S20 



MOSSES AND FERNS 



CHAP. 



Pfeffer's figures, the cotyledons do not develop their vascular 

 bundles until later. The early growth in length of the root 

 is mainly intercalary, as the divisions in the apical cell for some 

 time are not very rapid, and for a long time the root-cap con- 

 sists only of the two original layers. 



With the growth of the embryo the cell-formation in the 

 lower part of the spore continues until it is filled with a contin- 

 uous large-celled tissue, the contents of whose cells are much 

 less granular than the undivided regions of the spore, and as 

 the embryo develops, the foot crowds more and more upon them 



until it nearly fills the 

 spore cavity. 



On comparing Pfeffer's 

 account of v$\ Martensii 

 with my own observations 

 upon 5*. Kraussiana, the 

 main differences consist 

 first in the smaller devel- 

 opment in the latter of 

 the primary prothallium, 

 i. e., the prothallial tissue 

 formed before the spores 

 are shed, the archegonia 

 being only separated from 

 the diaphragm by a single 

 layer of cells instead of by 

 three or four, as in 5*. 

 Martensii. L. apus, which 

 FIG. 300. Longitudinal section of a fully- was also examined by the 



developed prothallium of S. Kraussiana wr j ter j g intermediate ill 

 with an advanced embryo (em), X?7', I, 



Hguia. this respect between the 



two. A second difference 



is the later period at which the cell division in the lower part of 

 the prothallium is completed in S. Kraussiana. In this species, 

 too, no rhizoids were seen, while Pfeffer observed them in 5\ 

 Martensii. Finally, in the latter the suspensor is much shorter 

 and straighter than in ^. Kraussiana. Miss Lyon (2) found 

 that in ,S\ apus no suspensor was formed, but the development 

 of the embryo is not described. 



In >S\ Martensii, almost as soon as the cotyledons are estab- 

 lished, the two-sided apical cell of the stem is replaced by a 



