DIABETOGENIC ACTION 387 



nals, the levels are not only restored but appear to be higher at 90 min. 

 More striking effects might have been noted if a shorter interval than 



45 min had been used. This was done by Binet et al. (1958) and the initial 

 decreases in glutathione are indeed greater (see accompanying tabulation). 



These results are not only quite different from those of Brada, but with 

 respect to blood and liver changes are different from those of all other 

 workers. 



The results obtained on the alloxan-induced changes in pancreatic thiols 

 are unfortunately divergent. We have seen that Brada (1951) observed a 

 significant fall in pancreatic glutathione 45 min following the administra- 

 tion of alloxan, but Hultquist (1956) could find no changes. It must be 

 realized that the /5-cells constitute only about 0.5 1% of the total mass 

 of the pancreas and, hence, analyses of the entire organ may be meaningless 

 with respect to the action on the ^-cells. MacDonald (1959) used Bennett's 

 red SH reagent to determine the SH group content of acinar and islet tis- 

 sue, and showed that alloxan lowers the /3-cell SH content relative to acinar 

 tissue, the difference being evident at 15 min and marked after 2 hr. Falk- 



