1902.] NATURAL SCIENCES OF PHILADELPHIA. 343 



affected, exhibiting a broAvn color, plainly the characteristic osmic 

 color. Those in the base of the large cells, on the other hand, take 

 the stains. Thus in Nos. 21, 25, etc., stained in irou-hsema- 

 toxylin, they appear brown and black respectively, while in No. 35, 

 stained in acid-fuchsiue, brown and red. After picro-formalin 

 also (No. 16) those in the large cells stain red in Biondi-Heiden- 

 hain, while those in the small cells remain unstained. 



In Nos. 1 and 2 there is a complete transition from one color to 

 the other, some retaining no stain, others retaining only a light 

 shade or only a peripheral ring of black, still others holding it 

 densely. Moreover, the transition does not stop at granules of the 

 same size ; among those in the large cells which hold the stain most 

 tenaciously thei'e is a perfect gradation from small granules to large 

 globular masses (fig. 20). The latter correspond to the contents of 

 the vacuoles mentioned above (fig. 22). We have evidence, 

 therefore, that the two kinds of cells are, as Frenzel holds, but the 

 young and mature phases of the same kind. Our evidence goes a 

 step farther. Frenzel found both zymogen granules and fat globules 

 in the same cell, and from this concluded that all the cells produce 

 both, the ferment during the early life and fat later. Now we 

 have been able to trace a complete transition from the zymogen 

 granules to the large globules, merely by securing a good fixation for 

 all the cell constituents. The indication is, therefore, that we have 

 to do not with two distinct products, but with different stages in the 

 formation of a single product. 



Before going farther with the present discussion it will be neces- 

 sary to present the changes which the cells undergo in secretion. 

 Fig. 21 represents a cross-section of No. 10 (Table I) preserved 

 twelve hours after feeding — this following a fast of eleven days. 

 In this case the large globules are not preserved, only their vacuoles 

 being seen. At the bottom of the cells are masses of zymogen 

 granules, some of which are becoming less distinct in outline, others 

 are represented merely by a dense mass of small granules. Fig. 

 20 (No. 20), sixteen hours, after a fast of twenty-one days, shows 

 the large globules preserved. In the former case, as well as in the 

 latter, the ends of some cells have broken down and are undergoing 

 a process of disintegration. Sometimes the whole end of the cell is 

 involved in this destruction, or the end may break up into large or 

 small fragments, or finally break off as a whole, and the larger or 

 smaller pieces are then found in the lumen as far down as the canal 



