1904] CHRYSLER—CENTRAL CYLINDER 163 
3. Does the structure of the young stele throw any light on the 
question of the origin of the monocotyledons ? 
ARACEAE. 
The number of forms which have been available in this inquiry 
has not been large, but they are sufficiently varied in their affinities, 
and appear to the writer to yield no uncertain result. 
Pothoideae. This subfamily is regarded by ENGLER (1) as the 
most primitive one in the family. Hence one of its most available 
representatives will be first described. 
Acorus CaLamus.?—In a seedling of this plant the central cylinder 
in its lowest region is a solid mass of vascular tissue, consisting of a 
core of xylem and a ring of phloem, surrounded by pericycle and 
endodermis, that is, it is a typical protostele. One trace is given off 
to the cotyledon, and usually three traces to each of the next three 
or four leaves, after which the number of foliar traces is increased. 
In the region where the traces of the second leaf are given off, the 
central cylinder is seen to possess a parenchymatous pith, which is 
continuous with the pericycle through the gaps in the vascular tissues 
caused by the bending out of the traces (fig. 1, which, however, 
represents a higher region of the stem). The endodermis does not 
nd inward through the gap with the pericycle, but remains 
unbroken, a portion of it surrounding the trace as it passes outward. 
Followed downward through the stele the pith either becomes nar- 
Tower and disappears above the point of exit of the cotyledonary trace, 
or in some cases enlarges at this point and communicates with the 
pericycle at the higher node; followed upward the pith widens out 
with the enlarging central cylinder. As the three traces of the third 
leaf bend outward, the pith again communicates with the pericycle; 
Since the median trace is the largest of the three, the gap it leaves in 
the vascular ring is the widest; in fact, the gap of one or both of the 
lateral traces may be filled only by a single row of parenchymatous 
cells or may not be present. Up to this point the vascular ring is 
Practically continuous, owing to the foliar gaps being so short, but 
in the higher regions of the young stem the gaps remain open longer, 
so that the central cylinder appears to be made up of a ring of sepa- 
*The nomenclature employed in this paper is that of Engler and Prantl. 
