ANATOMY OF CEETAIN SPECIES OF ENCEFHALAKTOs. 455 



and orientation and their more or less rudimentary development. Scott's explanation uf 

 the centripetal xylem in the peduncle of Stangeria is that it is a vestige of a structure 

 such as occurs normally in the vegetative stem of sucli fossil plants as Lyginodendron 

 and Calamopitys Saturni, Ung., and he parallels the occasional occurrcmce in Stawjeria 

 of internal phloem and entire inverted bundles with the case of Seward's LyginoOeudron 

 rohustum, where vascular tissue w ith inverted orientation also occurs on the inner side of 

 the primary xylem, as it also occasionally does in L. Oldhamium. The comparison thus^ 

 instituted I should distinctly support. But, on the other hand, I regard the vascular 

 tissues with inverted orientation, which occur regularly in L. rohustumy Sew., and 

 irregularly in L, Oldhamium, not as sportive and utterly abnormal, but as a reversion to 

 the typical ancestral condition. That the vascular stem-structure of Lyginodendron is- 

 really composed of the vestiges of a ring of concentrically-constructed ^ivmnh is distinctly 

 shown, as in the peduncle of Stangeria, by the curved contour of the bundles composing 

 the cylinder in L. Oldhamium. I am very far from regarding the stem-structure of 

 this plant (like Scott does) as derived from that of Heterangium, this latter being, 

 with its single large stele, really a variant on that of Mednllosa w ith its ring of several 

 steles, just as is also the case with Colpoxylon, which latter plant may, in fact, be 

 regarded as a Mednllosa possessing one or two large steles or concentric strands instead 

 of a number of small ones. Such forms as Heterangiiim are not necessarily the most 

 primitive; ^ polystelic member of the Cycado-tilices, Cladoxylon, is found at a geological 

 horizon quite as low as that at which Meterangiimi occurs. The latter plant probably 

 represents a distinct phylum of development. Lyginodendron more nearly resembles- 

 the typical Medulloseie and existing Cycads, its structure consisting, as in both these 

 groups, of a ring of steles or vestiges of such. Its ancestors never bid prhnary 

 tracheides occupying the so-called " j)ith." 



Scott says again : — " Extrafascicular zones occur in the same form in some MedulloseiB 

 as in certain recent Cycads, so it appears unnecessary to derive this part of the structure 

 from a reduced system of rings." Eut this part of his argument carries but little weight, 

 for it appears almost certain (is, at least, highly probable, in view^ of the transitional 

 structures found in other species of Medullosa) that the structm-e referred to has itself 

 been derived froni the dclinite polystelic arrangement of its own more internal strands 

 and of those of the more typical forms. 



I may add that valuable evidence for the origin of the stem-structure of Cycads from 

 that of the Medullosese has been observed in the roots of several genera— e. g., Cycas 

 revoluta, Thunb., C Seemauni, A. Br., Encephalartos (described above), and Macro- 

 zamia,— viz., in the upper or hypocotyledonary part of the organ near the transitional 

 region between stem and root, i would draw very special attention to the fact (for ou 

 this depends much of the weight of my whole argument) that this is the region par 

 excellence where the jirst-formed tissues are situated and therefore where ancestral 

 characters would be sure to preponderate. The vascular tissue arismg outside the 

 central stele in this region assumes the lorm, when it is first laid down, of, in some cases, 

 perlectly concentric, in others partially concentric, strands or portions of such. These 

 structures in the root are only found in those genera with more than one vascular ring 



3z2 



