Tissues in Certain Monocotyledons . 39 
continuous, though it may be so for long distances. That 
the divisions should pay no respect to the morphological 
distinction between stele and cortex cannot surprise us 1 . 
The case where the cambium at once appears outside the 
endodermis is more puzzling. Here both pericycle and 
endodermis may be very thick-walled, and, so far as the 
transverse section shows, there may be no interruption to their 
continuity. The secondary tissues therefore are from the 
first cut off from any direct communication with the primary 
cylinder at the same level. In this case it is by no means 
always the first cortical layer which divides ; sometimes it is 
even the fourth layer from the endodermis in which the 
first divisions appear. 
It is impossible to understand this type of secondary 
growth unless we trace the course of the cambium and its 
products in the longitudinal direction. The statements in the 
literature appear to show that the secondary tissues taper off 
regularly towards the root-apex, their maximum thickness 
being at the base of the root, where the process presumably 
started in the first instance. It has also been noticed that 
the thickening is almost always highly excentric, the upper 
side of the approximately horizontal root, according to 
Strasburger, receiving the first and greatest increment. The 
thickening near the apex is said to be always pericyclic, 
cortical cambium first appearing in a more advanced region. 
Although this account of the process is no doubt applicable 
to certain cases, or it may be to certain species, we have not 
found it confirmed by our observations on Dracaena fragrans 
and D. Draco. 
We only found one instance in which the secondary growth 
appeared to have started from the base only of the adven- 
titious root and advanced regularly towards its apex. This 
was in D. fragrans, and in this case the thickening remained 
pericyclic throughout. In all other thickened roots of these 
two species which we investigated (and in D. fragrans they 
were fairly numerous) the maximum thickness of the secondary 
1 Cf. Scott and Brebner, On Internal Phloem, Annals of Botany, vol. v. p. 287. 
