xvi FOSSIL ARCHEGON1ATES 589 



have closely resembled, except for their much greater dimen- 

 sions, such species of Lycopodium as L. cernuum or L. den- 

 droideum. The branching* was prevailingly dichotomous, and 

 the shoots thickly set with acicular leaves of a size correspond- 

 ing to the dimensions of the shoots. Sigillaria seems to have 

 been much less freely branched than Lepidodendron, and it 

 has even been supposed that in some species branching was en- 

 tirely suppressed. Of the living species of Lye op odium, L. 

 inundatum or L. saururus may be compared in habit to Sigil- 

 laria. Trunks of Lepidodendron a hundred feet in length have 

 been found, showing the genuine tree-like proportions of these 

 giant Club-mosses. 



The base of the stem in both Lepidodendron and Sigillaria 

 is often found connected with forking structures, which were 

 originally described as distinct fossils under the name Stig- 

 maria. It is clear, however, that these were the underground 

 parts of Lepidodendron and Sigillaria, probably rhizomes 

 rather than true roots. The name Stigmaria is given them be- 

 cause of the very regular scars upon the surface, and these have 

 been shown to be the points of attachment for roots or root- 

 lets, if the main Stigmaria branches are true roots and not rhi- 

 zomes (see Scott (i), Fig. 82). 



The slender pointed leaves were often of considerable 

 length, 15 centimetres or more, and resembled those of Selagi- 

 nella rather than Lycopodium in having a ligule near the base. 

 (See Scott (i), Figs. 48, 58). 



The internal structure is well known in a good many forms, 

 especially among the Lepiclodendracese (Scott (i)), and it is 

 evident that there was a good deal of difference among them, 

 especially in the degree of secondary thickening which occurred. 



In all known species of Lepidodendron (Scott (i), p. 123)' 

 there is always a single stele with centripetally developed pri- 

 mary wood. There may or may not be a central pith. In the 

 larger stems -there is usually a central medulla about which the 

 primary wood forms a ring. Probably the phloem, which is 

 rarely well preserved, formed a ring outside the xylem. The 

 cortex is relatively very thick, as it is in the living Lycopo- 

 dinese, and through it passed obliquely the leaf-trace bundles, 

 one being given off from the central stele of the stem to each 

 leaf-base. 



While in some species, e. g. f L. parvulum, there was appar- 



