354 BULLETIN 82, UNITED STATES NATIONAL MUSEUM. 



sphere which they form enlarges through the growth of the cells in such a way that 

 a free space remains in the interior. By its growth the sacculus is shoved near 

 the periphery and often causes a small external swelling. Reichensperger did not 

 observe at this stage any special epithelium as described by Seeliger. 



With the enlargement of the sphere the nuclei become more indistinct and a 

 nuclear body is no longer recognizable. At the same time fine vesicles appear. 

 The cells have now become enlarged and elongated, so that they are somewhat 

 tongue-like in shape, with the broad bases crowded together and the more slender 

 distal ends projecting into the lumen of the sphere. In these slender and elongated 

 portions of the cells the larger granules are found, and further increase in number. 



At a later stage certain cells everywhere on the periphery, chiefly lying toward 

 the outer side, are seen to produce no granules, but to become narrower and laterally 

 broader. These close together into a thin broad ring about the larger cells, en- 

 capsuling them. 



At the time of the commencement of the building of the terminal thread the 

 nuclei of the cells is moderately large, and in the cytoplasm can be differentiated 

 thread-like cords. 



In the fully developed sacculus the granules appear as if packed together and 

 lie clearly within the cell membrane, which is produced into a thread-like twisted 

 point. At first they are regularly spherical, but they later become flattened on two 

 or more sides and somewhat angled. 



At different developmental stages the granules show the widest variation in 

 their reaction to stains. 



Speaking of the variability in the number of sacculi in different species Reichen- 

 sperger says that in an apparently undescribed species of "Antedon " the sacculi 

 occupy by far the largest place in sections of the arms. They lie closely together, 

 separated only by a thin connective tissue layer, and are of considerable size, dis- 

 placing the other organs of the arm. 



Reichensperger is undecided whether the vacuolar cell complex which Harnann 

 first described in the Comasteridse represents in that family the sacculi of the other 

 comatulid groups. Their corresponding position in the disk and in the arms ap- 

 pears to indicate it. In many sections also it appears as if they opened to the ex- 

 terior through the rupture of their wall in the same way as the sacculi. 



P. H. Carpenter says that in some species of the Comasteridaj individual 

 vesicular bodies resembling the elements of the sacculi are scattered through the 

 ventral perisome, but they are not regularly arranged in groups along the sides of 

 the ambulacra. 



No less than 11 more or less widely different functions have been ascribed to 

 the sacculi. In chronological order they have been considered as 



(1) Excretory organs, secreting a reddish fluid, especially at the period of 

 sexual maturity (Dujardin and Heusinger). 



(2) Ovaries (Forbes). 



(3) Excretory organs (Perrier, Ludwig, Chadwick. and Reichensperger in 

 his more recent contributions). 



