Ill 



last one was obtained which showed very regular 



J O 



results. The whole experiment was carried out on one 

 day with the same blood-dilution and all the values 

 therefore are completely commensurable. They are gi- 

 ven in the following table, where the first column indi- 

 cates how many (n) cc. of antitoxin-solution (V4oo%) 

 were added to 2 cc. of a 2% tetanolysinsolution and 

 to (2 - - n) cc. water. Under the heading ,,x obs." is 

 indicated the "toxic quantity" i. e. the quantity ot this 

 mixture, which added to 10 cc. 2,5 % horseblood (dilu- 

 ted with saltwater) produced a haemolysis == 20. When 

 amounts over 1 cc. are indicated, only y ]0 of a 10 times 

 more concentrated solution was added. These cases 

 are marked by an asterisk in the table below. Under 

 the supposition that all the tubes, with the same degree 

 of haemolysis, contained the same amount of toxin, the 

 amount of toxin which was present in every cc. of the 

 solution added was calculated in arbitrary units. In 

 reality, we calculated how much toxin was present in 

 the whole mixture of 10 cc. blood and the added amount 

 of toxin (this amount of toxin being proportionate to the 

 total volume); and it was then supposed that if this 

 fluid, the blood corpuscles excepted, was concentrated 

 to the original volume of toxin added, this ought not 

 to cause any change in the loxicity - - or if a such 

 should really be present, that it would be of the same 

 relative, size in all cases. The observations below 

 mentioned show this supposition to be in concordance 

 with the experience. Is x the toxic quantity, the loxi- 

 city (Jr will be found after the following formula: 



1 10 + x 



