On the Relations between the „Islets of Langerhans'' etc. 87 



The first of these changes, namely the conversion of alveolus into 

 islet in exhaustion and inanition has been abundantly proved by the 

 work of Dale and ourselves. 



It remains to discuss the second change, viz. the reconstruction 

 of alveoli from islets. Dale states that a compartment in the islet of 

 the toad can often be found in which the cells show an alveolar 

 arrangement into two layers round an indistinct lumen. In such a case 

 the cells of the outer layer have spherical nuclei, which are of the 

 type found in secretory cells, while the cell-substance shows basophile 

 staining, and, in a few cases, a small number of eosinophile granules. 

 Such an appearance, Dale believes, represents the early stage of recon- 

 struction of an alveolus from the islet. But, so far as we are aware 

 no previous observers have given evidence of a similar reconstruction 

 in mammals. 



We are able to bring forward very strong evidence that such a 

 reformation of alveolus from islet may occur in the dog. 



After having made ourselves thoroughly familiar with the general 

 appearances and proportion of islet to alveolus in the pancreas of a 

 dog after a certain number of days' inanition, we have in three 

 experiments, allowed the animal to become restored to its normal 

 condition after the lapse of such a period. Again taking splenic 

 end for microscopic examination and comparison, we find that the 

 smallness and relative paucity of islets, characteristic of the normal 

 state are once more found. Being certain of what was the condition 

 during inanition, we can be equally certain that either alveoli have 

 been reconstructed from islets or that the islets have disappeared 

 altogether and new alveoli formed from the old ones, in order to 

 restore the gland to its original condition. The first alternative is by 

 far the most probable. 



In the pancreas of some of these animals, we have seen abundant 

 karyokinetic figures in the remaining islets and the alveoli immediately 

 surrounding them. The precise significance of these we are not pre- 

 pared to discuss, for it is obviously conceivable that the reconstruction 

 might occur witliout any increase in the number of cells. 



