178 Journal of Agricultural Research voi. vm, no. s 
dion, but this is not necessary. Needle wounds without injection— 
that is, where only air entered, gave no such results; neither did the 
injection of pure water. The bark and wood cylinder were not visibly 
affected by the injection—that is, there was no change in color and no distor¬ 
tion or abnormal increase in the diameter of this part of the stem—but 
the pith, in contact with the solution, proliferated into the pith 
cavity enormously as a compact cylinder of small cells narrowing the 
lumen almost to closure, especially at the upper end, the diameter of 
the stem wall being increased to more than double its normal thickness— 
that is, from 3 to 7 mm. The normal Ricinus stem of this age contains 
a large pith cavity and a relatively thin wall. The xylem-phloem cylinder 
in such a stem is well toward the surface, the cortex being thin and the 
pith relatively thick. (PI. 37, fig. 1.) The bundle is not bicollateral. 
Adding the proliferated pith tissue to that normally present gives a 
very thick cylinder of pith in the inner part of which, entirely surround¬ 
ing the pith cavity (source of the stimulus), a second xylem-phloem 
cylinder has developed. (PI. 37, fig. 2.) About the tenth or twelfth 
day death occurred in the middle part of the thickened pith cylinder as 
shown by the appearance on cross sections (PI. 37, fig. 2) of a wide white 
ring midway between the two complete xylem-phloem cylinders, indi¬ 
cating substitution of air for water, and soon after the inner xylem-phloem 
cylinder was tom from the outer part of the pith, probably by cessation 
of growth on the part of the inner pathological cylinder, the stimulus 
from the pith cavity having been exhausted, with continued growth of 
the outer xylem-phloem cylinder. The inner cylinder then lies loose in 
a large cavity, like the skin of a caterpillar in a cocoon, or like a rough- 
coated stick of macaroni (PI. 37, fig. 3), the surface being covered with 
coarse shreds of the dead, white pith. The interior of this cylinder is 
still living and the inner face is covered with a sheet of living, loosely 
connected, rounded, glistening, turgid cells—the final proliferations in 
response to the waning stimulus. The wall of the outer inclosing stem 
has now about the same total thickness and relative arrangement of 
tissues as the whole stem had at the beginning of the experiment (com¬ 
pare fig. 2, 3, and 4 of PI. 37) except that the pith is narrower (approxi¬ 
mately one-fifth cortex, two-fifths xylem-phloem, two-fifths pith); but 
its inner face is covered with coarse white flecks of dead pith and all its 
inner cells are more or less shriveled. The pathological stele in the 
tobacco stem occurs in the outer part of the fundamental connective or 
conjunctive tissue (the cortex); and this pathological stele in Ricinus 
occurs in the pith, another part of the same fundamental tissue. 
This second xylem-phloem contains primitive spirals, tracheae, wood 
fibers, cambium, and sieve-tube tissue, the numerous bundles that com¬ 
pose it being regularly separated by medullary rays. (PI. 38-41.) 
This pathological stele within the normal stem has a reversed polarity— 
that is, its spiral vessels face those of the normal xylem cylinder but are 
