Johnson — A Seed-bearing trish Pteridosperin. 3 



summarized account in Scott's "Studies in Fossil Botany" (vol. ii, 1909), in 

 which work references to the general literature of the subject will be found. 



Lygiiiodendron possessed a branching stem not more tlian 4 cm. in 

 diameter, armed with liook-like emergences, like a scrambling bramble. The 

 leaves were large and compound, being bi- to tri-pinnate. Emergences were 

 present on the leaf in all its parts, and appear in the pinnule impressions as 

 tubercles, occasionally, years ago, mistaken for sori. The pinnules were 

 slightly stalked, and divided into three or four conchoidal, rounded, slightly 

 lobed segments. A main vein enters each pinnule segment, passes througli it, 

 sending off branches at a wide angle, which, like it, may on tlieir way to the 

 edge of the segment brancli, i.e. the venation is pinnate with diverging 

 branches. The stem is characterized by a vertical reticulum of sclerotic bands 

 running in the outer cortex. The vascular system is most interesting. The 

 stem shows an interrupted ring of vascular bundles, enclosing a pith. Each 

 bundle is collateral and mesarch, as is the vascular bundle of the Gycad petiole, 

 and, in some cases, of the Gycad peduncle also. The bundle is open so that 

 secondary thickening occurred, as in Gycads. Both raetaxylem and secondary 

 xylem show trachese witli multiseriate bordered pits, also as in Gycads. Thus, 

 while the arrangement of the bundles is like that of such a fern as Osmunda, 

 the other characters are Gycadean. The leaf-trace system is also Gycadean 

 in its course, but becomes fern-like in structure and in tlie mode of branching 

 in the foliage. 



The venation of the pinnule is very like that of Tric/iomanes radicans, 

 without serving in this case as a sign of affinity. 



The adventitious roots, Kaloxylon Hookeri, have a primary structure like 

 that of a root of Angiopteris evecta, or otlier Marattiaceous Fern, but show, 

 superadded to this, secondary thickening as seen in the root of a Gycad. 



Thus the vegetative organs alone indicated clearly, as is now generally 

 admitted, that Lyginodendron was a member of the Cycado-Filices. For some 

 years this new group rested for its justification on the characters of the 

 vegetative organs of its representatives. It was not until 1903 that any 

 knowledge of its reproductive organs was obtained. 



Eeprodttctive Organs. 

 il/rtfc.— A species of CrossoUieca, founded in 1883 by Zeiller as a dimorphic 

 Marattiaceous genus, possessing sterile Sphenopteris-like fronds and peculiar 

 fertile ones, was shown by Kidston in 1905 to be the male state of Lygino- 

 dendron, and was named by him Grossotheca Honinghausi.^ A fertile male 



' This name will probably replace the now well-established name Zt/ffinodendroit OUhamium, 

 "Willni,, for our fossil, owing to Kidston's discovery. 



a2 



