PALEOBOTANY AND EVOLUTION. 115 



ourselves as having a tall, upright stem rising to a height of 

 several feet and bearing somewhat remote, spirally arranged, 

 highly compound, fern-like leaves. The base of the stem, 

 where the adventitious roots were given off, must have been 

 buried for some depth in the earth or mud. . . . The stem 

 is mechanically well constructed, and thus fitted to maintain 

 an upright position while bearing abundant foliage. We 

 have no distinct evidence for any rhizome-like organs, but 

 it is possible that some of the small shoots which are rather 

 deficient in sclerenchyma may have been creeping or sub- 

 terranean." Certain interesting examples of anomalous 

 structures in Lyginodendj'-oiL stems have been figured and 

 described, but it is impossible to deal with these in the 

 present article. 



Affinities. The leaf-trace bundle system of the stem 

 agrees in certain respects with that of Osjnimda. In both 

 genera the stele possesses a ring of collateral primary 

 bundles, and in both there is good evidence for believing 

 that in the lower stem internodes the xylem has the form 

 of a continuous ring enclosing a central pith. Another 

 point of contact between the fossil plant and recent ferns is 

 afforded by the Gleicheniaceous genus Platyzoma, in which 

 the monostelic stem possesses a distinct pith. The proto- 

 xylem of Lyginodendro7i is of the same type as that of the 

 foliar bundles of recent cycadean petioles, and in some cases 

 also of the stem bundles of Cycads. The term " mesarch," 

 proposed by Solms-Laubach, has been adopted for this 

 particular form of bundle. The characteristic feature of 

 mesarch bundle, is that the protoxylem is not on the inner 

 or outer limit of the primary xylem tracheae, but occupies 

 an intermediate position between the primary centripetal 

 and primary centifugal xylem elements. 



The secondary wood and bast are similar in a striking 

 degree to the corresponding tissues in cycadean stems. As 

 regards the cortex, the emergences of the peripheral tissues 

 suggest those of Ahophila and other recent ferns. Passing 

 to the petiole, we lose sight of the more distinctly cycadean 

 affinities, and find an obvious correspondence with the 

 anatomy of recent ferns. The concentric bundle is clearly 



