November i i, 1922] 



NA TURE 



639 



(Lepidocarpon) a single megaspore only came to matur- 

 ity, constituting the embryo-sac, while an integument, 

 like a seed-coat, grew up round the sporangium. 



The prothallus is sometimes well preserved, both in 

 the seed-like bodies and in the more ordinary mega- 

 spores. In the latter (Lepidostrobus Veltheimianus), Dr. 

 Gordon has recorded a perfectly typical archegonium, 

 showing that the details of reproduction in these old 

 Lycopods were the same as in their modern hetero- 

 sporous representatives. 



The best-known member of the Horsetail race was 

 Archseocalamites, remarkable for the long leaves, often 

 repeatedly forked, very different from the foliage which 

 we are accustomed to associate with the Equisetales. 

 The later Calamites were more or less intermediate in 

 this respect. Anatomically, the Calamites, whether of 

 Lower or Upper Carboniferous age, developed much 

 secondary wood, and, like many contemporary Lyco- 

 pods, became trees. The Lower Carboniferous Proto- 

 calamites is remarkable for possessing primary wood, 

 centripetally formed, thus presenting some analogy 

 with the Sphenophylls, in which this tissue is highly 

 developed. 



The cones attributed to Archa;ocalamites are, curi- 

 ously enough, intermediate in structure between modern 

 Equisetum cones and those of the Upper Carboniferous 

 Calamites, for sterile bracts were either absent, or 

 developed only at long intervals. In Equisetum, of 

 course, they are absent altogether, while in the well- 

 known Calamostachys and allied Upper Carboniferous 

 fructifications, the sterile whorls are equal in number 

 to the alternating fertile verticils. It must be ad- 

 mitted, however, that our knowledge of Lower Carbon- 

 iferous fructifications of this group is still somewhat 

 scanty. 



The Sphenophylls of the period were already very 

 advanced, and in the genus Cheirostrobus appear to 

 have reached their zenith. The great cones of this 

 striking plant, with their elaborate and perfect appar- 

 atus of compound sporangium-bearing organs, and 

 protective sterile appendages, are certainly the most 

 complex cryptogamic fructifications known, from any 

 period. Thus, in certain directions, the Lower Car- 

 boniferous plants had attained a height of development 

 which has never since been equalled. 



Sphenophyllum itself still had, for the most part, 

 the deeply cut leaves of the Upper Devonian species. 

 Where the anatomy is known (S. insigne, from Burnt- 

 island), it is of the same general type as in the later 

 Upper Carboniferous forms, but apparently somewhat 

 less specialised. It is worth remarking, that all the 

 Sphenophyllums formed secondary wood, though they 

 were small plants. Thus growth in thickness by 

 cambium was not confined to arborescent forms in 

 Palaeozoic times, any more than it is now. 



As regards the affinities of the Sphenophylls, some 

 relation to the Horsetail stock seems evident, as indi- 

 cated by the whorled leaves, the general organisation 

 of the cones, and the detailed structure of the sporangia. 

 Presumably these two lines sprang from a common 

 source, but what it was is still unknown. Further 

 affinities, once suggested, with the Lycopods and the 

 recent Psilotaceae have not been confirmed and are 

 probably illusory. Neither has Lignier's hypothesis of 

 a common origin of both branches of the Articulatas 



NO. 2767, VOL. Iio] 



from Ferns, gained any support from the fossil record. 

 The Articulatas, as a whole, remain a completely 

 isolated phylum. 



The Ferns of the Lower Carboniferous were well de- 

 veloped and varied. We meet with the usual difficulty 

 in distinguishing between the fronds of true Ferns, and 

 those of the so-called " Seed-Ferns," which simulated 

 them in habit. Where, however, anatomical characters 

 are available, we find no approximation whatever 

 between the two groups. Pteridosperms and Ferns 

 at all times show themselves perfectly distinct, whenever 

 our knowledge admits of an adequate comparison. 



We have fairly abundant structural material of 

 Lower Carboniferous Ferns, but it seems that practi- 

 cally all of it represents the group called Primofilices by 

 Arber, who by this name meant to suggest age, not 

 primitiveness. They were curious plants, and many 

 of them must have been very unlike any Ferns now 

 living. Unfortunately, our knowledge of their habit 

 is by no means equal to that of anatomical detail. 



The chief family in the Lower Carboniferous is that 

 of the Zygopterids, of which several genera are repre- 

 sented. As we have seen, this family had already 

 appeared in Upper Devonian times. The vascular 

 cylinder of the stem shows some differentiation of the 

 wood into a central region (either a mixed pith or a 

 core of small, short tracheids), and a wide outer zone 

 of larger elements. The petiole always has a peculiar 

 structure, with a bilateral strand (often of complex 

 form) giving off branch-bundles to the right and left. 

 It is remarkable that the genus Clepsydropsis, once 

 thought primitive on account of its simple petiolar 

 structure, has been shown to possess an exceptional!)' 

 high organisation of the stem. 



The most striking point is the morphology of the 

 frond. Even where there were only two series of 

 pinnje (as in normal compound leaves) their plane was 

 not parallel to that of the main rachis, but at right 

 angles to it. Moreover, in several genera there was 

 the greater peculiarity that the pinna? were in four 

 rows, two rows on each side, a condition unexampled 

 in ordinary leaves. In Stauropteris this quadriseriate 

 branching was repeated in successive ramifications, so 

 that the form of the whole frond was compared by 

 Lignier to a bush. In this genus it is practically certain 

 that the leaflets had no blade, and throughout the 

 family there is rarely any proof of its presence. 



The other Lower Carboniferous family of Primo- 

 filices, the Botryopterideas, is at present represented 

 for that period by a single species, the Botryopteris 

 antiqua of Kidston, a plant in all respects of simpler 

 organisation than the Zygopterids, and apparently 

 more like an ordinary Fern. 



Sporangia are known in several cases. Those of 

 Stauropteris were borne singly on ultimate branches 

 of the frond ; they had no annulus, and are very 

 similar to the sporangia associated with the Early 

 Devonian Asteroxylon. In the fructification attri- 

 buted to Diplolabis the sporangia are grouped in a sort 

 of synangium, while those associated with Botryo- 

 pteris have a biseriate annulus. 



Both families show some affinity with the older 

 members of the Osmundaceas, while a relation to the 

 Adder's Tongues has also been traced. But in both 

 directions the connexion seems to be somewhat remote. 



