24: P. ORPEN BOWER. 
both tangential and transverse.' Still the serial arrange- 
ment may often be followed, and even at the apex of the 
feeder may be seen traces of it, more especially towards 
the middle of the organ (fig. 19, the series are marked 
by letters). There is no fibro-vascular system belonging 
to the feeder itself. The bundles of the hypo-cotyledonary 
stem, curve outwards (fig. 18), but give off no branches. 
At the point of greatest curvature there are to be seen 
irregularities of the reticulated constituents, which seem 
also to have been drawn out laterally (fig. 20). 
We see then that the feeder is a structure formed by the 
lateral extension of the tissues of the hypo-cotyledonary 
stem, accompanied by cell divisions. Its origin is deeper 
than the epidermis, but it has no bundle system of its 
own. It is also an adventitious structure. It is therefore 
to be considered morphologically as an emergence. There 
are no sclerenchymatous elements in the feeder. All its 
tissues, but more especially those at the apex, are thin 
walled, with transparent protoplasm and a fine nucleus. 
The quantity of starch increases as we recede from the 
apex, till close by the fibro-vascular bundle the cells are 
densely filled with it (compare figs. 19 and 20). 
Almost immediately below the feeder the transition from 
the stem to the root type of structure occurs. As described 
by Strasburger for others of the group of the Gymno- 
sperms,” the epidermis in the mature embryo covers the 
hypo-cotyledonary stem; but as we pass towards the apex of 
the root it ceases. The successive outer layers of the root 
cap are the outer covering of the body of the root. In Wel- 
witschia, as in Ephedra, &c., there is no develoment of a 
pseudo-epidermis like that in Taxus. Accordingly, in the 
seedling, since the feeder is developed at the base of the 
hypo-cotyledonary stem, the epidermis is found to cease im- 
mediately below it; and it is even difficult to trace an 
epidermal layer on the under side of the feeder. 
The first change of the arrangement of the fibro-vascular 
system, in passing from the stem to the root, corisists in the 
extension of the phloem portion of the two bundles nearest 
one another. These finally meet and coalesce (fig. 15 @). 
Meanwhile they become separated from the xylem portions. 
These latter also pair off and coalesce to form the two xylem 
masses of the root (fig. 15G,H). Ata point where the phloem 
masses have already fused, but the xylem masses are still 
1 The terms are here used as referring to the axis of the plant. 
2 *Conifere,’ pp. 362, &c. 
