1891.] Notes on Osmunda and Botrychhtm. 39 



1 



tions (figs. 2, 3), the adjacent cells are sometimes of nearly the 

 same size and shape as the apical cell; but this position gen- 

 erally shows plainly that they are derived from it, and the 

 relation of the young segments to the apical cell may be 

 readily made out (fig. 2). It is, of course, not impossible 

 that here, as in Todea, there may be sometimes four initial 

 cells, but this view is not supported by my own observations. 

 Owing to the slowness of cell-division in the young seg- 

 ments, and the comparative irregularity of the same, it is 

 difficult to trace the limits of the segments beyond the 



youngest ones. 



The apical cell, as in O. regalis, is deeper than in most 

 Filicineae, but from its faces the segments are cut off much as 

 in the latter. The segments usually form a spiral, although 

 cases were met with that looked as if this were not the case, 

 but that they arose in pairs from opposite sides of the four- 

 sided initial-cell. From the outer face segments are also cut 

 off, and these contribute to the growth of the root-cap, but 

 this is also formed in part from cells cut off from the young 



lateral segments (fig. 1). 



While the earlier divisions of the young segments do not 

 follow any absolute rule, nevertheless they correspond in the 

 main to those observed in the other Filices. The first wall 

 in the young segment usually divides it radially into two cells, 

 one of which is deeper than the other, as the wall does not 

 exactly bisect the segment. This is followed in each semi- 

 segment by a transverse wall that separates an inner small 

 cell from an outer larger one. The former, as in the other 

 ferns, probably gives rise only to the plerome-cylinder; the 

 latter to periblem and dermatogen, and in part also to the 

 root -cap (fig. 1). As the root-cap is derived both from the 

 outer segments of the apical cell and the outer cells of the 

 lateral segments, its inner layers lack that regularity that is 

 so marked in most ferns; but as the cells enlarge this irregu- 

 larity is in great part lost, and the outer layers of cells show 

 a stratified appearance, nearly as regular as in the other Filices. 



The plerome cylinder is in this species especially large, 

 somewhat oval in section, but with poorly defined limits, so 

 that it is not possible to state positively whether or not it 

 owes its origin exclusively to the innermost cells of the seg- 

 ments. Occupying the center are several very wide cells that 

 early cease to divide and are very conspicuous. These are 



