MOSSES AND LIVERWORTS 113 



The spores arise from a single subepidermal layer 

 of cells (Fig. 28, D), which later becomes deeper 

 seated through the further division of the superficial 

 cells (E). Within the sporogenous layer, or arche- 

 sporium, is a central cylinder of sterile cells forming 

 the "columella," which both in origin and position 

 seems to represent the axial vascular bundle or strand 

 of conducting cells found in the young sporophyte of 

 the ferns, and it is not impossible that it may also serve 

 as a conducting tissue, thus representing a primitive 

 vascular bundle physiologically as well as structurally. 

 Owing to the absence of a root connecting the sporo- 

 phyte with the earth, it remains dependent upon the 

 gametophyte for its supply of water and also for cer- 

 tain food elements, and if the gametophyte perishes, the 

 sporophyte necessarily soon dies as the supply of water 

 is cut off. Otherwise, owing to the perfect assimilative 

 system, it is quite independent, and if a root were pres- 

 ent would be entirely so. 



THE TRUE MOSSES (Musci) 



The second class of the Bryophytes, while greatly 

 outnumbering the liverworts, shows very much less 

 range of structure and is evidently a much more spe- 

 cialized group. These "True Mosses," with few ex- 

 ceptions, show an almost stereotyped plan of structure, 

 the differences between them being mostly of minor 

 importance. There are a few, however, notably the 

 peat-mosses (Sphagnacese), which show affinities with 

 the liverworts, especially with Anthoceros. 



The gametophyte of the Musci usually exhibits two 



