164 H. HOLTER 



have learned from Dr. Porter's review that other interpretations are 

 possible) then it would certainly seem to indicate the existence of a 

 spatial continuity between pinocy tic vesicles and the endoplasmic reticulum. 

 In Siekevitz's diagram two alternatives are given for the fate of the 

 pinocytic vesicles. One is their fusion with the channels of the endoplasmic 

 reticulum, discussed above, the other is their disappearance into the 

 general cytoplasmic matrix. As we have seen, a third possibility would be 

 their transformation into granular cell organelles, perhaps of the lysosome 

 type. At present we cannot decide if only one of these possibilities is 

 actually realized or if several processes operate at the same time, perhaps 

 with shifting preponderance, according to the metabolic needs of the 



Fig. 5. Stylized representation of a cell, showing a channel of the endo- 

 plasmic reticulum (er) and its relation to the perinuclear space and pinocytic and 

 secretory processes. (After Siekevitz [28].) 



cell. All possibilities are highly interesting, but today the situation is not 

 ripe for too-detailed speculation. 



There is, however, another aspect of the matter that I should like to 

 mention. It is well known that at least in amoeboid cells the process of 

 pinocytosis has its counterpart in the vacuolar discharge of material. This 

 is realized for instance in the filling up of the contractile vacuole [22] by 

 fluid-filled vesicles that fuse with the main vacuole, and also, with certain 

 morphological modifications, in the process of defaecation (Fig. 6). One 

 might say that pinocytosis works both ways, and it has also been necessary 

 to assume processes of inverted pinocytosis in several cases where pino- 

 cytosis was considered to mediate not orily the introduction of substances 

 into cells but also the transport of substances t/irougli cells. As depicted in 



