1895.] Lyginodemlron and Heterangium. 199 



C. The Root 



1. Connexion between Root and Stem. We have already briefly 

 recorded onr discovery that " Kaloxylon Hookeri" is the root of 

 Isyginodendron* We have found that certain appendages of the 

 stem of Lyginodendron, most of which were formerly described as 

 " branches," are in reality of endogenous origin, as is shown by the 

 fact that the appendage, in passing through the cortex of the parent 

 stem, is surrounded by a well-defined cortex of its own. These 

 appendages are further shown to be roots, by the structure of their 

 central cylinder and their mode of branching. Sections of the free 

 part of the same organs, which are in connexion with stems of 

 Lyginodendron, show that they agree in all respects with " Kaloxylon 

 Hookeri" namely, in the structure and arrangement of both primary 

 and secondary wood, and in the details of the cortex, which is well 

 characterised hy its double or treble external or epidermal layer. 



We find then that the stem of Lyginodendron bore numerous 

 adventitious roots, of endogenous origin, and that these roots are 

 identical with the fossils previously described under the name of 

 "Kaloxylon Hookeri." 



2. Primary Structure of the Root. All the specimens of " Kaloxylon 

 Hookeri " have been re-examined, and are found to present a perfectly 

 typical root-structure. The stele varies from triarch to octarch 

 structure in different specimens. The protoxylem is external, showing 

 centripetal development of the primary wood. In favourable speci- 

 mens the regular alternation of the phloem-groups with those of 

 xylem is quite clear. The stele has no pith, but there is a consider- 

 able amount of conjunctive parenchyma. Both pericycle and endo- 

 dermis are present. The inner cortex contains abundant " secretory 

 sacs." 



The young roots much resemble the smaller adventitious roots of 

 Marattiaceae. 



3. Secondary Tissues of the Root. These are beautifully preserved, 

 and are found at all stages of development. The cambium is often 

 specially clear. Secondary growth began opposite the phloem-groups, 

 and the secondary wood is generally interrupted by large rays 

 opposite the protoxylem-strauds. The secondary tissues resemble 

 those of the stem. The whole process of secondary growth was per- 

 fectly normal, as in dicotyledons at the present day. 



4. Branching of the Root. The numerous specimens showing 

 branching prove that the rootlets were endogenous, and that they 

 arose opposite the protoxylem-groups of the main root. 



* ' Koy. Soc. Proc.,' vol. 56, 1894. 



