596 



NATURE 



[September 21, 191 1 



lu the discovery of traces of secondary growth in Ophio- 

 glossum, and not to that of Isoetes, a plant much more 

 related, as we now believe, to the Lepidodendracea . 

 and the structure of which had been so thoroughly in- 

 vestigated 1>> Hofmeister. Williamson, it is true, refers 

 to the secondary growth in the stem of Isoetes in his 

 in moir on Stigmaria, but compares it with the periderm- 

 forming cambium of that plant, and does not, therefore, 

 [nise any agreement in the secondary growth of these 

 two plants. 



Adopting Von Mohl's interpretation of the root-bearing 

 basi of the Isoetes plant as a " caudex descendens," 

 Williamson instituted a morphological comparison between 

 the latter and the branching Stigmaria, and came to the 

 conclusion that they were homologous structures, a view 

 which, as we heard at Sheffield, is supported by Dr. Lang 

 on the strength of a re-examination of the anatomy of the 

 stock of Isoetes. If we do not accept Williamson's inter- 

 im i . it ion of the Stigmarian axis as a downward prolonga- 

 tion of caulome nature, the question remains open whether 

 this underground structure represented a leafless modifica- 

 tion of a normal leaf-bearing axis such as is known in the 

 leafless rhizoms of Neottia and other saprophytic plants, or 



! her the Stigmarian axes were morphological entities 

 of peculiar character. Grand' Eury, in comparing them 

 with the rhizoms of Psilotum, accepted the former 

 alternative, and, apart from morphological considerations, 

 was led to this view by the fact that he had observed 

 aerial stems arising in many instances as buds on the 

 horizontal branches of Stigmaria. Confirmation of this 

 mode of growth is still required, but it is quite conceivable 

 that there may have been a mode of vegetative reproduc- 

 tion in the Stigmariae analogous to that of Ophio- 

 glossum.' 



The alternative interpretation of the Stigmarian axes as 

 special morphological entities has received weighty support 

 from Scott and Bower, who consider them comparable to 

 the rhizophores of Selaginella, which, as is well known, 

 may either be root-bearers, or in certain circumstances 

 become transformed into leafy shoots. This peculiarity- 

 has led Goebel to regard them as special members, some- 

 what intermediate between stems and roots. But though 

 they might therefore be regarded as of a primitive nature, 

 the rhizophores of the Selaginellacea? seem such specialised 

 structures that I incline to agree with Bower that, so far 

 as their correspondence with Selaginella is concerned, the 

 Stigmarian axes would agree most closely with the basal 

 knot formed on the hypocotyl of Selaginella spmulosa. 

 Seeing, however, that the nearest living representative of 

 the Lepidodendracea? is in all probability Isoetes, which 

 Bower has aptly summarised as like " a partially differ- 

 entiated Lepidostrobus seated upon a Lepidodendroid 

 base," we must inevitably consider the root-bearing base 

 of Isoetes as homologous with the branching axes of Stig- 

 maria, whatever their morphological nature may have 

 been, and perhaps we shall be on the safest ground if we 

 consider them both as different expressions of the continued 

 growth of the lower region of the plant, which appears 

 to have been a primary feature in the morphology of both 

 these members of the Lycopodiales. 



The somewhat considerable difference in external appear- 

 ance between the homologous organs of these two plants 

 may be considered bridged over by the somewhat reduced 

 axes of Stigmariopsis and by the still more contracted 

 base of the Mesozoic Pleuromoia, which, in spite of its 

 very different fructification, we may unhesitatingly com- 

 pare with Isoetes so far as its root-bearing axis is con- 

 cerned. 



1 was inclined at one time to seek an analogy for the 

 larian axis in that interesting primitive structure, the 



rm of Phylloglossum, and of embryo Lycopods ; 



but I now consider that the resemblances are largely super- 

 ficial, and do not rest upon any satisfactory anatomical 

 correspondence. 



One of the features which has caused some divergence 



1 It is of interest in this connection to note that Potonie has recently put 

 forward the suggestion that many of these vertical outgrowths from the more 

 or '-is horizontal Stigmarian axes, some of which, as figured and described 

 by Goldenberg, taper off rapidly to a point, without any trace of ramifica- 

 tion, may he comparable with the conical " knees" of Tax-odium, and repre- 

 sent woody tineumatophores so common in the Swamp Cypress and other 

 swamp-inhabiting trees. 



of opinion in the past as to the morphology of the Stig* 

 marian axis has been the definite quincuncial arrangement 

 and the apparent exogenous origin of the roots borne on 

 these underground organs. Schimper, indeed, considered 

 these two features so characteristic of foliar organs that he 

 suggested that these so-called " appendices " might possibly 

 be metamorphosed leaves. Not quite satisfied with this 

 view, Renault endeavoured to establish the existence of 

 two types of lateral organs on the Stigmarian axis, true 

 roots with a triarch arrangement of wood and root-like 

 leaves of monarch type. Williamson, however, clearly 

 showed that the apparent triarch arrangement was really 

 due to the presence at two angles of the metaxylem of the 

 first tracheids of secondary wood, and reasserted the exist- 

 ence of only one type of appendicular organs, agreeing so 

 closely, both in structure and in their orientation to the 

 axis, on which they were borne, with the roots of Isoetes 

 that it would be impossible to deny the root nature of the 

 Stigmarian " appendices " without applying the same 

 treatment to the roots of Isoetes. 



Still, so distinguished a Palseobotanist as Solms Lau- 

 bach, after a careful weighing of all the available evidence, 

 continued to uphold Schimper's view of the foliar nature 

 of these outgrowths, both in his " Pala?ophytologie " and 

 in his memoir on Stigmariopsis, in which he stated that he 

 was in complete agreement with Grand' Eury's conclusion : 

 " Que ces organes sont indistinctement des rhizomes et 

 que les Sigillaires n'avaient pas de racines reelles, ainsi 

 que Psilotum." Indeed, in reviewing the account I gave 

 of the occurrence of a special system of spiral trai 

 the outer cortex of the Stigmarian rootlets, Count Solms 

 directed attention to their similarity to the transfusion 

 tissue of Lepidodendroid leaves, and asserted that we have 

 here a further indication of the former foliar nature of these 

 rootlets. Personally, 1 still adhere to the belief, expressed 

 at the time, that these peripheral cortical tracheids repre- 

 sent a special development required by a plant with an 

 aquatic monarch root of the Isoetes type and a large 

 development of aerial evaporating surface. The fact that 

 the lateral outgrowths from the Stigmarian axis have 

 been generally considered to be exogenous is not a valid 

 argument against their root nature, as the same origin is 

 ascribed to the roots of Phylloglossum and to those pro- 

 duced on the rhizophores of Selaginella. Probably, indeed, 

 as Bower points out in his masterly exposition of the 

 "Origin of a Land Flora," in dealing with the Lyco- 

 podiales, " the root in its inception would, like the stem 

 of these plants, be exogenous." According to the "re- 

 capitulation theory," indeed, the exogenous formation of 

 the roots in the embryo of certain Lycopods, as well as of 

 the first roots of Isoetes and the first root of the Filicales, 

 might be regarded as the retention of a more primitive 

 character in these particular organs. The roots of Stig- 

 maria, even if exogenous, might therefore merely repre- 

 sent a more ancestral stage. This difference between the 

 roots of Isoetes and the rootlets of Stigmaria may. how- 

 ever, be more apparent than real, for my colleague, Dr. 

 Lang, has directed my attention to the fact that there appear 

 to be in Stigmaria remnants of a small-celled tissue on the 

 outside of what has generally been taken to be the super- 

 ficial layer of the Stigmarian axis, and a careful investiga- 

 tion of this point inclines me to agree with him that very 

 probably the Stigmarian rootlets were actually formed like 

 those of Isoetes, somewhat below the surface layer, which, 

 after the emergence of the rootlets, became partially dis- 

 organised. Should this surmise prove correct, when apices 

 of Stigmaria showing structure come to light, the last real 

 difference between the rootlets of Isoetes and the rootlets 

 of Stigmaria will have disappeared, and the view for which 

 Prof. Williamson so strongly contended will be finally 

 established. 



While a careful comparison of Isoetes with the extinct 

 Lycopodiaccous plants may be taken to settle finally its 

 systematic position, the Psilotacese have been son 

 disturbed by such comparisons. Placed formerly without 

 much hesitation in the phylum Lycopodiales, certain 

 features in their organisation, such as the dichotomy of 

 their sporophylls and the structure of their fructification 

 generally, have suggested affinity with that interesting 

 group of extinct plants, tin Sphenophyllales. Their actual 

 inclusion in this group 1a- Thomas and by Bower may 



NO. 2 1 86, VOL. 87] 



