58 DE. D. H. SCOTT ON A NEW TYPE OF MEDULLOSEZX 
The elements of the bundle as they appear in longitudinal section are shown in Pl. 10. 
fig. 18, taken from one of the larger strands, a near neighbour of that shown in 
fig. 14, and of similar structure. The various forms of tracheide, from the large pitted 
elements to the spiral protoxylem, are well shown, as well as the parenchyma lying 
between them. 
As regards the phloem there is a rather striking difference between the stele and 
meristeles on the one hand and the leaf-trace bundles on the other. In the former, as 
we have seen, the phloem-elements are generally narrow, and the sieve-tubes are by no 
means conspicuous in transverse section. The leaf-trace bundles, however (and the 
same applies to those in the petioles), show a ring of large elements in the phloem 
(figs. 14 & 15, s£), which strike the eye even under a slight magnification (fig. 16). 
Their diameter varies in different bundles; in that shown in fig. 15 the average is about 
100 u. These elements are usually separated from one another by small cells, but 
sometimes two or more occur in contact. Longitudinal sections show them simply as 
long tubes, and I have not been able to detect any characteristic markings on their 
walls. In the unilateral bundles of the stem the phloem is usually preserved on one 
side only—that remote from the protoxylem. On the other side there is a gap (fig. 15), 
such as one so often finds in the familiar case of Myeloaylon. Some of the petiolar 
strands are better preserved, and we can there see that the large phloem-elements 
extend all round the bundle, lying a little further out opposite the protoxylem, from 
which they are separated by small-celled tissue (see Pl. 10. fig. 22). "Towards this side 
the phloem-elements gradually become smaller. In another case, where two bundles 
were close together, the large phloem-elements appeared to extend in an interrupted 
row from one to the other. | 
The question arises whether these elements were of the nature of sieve-tubes or 
secretory canals. Where they are all scattered, each is surrounded by a ring of small 
cells, recalling the secretory canals of the pericycle and cortex. On the other hand, it 
is not uncommon for two to be in contact, apparently separated by a single cell-wall. 
Further, they are entirely destitute of the dark carbonaceous contents which are so 
characteristic of the canals. 
Another point is the gradation in size of the phloem-elements, as one proceeds from 
the stele to the foliar bundles. The elongated elements (presumably sieve-tubes), 
such as are shown in longitudinal section in fig. 20 (Pl. 10), scarcely exceed 30 pin 
diameter. This is the size characteristic of the stele and meristeles, though somewhat 
larger sieve-tubes may occur. In some of the radially symmetrical leaf-trace strands 
there is little noticeable advance. In others (e. g. Pl. 9. fig. 14) the large elements 
make their appearance, but are only about 50 pin diameter. In the unilateral strands 
of the leaf-bases (e. g. fig. 15) the diameter averages about 100 », while in the bundles - 
of the petiole 130 u is reached (Pl. 10. fig. 22). This gradation certainly suggests that 
the elements in question are of the same nature throughout. 
Further, I am not aware of any case of a vascular 
80 predominant a constituent of the phloem. 
For these reasons I am inclined to regard the large elements of the phloem as sieve- 
strand where secretory canals form 
