302 MOSSES AND FERNS chap. 



In all the Hepaticse the axial row of cells of the archegonuim 

 terminates in the cover cell, which by cross-divisions forms the 

 group of stigmatic cells of the neck. In the Anthocerotes this 

 terminal group of cells is the only part of the archegonium neck 

 that is free, the lateral neck cells being completely fused with 

 the surrounding tissue. This arises from the archegonium 

 mother cell not projecting at all, but we have seen that in cross- 

 section a similar arrangement of the cells is presented to that 

 found in the young archegonium of other Hepaticse. In the 

 Filicineae a similar state of affairs exists, but the divisions in the 

 mother cell are, as a rule, not so irregular. Still, e. g., Marattia, 

 it is sometimes easy to see that the mother cell (so-called) of 

 the archegonium is triangular when seen in cross-section, and 

 cut out by intersecting walls in exactly the same way as the 

 axial cell in the Bryophyte archegonium. In short, what is 

 ordinarily called the mother cell of the archegonium in the Ferns 

 is really homologous with the axial cell only of the young 

 archegonium of a Liverwort. A comparison of longitudinal 

 sections of the young archegonium of Marattia, for instance, 

 with that of Notothylas, will show this clearly. From this it 

 follows that the four-rowed neck of the Pteridophyte arche- 

 gonium does not correspond to the six-rowed neck of the 

 Bryophyte archegonium, but only to the group of cells formed 

 from the primary cover cell, and is a further development of this. 

 The relatively long neck of the archegonium in the more special- 

 ised forms, e. g., Botrychium Virginianum, and especially the 

 leptosporangiate Ferns, must be regarded as a secondary de- 

 velopment connected probably with fertilisation. The shifting 

 of the archegonium to the lower surface of the gametophyte has 

 probably a similar significance. In B. Virginianum, however, 

 the archegonia are borne normally upon the upper side of the 

 thallus, as in the thallose Liverworts. 



It is possible that a similar relation exists between the 

 antheridia of the eusporangiate Ferns and that of the Antho- 

 cerotes. In both cases the formation of the antheridium begins 

 by the division of a superficial cell into a cover cell and a central 

 one. The former divides only by vertical walls in the Marat- 

 tiacese, but in Botrychium and the Anthocerotes it becomes 

 two-layered. In the latter the central cell may form a single 

 antheridium, or it may produce a group of antheridia, but in 

 the others it diyideg at once into a mass of sperrn cells. By the 



