hæmoglobin, and in doubtful cases the spectroscope should be used.
The serum must be completely freed from its form-elements, a point which is often neglected. A serum may appear absolutely clear, and yet contain millions of red blood corpuscles. The serum must be treated with a good electric centrifuge until the tube shows no trace of blood corpuscles, either on its sides or at the bottom. The serum, after each treatment with the centrifuge, is drawn off with a pipette and transferred to another tube, and during this operation, in order to avoid any contact of the pipette with the red blood corpuscles, the tube is placed upon a mirror. One can then see exactly where the end of the pipette is at any moment. The blood is best taken with an absolutely dry needle and placed directly into a sterilized centrifuge tube, or, better still, into a small Erlenmeyer flask. The blood is allowed to clot spontaneously, and is watched until the serum separates out. Any mode of procedure which accelerates the separation of the serum increases the risk of hæmolysis. The blood should not be placed either in an ice-chest or in an incubator, but should be left simply at room temperature. In the first case, the risk of hæmolysis is very great; in the second, autolysis of the form-elements generally results. Serum is generally obtained in a considerable quantity after five or six hours, but if enough has not
