io6o 
de Fraine. — On the Structure and 
ment centres round the nature of the primary strands, for in agreement 
with Bertrand and Corneille, 1 Chodat regards these in Lyginodendron 
as ‘divergeants du type Qsmunda en w renverse, les deux ailes du di- 
vergeant se rebattent en arriere et finissent par se reunir par leurs poles ’ ; 
this is a not uncommon type in the Ferns, whereas in the stem of 
Cycads no ‘divergeants’ occur, the stem, as in other Gymnosperms, 
being composed of a ‘ couronne de faisceaux endarques The views 
of Chodat thus summarized have been considered at some length by 
Weiss, 2 who concludes that the criticism seems £ sufficiently weighty to 
demand a careful reconsideration of the structure and affinities of 
Lyginodendron ’. 
The theory which would derive the Cycadaceae from the Medullosean 
line has been most fully discussed by Worsdell. 3 He regards the normal 
vascular ring in the Cycad as composed in reality of ‘ the one-sided remnants 
of a number of steles’, while the extrafascicular arcs and concentric bundles 
characteristic of some Cycadean genera are ■ remnants of some ancient 
structure ’ which ‘ consisted of rings or layers of concentric vascular strands ’. 4 
The evidence adduced by Worsdell in favour of the Medullosean ancestry 
of the Cycads is based upon structure observed (i) in the cotyledonary node, 
and (2) in the ‘ flowering’ axis. It is not intended to enter here into a full 
discussion of the evidence afforded by the c flowering ’ axis, but it would 
appear that the incurved and horseshoe-shaped bundles of Stangeria 
described by Worsdell have had far too weighty phylogenetic significance 
attached to them ; 5 it has yet to be proved that far-reaching conclusions of 
such a type can be based, legitimately, upon evidence afforded by a structure 
such as the ‘ peduncle ’, in which special circumstances 6 and physiological 
needs must play an important part. 
The evidence brought forward as supplied by the cotyledonary node is 
that given by the concentric structures around the main axis in certain 
species of Cycas, Encephalartos , Macrozamia , and Bowenia , which give the 
seedling an appearance of ‘polystely’ in this region. The strands are 
1 Bertrand, E. C., et Corneille, F. : La masse libero-ligneuse el^mentaire des Filicinees 
actuelles. Travaux et Memoires de l’Universite de Lille, 1902, t. x, Mem. no. 29. 
2 Weiss, F. E. : Presidential Address to the British Association for the Advancement of 
Science, Section K, Portsmouth, 1911. 
3 Worsdell : loc. cit., 1906, p. 137. 
4 The supporters of the Medullosean ancestry for the Cycads had not, at the time they wrote, 
the advantage of considering a relatively simple member of the group such as Sutcliffia. 
5 It may be pointed out that Scott considers that the mesarch bundles which occur in the 
peduncles of the cones in certain genera of the Cycads, e. g. Stangeria , represent the retention of 
a primitive character by the ‘ floral axis \ Scott, D. H. : The Anatomical Characters presented 
by the Peduncle of the Cycadaceae. Ann. of Bot., vol. xi, 1897 ; and Studies, pp. 365-6. 
6 The observations of Thoday on the inflorescence axis in Gnetum sp. are of considerable 
interest in this connexion. Thoday (Sykes), M. G. : Note on the Infloiescence Axis in Gnetum. 
Ann. of Bot., vol. xxvi, 1912, p. 621. 
