110 MR. W. C. WORSDELL ON THE 
more than merely to supplement that treatise by one or two short additional remarks. 
The seedlings I examined (at a date prior to the appearance of Mr. Pearson’s paper) 
were slightly younger than those of that author. 
The bundles in the lower part of the cotyledon, which is still enclosed within the seed, 
exhibit a very decided tendency towards a concentric structure, as does also one of those 
in each cotyledon in the uppermost portion of the latter. The bundles are curved in 
contour so as to be quite three-fourths concentric. Fig. 1 (Pl. 15) represents such a bundle 
which exhibits a central protoxylem ; the tracheides 4? I regard as having been cut 
off by a cambium whose cells are too indistinct to be drawn accurately, but which 
appear to extend entirely rouud the bundle, as I have endeavoured to indicate in the 
drawing. Although the bundles in the cotyledons of all Cycads are more or less curved 
or horseshoe-shaped in contour, those of Bowenia and Stangeria exhibit a much greater 
tendency towards the concentric structure than is the case with the rest. As the 
cotyledon is one of the three sets of primitive foliar organs in which ancestral characters 
are to be first of all sought for, it is a matter of great phylogenetic interest to discover 
in the cotyledons of the above-mentioned two genera this concentrice type of structure in 
the bundles, which I regard as distinctly a primitive ancestral character. The course of 
the two cotyledonary bundles through the cortex of the stem to join the central cylinder 
is also a point worthy of mention. As I have already stated for the same bundles of the 
cotyledons and first scale-leaves of Cycas *, their course is direct, and does not assume a 
girdle-like direction, which is apparently a recently acquired character. The direct 
radial course of the bundles is the ancestral one, and is exhibited not only in connection 
with these most primitive of vegetative foliar organs in the seedlings of modern Cyeads, 
but as a normal character throughout the vegetative axis of the Wealden Cycad 
Bennettites, as also in the more ancient Medullosexe and Lyginodendre:e. 
Older Seedlings. 
Two-year-old plants of a varietal form (var. serrulata) grown by Messrs. Sander of 
St. Albans, and obtained from them by the authorities of the Royal Gardens, Kew, 
were carefully investigated. 
In the extremely short young stem the central cylinder contains about three stout, 
bundles, consisting of a great deal of secondary tissue and a large amount of parenchyma. 
Leaf-trace bundles occur in the cortex. Lower down, just beyond the transitional 
region between stem and root, or primary node, and where root-structure already 
prevails, the central stele is of considerable width, and the constituent bundles farther 
apart, owing to the expansion of the parenchymatous tissues. The large bundles 
composing the secondary tissues of the stele are now in places connected by delicate, — 
few-layered. bands of secondary vascular tissue, whose course is rather difficult to ` 
follow. 
Outside the central stele are seen, in transverse section, other strands, which, there 
ean be no doubt, represent the extrafascicular vascular tissue of Cycas, Encephalartos, 
* « The Comparative Anatomy of certain Genera of the Cycadacee,” Journ. Linn, Soc., Bot. vol, xxxiii. 1895, 
pp. 437-457. 
