1912.] 



Method for Opsonic Index Estimations. 



139 



In my work this error has been studied and largely reduced. The first 

 idea was that the number of leucocytes in the mixture might vary in the 

 successive sets of materials. To test this point I repeatedly estimated the 

 opsonin of normal serum to staphylococci, and also counted the leucocytes 

 per cubic millimetre in each opsonic mixture. The results are shown in 

 Table A. 



Table A.— Opsonic Index. (Old Method.) 

 Repeated estimation of the opsonin to Staphylococcus aureus in normal serum 

 and simultaneous enumeration of the leucocytes per cubic millimetre in 

 each opsonic mixture. 





Leuco- 

 cytes per 

 cu. mm. 



Opsonic 



50 leuco- 

 cytes. 



Error 

 above 

 average. 



Error 

 average. 



Mean 



Blanks 



per 

 50 leuco- 



Worst 

 average. 



Worst 

 error 

 below 

 average. 









per cent. 



per cent. 







per cent. 



per cent. 





15 x 200 



279 





4 





9 









9 „ 



337 



15 







9 



15 





© 



14 „ 



285 





2 



| 292 ^ 



7 









11 „ 



268 





8 





6 





8 



2 



18 „ 



190 



11 







10 









16 „ 



258 



52 







3 



52 





© 



19 „ 



118 





30 



H 



20 









10 „ 



114 





33 





20 





33 



3 



14 „ 



185 



8 







16 









9 » 



155 





8 



1 f 



14 







© 



6 „ 



213 



25 







13 



25 







16 „ 



137 





19 





21 





19 





36 „ 



180 



5 





J I 



13 









12 „ 



155 





8 





14 







4 



12 „ 



151 





20 





11 





20 





7 „ 



191 









7 







© 



25 „ 



191 







H 



8 







10 „ 



224 



17 









17 









133 



132 





Average error above or below the mean value = 265/18 = 14 - 7 per cent. 



Worst error above = 52 „ 



Worst error below = 33 „ 



The encircled numbers indicate the highest leucocyte content in each series and the lower figure 

 = the number of times it occurred. 



All the materials were identical in each series, and the opsonic estimations 

 by the same operator, and the observations made at about half- hour 

 intervals. 



