It is possible by means of curves to depict graphically the behavior of the various antibodies under various conditions. Logarithmic curves based on the data presented in the tables are submitted in Text-fig. 1.
In general it is evident that antibody destruction goes on gradually as the temperature is increased. Thus 65°C. for 20 minutes diminishes the activity of all the antibodies with the exception of red cell agglutinin, and in this case although the final titer was the same evidently some of the antibody was inactivated, since the reaction was weaker in the higher dilutions. It can, then, be said that 65°C. for 20 minutes appreciably affects the activity of all the antibodies tested. When the temperature is increased to 70°C. more marked differences are apparent. Here both types of the bacterial agglutinin and the precipitin are fairly stable when compared with hemolysin and red cell agglutinin. In both instances there is a sharp decline in the activity of the antibody. 75°C., however, is even a more critical temperature since at this point the somatic bacterial agglutinin and the precipitin are completely inactivated. The hemolysin and hemagglutinin behave alike. The flagellar agglutinin is the most resistant of the group to this temperature. When the temperature is increased to 80°C. the red cell agglutinin is completely inactivated, but sufficient hemolysin still remains to give a slight reaction at the lowest dilution. A further increase to 85°C. completely destroyed the hemolysin but left a definite amount of flagellar agglutinin; in fact, 90°C. for 20 minutes did not completely destroy this substance, since well defined clumps in the lower serum dilutions could be detected on microscopic examination. In this respect, then, the observations of Beyer and Reagh and Orcutt that there is a well defined difference between the two agglutinins for the hog cholera bacillus have been confirmed. However, each substance tested, with perhaps two exceptions, differs in its behavior to heat. It is of interest to point out the similarities in the reaction of somatic agglutinin and precipitin. Both are diminished when heated to 65°C.; 70°C. further affects the agglutinin, but not the precipitin; 75°C. completely inactivates both.
The assumption that the substances are apparently destroyed when they cease to react visibly with their respective antigens seems well founded since they cannot be reactivated with normal serum and no longer react to divert complement when combined in a hemolytic system.
It might be of interest to mention briefly other experiments in which the temperature was kept constant and the time varied. Thus temperatures of 50–55°C. and 60°C. maintained for 8 hours had no effect on antibody. 60°C. for 4 days failed to alter materially the flagellar agglutinin, although the same temperature for 24 hours inactivated the somatic agglutinin and the cow serum precipitin. Hemolysin deteriorates slowly at 60°C., so that after 4 days the serum, which originally reacted at a dilution of 1:10,240, only titered 1:160. The red cell agglutinin was about as resistant as the hemolysin in that a little still remained at the end of the test period. The experiments while incomplete add further proof that the somatic agglutinin and the precipitin are the least resistant to heat, while the flagellar agglutinin is on the whole comparatively stable.
A final experiment was performed to determine, if possible, at what temperature rabbit serum globulin was inactivated. With this in view, globulin was obtained by precipitation with ammonium sulfate, and a series of guinea pigs and chickens received several intraperitoneal injections. In no instance was a globulin precipitin obtained. By immunizing fowls in a similar manner with small quantities of rabbit serum good precipitin was obtained. The diluted rabbit sera heated at various temperatures for 20 minutes were tested for their antigenic activity with rabbit serum precipitin. It was found that diluted serum heated to 90°C. for 20 minutes reacted to about the same antigenic level as that not heated. Even boiling for 20 minutes failed to reduce greatly its antigenic properties. Paradoxically the visibility of the reaction was more intense with the antigen heated at the higher temperatures. The phenomenon was altogether so opposed to the usual conceptions of the inactivation of antigens that the subject will be gone into with more detail in a later communication. Although it is not possible to show definitely in the experiments that the globulin is or is not inactivated at certain temperatures, nevertheless it appears probable that certain of the antibodies are destroyed at temperatures below that capable of greatly altering globulin.
It must be recognized that comparisons cannot be made between similar antibodies in the serum of different species, since somatic agglutinin in rabbit serum resisted 70°C. for 20 minutes, although the same agglutinin in cow serum was destroyed at 65°C.