Being personally satisfied by clinical observations that certain of the acute insanities are toxic in their ætiology, it occurred to me that it might be possible to demonstrate the presence of the toxæmia by means of the delicate test of complement-deviation. I argued that if a toxin existed in the blood, it would probably be excreted in the urine, and that at certain stages of the disease process a corresponding immune body would be found in the serum. In a series of observations carried out on these lines, I used urines as the antigen, and serum as the immune body.