Yd. 6^.~] XEROPHYTIC CHARACTERS OE COAL-PLANTS. 287 



reasonably expect to find some analogous degeneration to be more 

 or less apparent. 1 



Dr. Scott's figs. 83, 86, & 112 (op. tit. pp. 222, 230, & 328) 

 represent sections of the root and rootlet of Stigmaria and Lygino- 

 dendron. They do not indicate any adaptations to water ; but in 

 the frontispiece, illustrating a restored tree of the latter, there are 

 adventitious roots. The conclusion therefore is, from Dr. Scott's 

 illustrations, that the latter genus grew both on dry ground and in 

 marsh or water ; or else was originally a hygrophyte, but retained 

 the adventitious roots on land, just as JPandanus, all palms, and 

 other Monocotyledons have done. 2 



In Dr. Scott's figure of the root of Stigmaria ficoides, there is an 

 outer and inner cortical layer, and a third zone of thin-walled tissue 

 surrounding the stele. I would suggest that this thin-walled 

 cylinder, with the innermost sheath to the xj'lem, was water - 

 storing cellular tissue, protected externally by the inner, cortical, 

 and thick-walled zone. 



In laying stress on the absence of an axial root, as being charac- 

 teristic of aquatic plants, it must be borne in mind that the cause 

 is quite different from that of the absence of an axial root in existing 

 dicotyledonous timber-trees, and presumably in Stigmarias. 



When a dicotyledonous seed germinates in the mud at the 

 bottom of water, the radicle is arrested, in consequence, as sug- 

 gested, of insufficient aeration. Adventitious roots then arise from 

 the nodes in ascending order, the stem assuming the form of an 

 inverted cone. 



In terrestrial dicotyledonous and gymnospermous trees, the secon- 

 dary roots arise from the primary or axial root in descending 

 order, the trunk being broadest at the base, whence the first- 

 formed and largest lateral roots arise. They grow large, at the 

 expense of the primary root, in order to spread out more or less 

 horizontally for superficial moisture and so support the tree. Had 

 the central vertical root kept pace with the trunk of the tree, 

 it would have equalled it in size below ground. This is never the 

 case, and therefore the whole of the morphology is exactly the 

 reverse of that which characterizes an aquatic plant. The root of 

 Stigmaria agrees thus in every respect with the roots of an ordinary 

 timber-tree growing on dry ground ; while the anatomical features 

 are entirely devoid of all aquatic adaptations such as aerenchyma or 

 pneumatophores, as will now be pointed out. 



My attention has been called to a paper by Dr. F. E. Weiss on 

 ' The Vascular Branches of Stigmarian Eootlets ' 3 as apparently 



1 In a paper entitled ' Influence de lAir sec & de l'Air huniide sur la Eorme 

 & sur la Structure des Vegetaux' by Ph. Eberhardt, Ann. des Sci. Nat. ser. 8, 

 ■vol. xviii (1903) p. 61, the author shows how, in every case, the wood and 

 sclerenchyma are greatly arrested under the influence of humid air, and, on the 

 other hand, intensified in yery dry air, as compared with the results of growing 

 normally. 



2 I would refer the reader to my paper on ' A Theoretical Origin of Endogens 

 from Exogens, through Self- Adaptation to an Aquatic Habit' Journ. Linn. Soc, 

 Bot. vol. xxix (1891-93) p. 485. 



3 'Annals of Botany ' vol. xri (1902) p. 559. 



