i8o 
Journal of Agricultural Research 
Vol. VIII, No. s 
not hitherto described their component parts. They are within the 
vascular cylinder of the tumor in what I take to be modified pith and 
they also have the same order and arrangement of parts as the Ricinus 
steles just described—viz, spiral vessels on thdr periphery followed by 
tracheae and cambium within which are sieve tubes, and sometimes in 
the larger ones a small amount of the fundamental tissue forming a 
coarse-celled center. 
I observed the same phenomenon in another Ridnus stem the pith 
cavity of which had recdved ammonia. Here no less than dght con¬ 
centric medullary bundles (phloem strands surrounded by xylem, as 
in crowngalls) developed in the outer pith near the vascular cylinder 
(PI. 44 and 45) at one level and 16 at another level in a circumference 
of less than 1 cm. In this connection, however, see Plate 65. 
Furthermore, in studying the outer part of the same cross sections 
(middle part of the Ricinus internode shown on Plate 5) I discovered 
under the dead pith cells at either side of the proliferating cells, shown 
in the center of the picture, cushions of proliferating tissue derived from 
the inner (parenchymatic) face of the uninjured xylem bundles and in 
the innermost part of these cushions close to the layer of killed tissue 
a small group of phloem cells under which (that is, farther from the pith 
cavity) are spiral vessels, with what appears to be cambium between 
them (PI. 46). In other words, here is another example of a pathologi¬ 
cally induced second xylem-phloem cylinder, much more imperfect than 
that produced by the ammonium phosphate but oriented in the same 
way—that is, it is the reverse or mirror image of the normal xylem- 
phloem, but is separated from the latter by a considerable abnormal 
growth of parenchyma. In Plate 47, figures 1 and 2 show cross sections 
of the central strongly proliferating part of Plate 5. Only tissues near 
its outer part are here shown, but in them are tracheae and twisted cells 
strongly suggestive of crowngall distortions, while deeper in (PI. 45, 
fig. 1) are the normal elements of the bundle. Calcium-oxalate crystals 
are very numerous in and near the killed pith tissues, are not elsewhere 
present in the sections, and have never been seen by the writer in normal 
Ricinus pith of this age. This suggests that normally in dividing cells 
ammonia is perhaps the first term in a series of reactions leading, in the 
plant, to the production of oxalic acid, which, as we know, is an almost 
universal product of its growth. 
In one case also where limewater was injected I succeeded, in the 
lower third of an internode, in closing the pith cavity altogether with 
normal pith proliferations in which no wound tissue is visible (PI. 14, 
fig. 3), but no second cylinder of xylem-phloem was formed in it; nor 
have I thus far succeeded in reproducing the phenomenon with the 
monobasic ammonium phosphate. My later experiments, I now believe, 
were done on internodes which were too old. 
