228 



(10—12 (lays nl'lcr the l)c'fTinniii}^f of llic cxpcriincnl ;. Al'lor- 

 wards. llio curve declines slowly. This and some oilier 

 experiinenls lig. 50 remind one of some carried out by 

 .lor gen sen and M a d s e n -'"';, in which Ihey show Ihc 



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 

 Passage 



Fig. 49. 

 Curve showing the variations of the degree of immunization of fibroblasts 

 through seventeen passages. Human ascitic fluid was used as antigen in 

 a concentration of 7 per cent. The ordinates represent the quotient of the 

 rate of growth of the immunized strain divided by that of the non-immu- 

 nized strain in a high concentration of antigen, and tlie abscissae the num- 

 ber of passages. 



amounl of agglutinin produced hy a goal under the in- 

 ralher abruptly and reaches a maximum al Ihe 51 h passage 

 flucnce of continued injections of the antigen, cholera ba- 

 cilli; Ihe conditions being almost the same as in my own 

 experiment. 



