EXPERIMENTS ON THE PRODUCTION OF SPECIFIC ANTISERA FOR INFECTIONS OF UNKNOWN CAUSE : I. TYPE EXPERIMENTS WITH KNOWN ANTIGENS—A BACTERIAL HEMOTOXIN (MEGATHERIOLYSIN), THE PNEUMOCOCCUS, AND POLIOMYELITIC VIRUS.

Attempts to produce antisera in animals to combat specific infections are usually deferred until the cause of the infection has been isolated and grown in pure culture to furnish antigen. It has seemed to us that the fulfillment of these conditions might in some cases be rendered unnecessary through the use of infected tissue itself as an antigen, combined with selective absorption of the antiserum to rid it of elements injurious to the species furnishing the tissue. In order to test this possibility type experiments have been carried out with immune sera effective against known antigens of three different sorts: 1. Sera resulting from the injection of rabbits and a goat with normal guinea pig tissues and a bacterial hemotoxin, the megatheriolysin described by Todd, which hemolyzes guinea pig cells. The sera possessed strong antitoxins for the megatheriolysin but were fatal to guinea pigs. By the method of selective absorption they were rendered innocuous to these animals and were successfully used to protect them from lethal doses of the megatheriolysin. 2. Anti-rabbit dog sera containing antibodies protective against pneumococcus infection. Such sera, subjected to repeated absorption with rabbit red cells, proved capable of protecting mice from pneumococcus infection in exactly the same degree as the unexhausted serum; that is to say, they protected against 100 times the dose of pneurnococci that was fatal with normal dog serum. 3. The serum of a monkey recovered from poliomyelitis and repeatedly injected with human red cells and extract of placental tissue. This serum, after selective absorption with human red cells, protected a monkey against an intracerebral dose of poliomyelitic virus fatal to eight other monkeys given it with normal monkey or human serum. The results in these instances, purposely chosen for their simplicity, would seem to indicate for the absorption method some usefulness in the study of immunity to infections of unknown cause. In Part II of our paper the method is applied to one such infection; namely, a sarcoma of the fowl engendered by a filterable agent. A general discussion will be found in connection with this portion of the work.