RG en 
Knox: Stem oF IBERVILLEA SONORAE 339 
It is treated by Leisering (29), deBary (6), Morot (18), Hérail (19), 
VanTieghem (24), and Weiss (15). Weiss pictures such bundles 
in the root of Bryonia, where masses of leptome are found to be 
associated with hadrome elements. He concludes that tracheae 
are separated by dilatation-parenchyma from the primary masses 
of the hadrome, and serve as centers of meristematic division in the 
ground-parenchyma, while the semicircular meristem gives rise to 
the leptome. The meristematic parenchyma:cells are said to come 
originally from the primary cambium. Scott and Brebner (25) de- 
scribe their work with TZhladiantha. They find, elaborating 
Dutailly’s treatment of the same form (7), the elements of the 
hadrome separated out by dilatation-changes, but differ from Weiss 
in finding that the leptome bears no constant relation to the 
tracheae, and that it is partly formed directly from the cambium, 
This work in each case is on the root. In /bervillea the general 
dilatation extends to the unlignified parenchyma of the hadrome, 
and of the parenchyma adjacent to the bundle. Such parenchyma 
lateral to the hadrome begins to divide and sometimes tracheaé are 
“nipped” off by the changes and isolated with the meristematic 
cell. In either case the formation of new cells in the dilatation 
pushes the meristem out into the medullary ray. The meristem 
then produces leptome but never hadrome, and as the stem in- 
creases in size the course of the bundles becomes irregular, so that 
it is often easy to see the sieve-tubes in longitudinal position. 
Figure 12 shows such a meristem in the midst of a starch-filled 
parenchyma. — 
One of the characters of the older secondary stems is the 
anastomosing of the bundles. Both the leptome and the hadrome 
frequently run a horizontal course from bundle to bundle, so that 
the center of the stem is a medley of supernumerary sieve-regions 
and it is impossible to trace any regulararrangement. As a whole 
the stem retains the character of its first secondary growth. The 
Parenchma of the medullary rays continues to divide tangentially 
and radially and the increase in pericycle and cortex is through 
the same sort of change. Minute fragments are all that remain of 
the stereome-ring, though opposite these one can still trace the 
Old cortical parenchyma. A relatively large production of phel- 
loderm adds most of the soft tissue in the outer part of the peri- 
