258 , ERNST F. ARTSCHWAGER 
as such, since they mature into vascular tissue. The ring of cambium is 
thus broken up. 
The cells of the cambium which are not concerned with the formation 
of these phloem groups are dividing very actively meanwhile, so that the 
cambium ring becomes undulate, the phloem groups occupying the depres- 
sions. Very soon after the initials of these phloem groups appear, a cam- 
bium forms in the parenchyma adjacent to the outer face of these groups 
(fig. 2, B). This new cambium becomes connected laterally with the cam- 
bium ring of the vascular cylinder. Sometimes the appearance of these new 
cambium segments is belated, and not until the unequal activity of the 
cambium ring has produced the undulate appearance and the depressions 
does a cambium layer appear at the outer face of each new group (fig. 3) . 
In such a case the cambium formation begins at the margin of the depres- 
sion, advances laterally, and when united undergoes reciprocal tangential 
division, thus giving rise to xylem and parenchyma which mature in the 
normal manner. 
The formation of new cambium initials which mature into groups of 
phloem completes the growth of a zone of thickening which is succeeded 
by a new similar zone. The same process is repeated, and thus arise the 
undulate zones of vascular tissue so characteristic of the members of the 
Chenopodiaceae . 
Conclusions 
The study of the anatomical features of the vascular tissue of the stem 
in part confirms and in part modifies and extends the results obtained by 
earlier investigations. In the discussion of the histological features of the 
phloem it was shown that sieve tubes and not phloem parenchyma make up 
the larger portion of that tissue. Why earlier investigations limit or even 
deny their occurrence is hard to understand. Even Gheorghieff in his 
detailed anatomical researches of the group simply states: ''Die Phloem- 
partie ist vorwiegend aus parenchymatischen Elementen zusammenge- 
setzt. Siebrohren habe ich nur selten gefunden." It must be admitted, 
however, that the elements of the phloem are comparatively small, and that 
the sieve tubes especially are narrow and easily mistaken for plasma-rich 
cambiform elements unless staining reactions show the sieve plates or the 
callus deposits over the plates. The typical staining reaction of this 
substance is a further aid in identifying the sieve tubes. 
The elements making up the conjunctive tissue exhibit such a variety 
in form and arrangement that they could not be conceived of as ray cells 
in the morphological sense. That they may function, however, as rays is 
not at all unlikely. 
Above all, however, this study has shown that the anomalous growth 
of the stem is produced by a periodically acting cambium which is pro- 
gressively renewed at places where new phloem groups originate. In the 
