S. p. KRAMER 561 



392. — Guinea pig, weight 550 gm., received 4 cc. =400+ M.L.D. or eight L4- doses of 

 filtered toxin. 



393. — Guinea pig, weight 540 gm., received 5 cc. = 500+ M.L.D. or ten L+ doses of 

 filtered toxin. 



All of the guinea pigs receiving toxin filtered through the plaster of Paris-mag- 

 nesia filters were alive and well December 16, 1926, except that each pig had developed 

 an ulcer with denudation of hair at the site of injection. The size of the necrotic area 

 varied directly with the amount of filtered toxin injected. 



It was evident that the filter had removed that fraction of the toxin which caused 

 the acute death and neurotoxic symptoms, leaving the fraction which caused the local 

 tissue changes. This tissue toxin was removed by filtration through the Berkefeld 

 filter charged with gelled silicic acid. Animals that received toxin which had been 

 filtered through both filters in amounts up to ten L+ doses developed no symptom 

 whatever — either local or general. 



When diphtheria toxin is filtered through siliceous filters alone the animal dies 

 promptly of general intoxication with doses approximating the lethal dose of un- 

 filtered toxin. Botulinum toxin is also rendered innocuous by filtration through filters 

 made of plaster of Paris and magnesia. Abrin, the toxin from jequirity seed, is re- 

 moved by filtration through siliceous filters. Tuberculin, when passed through a 

 Berkefeld filter or a Berkefeld filter charged with colloidal silica, is still lethal to tu- 

 berculous guinea pigs. Tuberculin passed through a filter made of plaster of Paris 

 with 25 per cent magnesium oxide is no longer lethal to tuberculous pigs. This is il- 

 lustrated by the following experiments: 



October 12, 1926. — Guinea pig A, weight 450 gm., received 2-mg. tubercle bacilli culture 

 H. 37 in the axilla. 



November 4, 1926. — This pig received 5 cc. of a solution of tuberculin, 10:30 a.m., 

 equivalent to 0.625 cc. of old Koch tuberculin. This pig died during the night and autopsy 

 showed intense tuberculin reaction. 



October 12, 1926. — Guinea pig B, weight 430 gm., received a like amount of tubercle 

 bacilli. 



November 30, 1926. — At 10:00 a.m. guinea pig received 5 cc. of the above solution of 

 tuberculin which had passed through a Berkefeld filter charged with colloidal SiO^. This pig 

 died at 2:30 p.m. of the same day with a marked tuberculin reaction. 



October 12, 1926. — Guinea pig C, weight 420 gm., received a like amount of tubercle 

 bacilli. 



November 30, 1926. — At 10:00 a.m. guinea pig received 5 cc. of the same tuberculin 

 solution which had passed through a filter of plaster of Paris with 25 per cent MgO. The pig 

 was unaffected and remained as before injection. 



December 2, 1926. — At 9:10 a.m. guinea pig C received 5 cc. of the same tuberculin 

 filtered through the Berkefeld filter charged with colloidal silica and died nine hours later 

 with a marked tuberculin reaction. 



The gypsum-magnesia filtrate of tuberculin will not produce a skin reaction, nor 

 will it produce a tuberculin reaction in tuberculous guinea pigs in any amount. The 

 pH before filtration was 6.0 and that of the filtrate was 8.0. Guinea pigs that had re- 

 ceived twice weekly for three months i cc. of the gypsum filtrate of a solution of O.T. 

 1-8 were found to be immune to a fatal dose — 0.5 cc. — of unfiltered old tuberculin 

 undiluted. 



