54 Scott and Brebner. — On the Secondary 
Fig. 12 is drawn from a section at 5 mm. from the apex. 
The differentiation of the bundles with metaxylem only, takes 
place rather later, and was studied in other series of sections. 
Returning to our first stem, we find at 23 mm. from the 
apex that the primary development of tissues is completed, 
and we have the normal structure of the central cylinder in 
its fully differentiated condition. This is the stage at which 
development would cease, if we were dealing with an ordinary 
Monocotyledon. Two points, however, must be noticed. Near 
the periphery of the cylinder, bordering directly on the 
pericycle, we still find a few unfinished bundles ; some are 
quite procambial, in others about the proximal half of both 
xylem and phloem is differentiated. Secondly, in the pericycle 
itself we find here and there a very few scattered cells in 
which a recent tangential division has taken place. These 
two points indicate that a further process of development 
is still to follow. Otherwise the structure has been sufficiently 
described at p. 47. At 31 mm. from the apex we found the 
tangential divisions in the pericycle more frequent, and on 
one side of the cylinder the dividing cells formed a continuous 
tangential band, which already suggested a cambium. 
At 41 mm. the cambium was well established in the 
pericycle, on both sides of the cylinder, but not at its ends 
(as seen in transverse section). At this stage we determined 
a fact which was confirmed by many other observations ; 
the tangential divisions are not limited to a single layer 
of pericyclic cells, but two such layers may begin to divide 
simultaneously. Hence, from the very first, there is no single 
initial layer present. It will be remembered that a per- 
fectly similar fact was noticed in the roots of Dracaena (cf. 
Fig. 8). 
At 47 mm. from the apex we found that the development 
had reached a very instructive stage. The formation of 
secondary tissues had just been started. In one place a few 
tracheides had been added to complete one of the unfinished 
primary bundles ; at another a new, altogether secondary 
bundle had been started. In fact, after the long pause, tissue- 
