cm. of the culture; dilution i : 1000 = 15.3 c.cm. of the culture. The now recognized specific reaction is supposed to take place in dilutions of i : 50, which would require 0.71 4- c.cm. of the bouillon or diluted agar culture. The culture is measured into the little crucible, the blood-containing portion of the capillary tube broken off, dropped in, and subsequently crushed to minute frag- ments and stirred about with a clean, rounded, glass rod, and a drop of the mixture placed as a "hanging drop"