76 MARY J. ROSS. 



as in Figs. 4, 9, 1 2, the more advanced and earlier stages are often 

 seen in the same section. In no stage, however, were the two 

 extremes, i. c., a completed gland and a beginning one, found. 

 In a larva of 16 mm. the stomach is entirely formed, but as yet 

 no glands open on the surface. Figs. S-ii show glands further 

 developed than those just described. By the increase in size of 

 the gland and the multiplication and rearrangement of the cells 



W^>-" : :,^'^ '-.. '^f^--,-':'--.:.:^: :'. <^-vs 



^ipi;|ll?^Sig^i^ 



Fir,. 10. Same as Fig. 8. 



lining the stomach, the excretory duct is being formed (Fig. 9, d}. 

 Before the connection between the fundus of the gland and the 

 excretory duct is established, a lumen is formed in the gland 

 fundament (Figs. 8, 9, /). This appears to be caused by a re- 

 arrangement of the cells and not by a cell disintegration. The 

 connection between the duct and fundus, forming the neck of the 

 gland, is also brought about by a rearrangement of the cells and 

 not by a cell disintegration. In Figs. 12-17 some of the glands 



