112 BOTANICAL GAZETTE [FEBRUARY 
in all cases seem to have been reached without the sure evidence 
of the mitotic figure. 
Having shown thus that some cells are added to the axial row 
by segments cut from the lower face of the terminal cell, it must 
next be determined whether all or only part of the remainder are 
thus formed, intercalary divisions being the alternative. Here 
too the evidence of the mitotic figures will be appreciated, for if 
found in the row intercalary growth can hardly be questioned. 
Fortunately these were found, and four such cases are shown 
(figs. 24, 31-33), proving that there are additions to the row by 
intercalary divisions. 
How long the terminal cell continues to act as an apical cell, 
and whether the cutting off of canal cells ceases before the 
cutting off of peripheral cells are questions to be asked at this 
point. The number of cells constituting the row is variable. 
That the cutting off of cells from the lower face of the terminal 
cell ceases early, in some cases at least, is evident from jigs. 24; 
32, 33. Here a longitudinal wall through the center of each 
terminal cell clearly prevents any further additions to the axial 
row from this source. The presence of such a wall in the ter- 
‘minal cell when the number of cells composing the row was only 
eight or ten is very common; that in fig. 2¢ is exceptionally 
early. There is no evidence to show that addition to the row 
from the terminal cell continues in any archegonium later than 
the age indicated in figs. 32~34, and much to show that it does 
not. Goebel (18, p. 244, fig. 138, 1v) reports one case where 
such a division has occurred in a row of eighteen or more cells, 
but the absence of a mitotic figure leaves some room for question. 
On the other hand, the production of neck cells by the tet 
minal cell continues till late in the development of the arche- 
gonium, but the last divisions for this purpose are always inter- 
calary (fig. 30). In many cases the number of cells in the axial 
row is noticeably less than those in the peripheral rows which 
inclose the canal row. There does not appear to be any relation 
between neck rows and canal row in this regard, for in other 
cases the numbers are nearly equal, and in a few cases those of 
the neck rows were less than those of the axial row. 
