VETEEINAEY MEDICINE. 779 



the fact that the test tube reartious, such as agglutination and precipitation, 

 prove ineffective. In infectious disease tlie coexistence of the antigen (tlie 

 infectious agent or its product) in the blood might be expected to produce this 

 result." 



XV. Equilibrium in precipitation reactions. Eqtiilibriiim in combination (pp. 

 1&-34). — The supernatant fluid resulting from the preparation of precipitates 

 with rabbit serum, previously immunized against horse serum, and normal 

 horse serum was found to contain both antigen and antibody. Similar results 

 were obtained with raw egg albumin. Purified egg albumin, however, exhibited 

 entirely different results. When treated as above the supernatant fluid never 

 contained both antigen and antibody simultaneously. Raw egg albumin was 

 found to contain more than one antigen, and the antiserum, therefore, corre- 

 spondingly contained more than one antibody. The protective action of a third 

 colloid, preventing complete interaction of antigen and antibody, has not been 

 demonstrated. Cross precipitations could not be induced by mixing the sera 

 of different rabbits immunized against crystalline egg albumin, but were possi- 

 ble when raw egg albumin was used as antigen. 



From the experimental results it is concluded that precipitin and precipi- 

 tinogen can not exist in the same fluid without undergoing union and produc- 

 ing precipitation. The instances of the apparent coexistence of antigen and 

 antibody reported in the literature must be explained on the basis of multi- 

 plicity of antigens. 



XVI. Equilibrium in precipitation reactions. Dissociation (pp. 35-46). — Ex- 

 perimental data show that "precipitates contain both antigen and antibody, 

 as shown by the fact that they sensitize both actively and passively. If pre- 

 cipitates ai"e treated with salt solution in the incubator, the extracts are found 

 to contain a small amount of antigen, but no antibody. If precipitates are ex- 

 tracted with solutions of sodium carbonate, antigen is readily demonstrable in 

 the extracts. Precipitin can not be demonstrated, but antibody is demonstrable 

 in large amounts, by the method of passive sensitization (sensitizin). Extrac- 

 tion with trypsin and with leucocytes yields both precipitin and precipitinogen." 



XVII. On tlie coexistence of antigen and antibody in the body (pp. 47-49). — 

 The author concludes that antigen and antibody may coexist in the same fluids 

 In the test tube in reactive form, and also in the blood and in the cells of the 

 living animal. Antibody, even if in combination with antigen, is still capable of 

 reacting with fresh antigen. 



Studies in nonspecific complement fixation, I— V (Jour. Infect. Diseases, 18 

 (1916), No 1, pp. 20-87). — The studies here reported consist of a number of 

 papers as follows: (1) Nonspecific Complement Fixation by Normal Rabbit 

 Serum, by J. A. Kolmer and Mary E. Trist (pp. 20-2G) ; (2) Nonspecific Com- 

 plement Fixation by Normal Dog Serum, by J. A. Kolmer, Mary E. Trist, and 

 G. D. Heist (pp. 27-31) ; (3) The Influence of Splenectomy and Anesthetics on 

 the Nonspecific Complement Fixation Sometimes Shown by Normal Rabbit and 

 Dog Sera, by J. A. Kolmer and R. M. Pearce (pp. 32-45) ; (4) The Relation of 

 Serum Lipoids and Proteins to Nonspecific Complement Fixation with Normal 

 Rabbit and Dog Sera, by J. A. Kolmer (pp. 4G-63) ; and (5) The Effect of 

 Heat on Normal Rabbit and Dog Sera in Relation to Antilytic and Nonspecific 

 Complement-Fixation Reactions, by J. A. Kolmer and Mary E. Trist (pp. 

 64-87). 



It is indicated that the sera of rabbits which are intended for use in com- 

 plement-fixation tests should be tested several times before the animals are in- 

 oculated, preferably with the particular antigen to be used, and only those 

 selected that react negatively. It is further emphasized that great caution 



