r 



1002.] STENGEL — SPECIFIC PRECIPITINS. ill 



The solution is made by first diluting the blood with distilled 

 water and then with an equal quantity of double isotonic salt solu- 

 tion. If the diluted blood is not perfectly clear, it is allowed to 

 settle and the supernatant blood is tested. A few drops of the 

 serum of the immunized rabbit are added to the clear solution of 

 blood in a narrow test-tube and a precipitation is generally ob- 

 served at once or may form after standing for a short time. A 

 later test may be performed by allowing the serum to flow down 

 the sides of the tube under the diluted blood, when a haziness at 

 the point of contact indicates the formation of a precipitate. The 

 precipitate increases somewhat after its first formation, especially 

 when the fluids are kept at 37-^ C, but precipitates formed twelve 

 to twenty- four hours or more after the mixture is made should be 

 regarded as doubtful. The same form of test may be applied with 

 dried blood. In this case care must be taken in making solutions 

 of the blood stains. The blood stains may be dissolved with 

 normal salt solution, or in the case of old blood stains the method 

 of Ziemke may be employed. In this method concentrated solu- 

 tion of potassium cyanide is used as a solvent. The mixture is 

 shaken with a few granules of tartaric acid until it becomes almost 

 neutral to litmus paper and then filtered. The mixture is diluted 

 to a yellowish-red color and the serum of the rabbit is then added. 

 Solutions of soda have also been used for extracting the blood clot. 

 Uhlenhuth demonstrated the reaction in blood (i) dried for three 

 months, (2) allowed to undergo decomposition for three months, 

 (3) washed with weak alkaline soap, (4) frozen in snow for four- 

 teen days at 10° C, and in (5) blood solutions in which the hemo- 

 globin was changed to carbon oxide hemoglobin. 



This test has been confirmed in more or less detail by investiga- 

 tors in all parts of the world and may said to be fairly established, 

 though the limits of its usefulness cannot as yet be drawn with pre- 

 cision. My own experiments with Dr. C. Y. White have consti- 

 tuted injections of a number of rabbits with defibrinated human 

 blood and with blood squeezed from placentae. In both series of 

 experiments, after from five to eight injections, we found the blood 

 serum markedly active in precipitating diluted human blood and 

 without effect upon the blood of dogs, horses and cattle. The first 

 eff'ect of adding the serum was the production of a flocculent preci- 

 pitate, which invariably increased after a short time. We dis- 



PROC. AMER. PHILOS. SOC. XLT. 171. AA. PRINTED JAN. 10, 1903. 



