1902.] NATURAL SCIENCES OF PHILADELPHIA. 317 



in the lumen, I would not go so far as to say that it may not escape 

 from the cell. The structure of the cell plainly admits this possi- 

 bility. All we can say at present is that where we might expect 

 evidence oi its passage into the lumen, no evidence is obtained. 



To sum up the evidence thus far : The granules as such come 

 from the nucleus; they are associated with food (albumose) in the 

 cytopkism; they stain only with difficulty; finally, there is no clear 

 evidence that they escape from the cell. All these indicate an 

 intracellular ferment. If it is such we should expect that it would 

 appear in the cell before the food does, or even before feeding. 

 Fig. 10, B, four hours after feeding, contains no albumose; it 

 rarely appears in the cell earlier than the eighth hour after feeding. 

 In the table. No. 3 is one of several intestines from animals starved 

 two weeks and not fed before killing; small granules were found in 

 considerable number which do not take the Biondi-Heidenhaiu 

 stain. Again, McMurrich reports for starved specimens of 

 Annadillldium some yellowish granules (which he regards as disin- 

 tegration products appai'ently identical with these). Finally, if 

 it is an intracellular ferment it should be found in all absorbing 

 cells. The definite aggregations of granules which are found in 

 the anterior cells are seldom met with in the cells of the median 

 part. However, No. 44 of the table, killed in sublimate, stained 

 in Biondi-Heidenhain, exhibited in these cells " small poorly 

 staining granules on the coelomic side" (also Nos. 13 and 18). 

 In neither of these cases are the granules in masses ; they are scat- 

 tered all through the coelomic end of the cell. While, therefore, 

 the granules are abundant and definitely aggregated in all stages 

 of absorption only in the anterior cells (see Table I), they do occur 

 in all absorbing cells. There is evidence, which I shall consider 

 presently, that some of the small granules of feeble staining capacity 

 may represent a stage in the transformation of the food itself. 



From eight hours onward to one hundred and twenty hours after 

 feeding, albumose granules may be found in greater or le^^s quantity 

 in all the cells. The amount met with in any particular cell will 

 depend upon the accidents of its formation in the lumen. The 

 coui-se it takes, once in the cell, seems to have no distinct refercuce 

 to the organization of the cell, except so far as that may influence 

 it mechanically. Fig. 2, Plate XVI, twenty-four hours, and Fig. 9, 

 fresh, show the granules arranged in rows, following in a general 



