334 RUEDIGER. 



render the filter impermeable. To insure thorough cleansing, the follow- 

 ing procedure has been found satisfactory : immediately after having been 

 iised, the surface of the filter is scrubbed with a brush under running 

 water and the serum is removed from the pores of the filter by passing 

 through it a liter or more of distilled water or an aqueous solution of 0.75 

 per cent sodium chloride. The filter is then boiled in distilled water, and 

 after the boiling, is again washed by passing water through it. It has 

 been my experience that if the boiling is omitted the filter gradually 

 becomes impermeable, and to insure a thorough cleansing none of the 

 steps mentioned should be omitted. 



Martini,^ in studying the effect of filtration on antidiphtheritic serum, 

 demonstrated great loss in the potency of the serum after it had passed 

 through Chamberland filters. Serum which before filtering contained 

 140 units of antitoxin per cubic centimeter, after filtering contained 100 

 units per cubic centimeter. Another lot of serum which contained 135 

 units of antitoxin per cubic centimeter, lost 75 units per cubic centimeter 

 through filtering. By evaporating he reduced antidiphtheritic serum to 

 the consistency of sjTup; the concentrated serum contained 850 units of 

 antitoxin per cubic centimeter; after passing it through a Chamberland 

 filter it contained only 40 units of antitoxin per cubic centimeter, having 

 lost 810 units per cubic centimeter. 



Dziergowsky ^ passed antidiphthetric serum through Chamberland 

 filters and could not demonstrate any loss of antitoxin thereby. 



Cobbet ^ repeated the work of the previous authors and showed that 

 antidiphtheritic serum may be passed through Chamberland filters with- 

 out losing its potency provided the filters are clean and the serum 

 passes through readily. On the other hand, if a filter is allowed to clog, 

 the serum passing through drop by drop invariably suffers great loss 

 of antitoxin. 



I have studied the question as to whetlier a serum suffers any loss 

 in immunizing value by being passed through Berkefeld filters. The 

 following experiments show the immunizing value of unfiltered and 

 filtered antitetanic and antidiphtheritic serums and the agglutinating 

 value of unfiltered and filtered typhoid serum. 



EXPERIAIENT NO. I. 



A quantity of about 1,500 cubic centimeters of antitetanic serum was divided 

 into four portions A, V, N and W. Portion A was neither centrifugated nor 

 filtered, portion V was passed through Berkefeld filter V, portion N was passed 

 through Berkefeld filter N and portion W was passed through Berkefekl filter 

 marked W. 



'Centralb. f. Bakt. etc. (1896), 20, 796. 

 '/bid. J 1897), 21, 333. 

 Ubid. (1898), 24, 386. 



