SOME PHYSICAL PROPERTIES OF PTYALIN. 101 



inverted 30 c.c. of 1 per cent, starch solution in 40 minutes, while the 

 filtrate was entireh' without effect in 45 hours. 



The sacs Avere again tested and found to be free from leaks. The 

 ptjalin had been completely retained by the collodium filter. The same 

 result was obtained in numerous similar experiments. 



The sacs after being used as filters, were washed thoroughly and used 

 for the purpose of dialysis. The same sacs, which were found to deny 

 any passage whatever to the ptyalin when used as filters, were found to 

 readily dialyze the enzyme. 



As an example, the following dialysis experiment was performed with 

 the sac of Filtration 2 (see above), — and was controlled with boiled 

 saliva in another sac under identical conditions. The dialysis and the 

 tests with iodine were performed as described above. 



The sacs were again tested and found to be free from leaks. 



The retention by the filter is not due to fixation, as the enzyme 

 traverses the sac in the longer process of dialysis. 



We have thus observed the fixation of ptyalin by hardened filter 

 paper; its dialysis through the collodium sac; and the impermeability 

 of the collodium sac to ptyalin, when used as a filter. 



It might be offered as a mere suggestion that the indialyzability of 

 enzymes through parchment is due to some such fixation as we have 

 here observed. We know that enzymes are dialyzable. We have demon- 

 strated this in the case of others besides ptyalin. We also know that a 

 fixing action is possessed by certain papers. It would be interesting to 

 determine whether this suggestion would be substantiated by the facts. 



Also as a suggestion, the idea might be advanced that these experi- 

 ments throw some light on the physical structure of the enzyme molecule, 

 arguing a complex phj^sical form or structure for the molecule, which 

 may be imagined as branching or cob-web like. It may be assumed that 

 the molecules in the filtration experiments are impelled by a force and 

 become plastered against the filtering wall, mutually hindering each 

 other from passage, while in. the case of dialysis, impelled by no such 

 pressure and subjected to no such plastering, the molecules can adjust 

 themselves to the shape and size of the pores or molecular interspaces of 

 the collodium septum, and insinuate their way through it, in some such 

 way as the blood corpuscle escapes through the containing capillary wall 

 in diapedesis. 



The results presented in this paper constitute a part of a more ex- 

 tended investigation which will be published elsewhere. This work has 

 been rendered possible by the aid granted by the Rockefeller Institute 

 for Medical Research. 

 David J. Levy^ A. B., 



Assistant in Bacteriology. 



Late scholar of the Rockefeller Institute for Medical Research. 

 Laboratory of Hygiene, University of Michigan, 

 Ann Arbor. 

 21 



