BY H, G. CHAPMAN. 531 



The animals were killed 7 to 12 days after the last injection. 

 Huttall's method* was followed. The blood was caught in a 

 large sterile dish which was at once covered. The serum was 

 removed from the dish in 7 or 8 hours, and either sealed up in 

 sterile tubes and stored on ice, or drie(it in vacuo over calcium- 

 chloride at 37°C. Dried antisera were kept in a .securely closed 

 sterile bottle. Every precaution was taken to work under 

 aseptic conditions. All the tubes used for the interactions, all 

 pipettes, measuring cylinders, and other apparatus were sterilised 

 ill a Koch's steam steriliser. The salt solution used as the 

 diluent was sterilised by raising to the boiling point on three 

 successive days. The trouble of these precautions was amply 

 repaid by the tubes remaining sterile for 7 or 8 days after being 

 set up. All the tubes were plugged with sterile cotton wool. 

 When a l)acterial deposit may have simulated or concealed a 

 slight precipitate, the tubes have been rejected. 



Measurements have been made with pipettes graduated to 

 1/lOOth of a cubic centimetre, and corrected by weighing the 

 quantity of mercur}' delivered by them. 



Tiie tubes employed for the interactions were made of ordinary 

 glass tubing of about 5 mm. bore. The ends were drawn out and 

 ruunded off. The bottom of each tube was thus shaped as a cone. 

 For certain experiments, tubes of a larger bore — 8 mm. — were 

 employed. 



The Relations of the Interacting Substances. 



The characters of the interaction between the antiserum and 

 the protein employed for its production, have been studied to 

 determine the general laws governing the reaction. If a given 

 weight or volume of antiserum be allowed to interact with 

 increising quantities of protein, the precipitate formed increa.ses 



* Nuttall, Blood Immunity and Relationship, Cambridge, 1904, p.o9. 



+ On the advantage of using dried antisera, see Cha.pman, Proc. Linn. 

 Soc. N. S. Wales, 1905, xxx., p.392. 



