

apparenil;; must tiike part in i.iic ioniution oi ;ho ctalk 

 of the inciusinm. Tlie dcnnatofron oi' boMi llieso sections 

 grows very rapidly in a radial direction, keeping P'-'^ce 

 v;it]i the lengthening of the- sonis, and giving rise thus 

 to a part of the indusiitm surrounding the latter (Figs. 

 51 - n4), becoming finally more tlian half the v/idth of tlie 

 capsule (i.ind,,o.ind,. Fig, 55). The outer end of these 

 layers later spread out peripherally to meet the ventral 

 end of section VI and thus enclose the cells of the so- 

 ruG. This portion of each of these tv;o sections becomes 

 several cells in thiclmess (Figs, 54, 55), and gives rise 

 to a part at least of the ventral portion of the gelatinous 

 ring, v;hilo the inner portions (o.ind«,i.ind.. Fig, 54) 

 remciin one cell thick even till matiu'ity, Tliis latter 

 statement, however, is strictly true only of the basicopic 

 portions of those segments, as we shall see later, 



Tlie Sporangia. 



It remains to describe the developnont of the most 

 important division of the soral sef^onts, the basicopic 

 ultimate marginal cell. This has a very interesting and 

 significant history, since each is ihe mo' her cell of all 

 the sporangia of the sorus formed in its segment. It is 

 the 'sorus mo '.her cell" of Bi^sgon, but as it give 



