160 THE MICHIGAN ACADEMY OF SCIENCE. 



at 37°. At the end of three hours, a portion consisting of 2 c.c. was 

 removed from each tube and tested with iodine with the following 

 result r 



TABLE IV. 



Tube containing control sac / Blue, 



" " sac 1 Achromic, 



(( 1 1 ii 2 (' 



I( (I (1 O tl 



(( (( II ^ ' u 



" " ''5'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'. Pink, 



" " "6 Achromic. 



When examined at the end of another hour, the liquid from the tube 

 containing sac 5 also gave an achromic reaction. 



Tests of the empty sacs by distention under water with air showed 

 complete absence of leaks, both before and affer the experiments. The 

 only way in which the ptyalin could have escaped from the sacs into 

 the surrounding liquid was by dialysis. Repeated experiments con- 

 firmed this result. In no instance did we fail to obtain a ready dialysis. 



Having dialyzed the enzyme, we next attempted to filter it through 

 the collodium sac. We know of no previous attempts to use the col- 

 lodium sac as a filter. Crendiroupoulis and Buffer (1900) called at- 

 tention to the imperfect permeability to diphtheria toxin of the collo- 

 dium wall when used as a filtering septum. Their method, however, 

 was one of dialysis and not of filtration. The same may be said of the 

 work of Rodet and Gu(^choff (1900) on the permeability of collodium 

 sacs to peptone, albumin and sugar. The first meutioued writers state 

 that ''the collodium sac is not as permeable as is generally believed by 

 bacteriologists," and the work of these men is of interest in calling 

 attention to this fact. 



The filtering apparatus employed is the device of Dr. Frederick G. 

 Novy. It consists of a sac, such as previously described, mounted on a 

 glass tube, the joint here being further strengthened by winding with 

 silk thread and recoating with collodium. The tube then passes through 

 a rubber stopper, the whole of which makes an air-tight connection with 

 top of a test tube on foot. This test tube on foot is provided with a 

 side arm, through which attachment is made to a suction pump. 



The sac was tested before use by distention under water with air at 

 the pressure of the ensuing filtration, and found free from leaks. Five 

 c.c. of actively diastati& saliva, previously filtered through a single 

 thickness of ordinary filter paper, were then placed within the sac, and 

 suction was begun. The sac swelled and droplets appeared on the out- 

 side like beads of perspiration. These became* confluent, collected at the 

 bottom of the sac and fell to the bottom of the tube. The filtrate was 

 then compared in diastatic activity with the contents of the sac, which 

 served as control. 



Experiment 1. — II/2 c.c of liquid were filtered through the sac in 1 

 hour and 30 minutes at a pressure of 2% inches. 0.1 c.c. of the contents 

 of the sac was found to invert 30 c.c. of 1 per cent, starch solution in 22 

 minutes, while the filtrate was entirely without effect in 22 hours. 



Experiment 2. — 3.4 c.c. of liquid were filtered through the sac in 1^^ 

 hours of a pressure of 2%-3^ inches. 0.1 c.c of the contents of the sac 



