NATURAL AND IMMUNE ANTIBODIES 149 



quently hemolyzing the red cells from another person. When 

 antibodies are present in the blood of one member of a species 

 for the red cells of another member of the same species the anti- 

 bodies are called isohemolysins and isohernagglutinins, respectively. 

 It is quite important in transfusing a patient to select as a donor 

 some one whose blood cells are not clumped by the serum of the 

 recipient. Downs, Jones, and Koerber (1929) have shown that 

 these isohemolysins require complement for their action as do other 

 hemolysins produced by injection of foreign red cells. Isohemo- 

 lysins and isohemagglutinins have been studied in rabbits, cattle 

 and other animals. These substances and the inheritance factors 

 involved will be discussed in another chapter under Blood Group- 

 ing. 



Isohemolysins in Paroxysmal Hemoglobinuria.— Another type 

 of isohemolysin is described by Donath and Landsteiner (1904). 

 This interesting type of hemolysin is apparently responsible for the 

 clinical condition known as paroxysmal hemoglobinuria. They 

 found in the serum or plasma of patients suffering from this dis- 

 ease, antibodies (isolysins) which would combine with the patient's 

 own red cells only at low temperatures. They reasoned that the 

 red cells of patients liaving tliese antibodies would become sensi- 

 tized in the skin capillaries when the surface of the body was 

 exposed to cold and would combine with complement after sensi- 

 tization. When the sensitized red cells with their complement were 

 transported to the internal organs where a temperature of 37° C. 

 would be encountered, hemolysis would occur. Mackenzie (1927) 

 has reviewed this subject and gives an excellent discussion of the 

 work of Donath and Landsteiner and others. 



Apparently the blood serum of almost every case of paroxysmal 

 hemoglobinuria reacts with Wassermann antigens to give a posi- 

 tive serological reaction. While 24 to 25 per cent give a history 

 of syphilis, it appears that approximately 75 per cent are negative 

 as far as history goes; the diagnosis being made upon the positive 

 serological findings. This is interesting in view of the work on 

 natural antibody-like substances (reagins) that are found quite 

 extensively in the blood of lower animals and to some extent in 

 the blood of normal human individuals. Kahn (1940, 1941) has 

 shown that these antibodies which unite consistently with Kahn 

 antigen and less extensively with Wassermann antigen do so more 



