CHAPTER XIX.

Seed-bearing plants closely allied to members of the Lycopodiales.

i. Lepidocarpon.

In 1877 Williamson[662] published an account of some fossil seeds which he referred to Brongniart’s genus Cardiocarpon[663], a generic title for certain Gymnospermous seeds. Some of these he identified, on the authority of the author of the species, with Cardiocarpon anomalum Carruthers[664]. Several years later Wild and Lomax described a new type of strobilus from the Lower Coal-Measures of Lancashire[665]. The result of this discovery and of the subsequent examination by Scott of additional material, was to establish the fact that the seeds described by Williamson and generally accepted as Gymnospermous, are in reality sporangia belonging to a Lycopodiaceous cone. The seeds to which Carruthers gave the name Cardiocarpon anomalum are, however, distinct from those described under the same name by Williamson and are those of a true Gymnosperm. For this seed-bearing strobilus Scott[666] instituted the generic name Lepidocarpon, which he thus defined: “Strobili, with the characters of Lepidostrobus, but each megasporangium was inclosed, when mature, in an integument, growing up from the superior face of the sporophyll-pedicel. Integument, together with the lamina of the sporophyll, completely enveloping the megasporangium, or nucellus, leaving only an elongated, slit-like micropyle above. A single functional megaspore or embryo-sac developed in each megasporangium, occupying almost the whole of its cavity. Megaspore ultimately filled by the prothallus or endosperm. Sporophyll, together with the integumented megasporangium and its contents, detached entire from the axis of the strobilus, the whole forming a closed, seed-like, reproductive body. Seed-like organ horizontally elongated, in the direction of the sporophyll-pedicel, to which the micropylar crevice is parallel.”

Lepidocarpon Lomaxi, Scott. [Fig. 218].

An immature cone of L. Lomaxi is practically identical with a Lepidostrobus; its sporangia are naked and only acquire their integuments at a later stage. A mature strobilus has a diameter of at least 3 cm. and is about 4 cm. in length. As in typical Lepidostrobi, the axis bears spirally disposed sporophylls, and each sporophyll has a long narrow pedicel approximately at right angles to the cone axis with its distal end expanded into a broad and thick lamina ([fig. 218], B).

At the distal end the pedicel has a thin marginal wing ([fig. 218], C, right-hand half) continuous with the upturned protective lamina. To the upper face of each sporophyll is attached along the whole length as far as the ligule, a single large sporangium; on each side of the base of the sporangium the sporophyll forms a supporting cushion. The relation of the sporangium to the ligule, l, is shown in [fig. 218], B, and in the tangential section, C, which illustrates the triangular form of the sporangium near its distal end.

In mature cones, the sporangia assumed the form of seeds, the change being due to the growth of an investing integument from the upper face of the sporophylls on each side of the sporangia. [Fig. 218], A, illustrates the form of a sporangium as shown in tangential sections; the vascular bundle is seen below the base of the sporangium and the gaps right and left of it probably mark the position of parichnos strands. On each side of the sporangium, b, a fairly thick wall of tissue has grown up from the sporophyll, forming an integument which overtops the apical ridge of the sporangium, leaving a narrow micropyle in the form of a long crevice (m, [fig. 218], B). At the proximal end of the sporangium the integument forms an enclosing wall; at the distal end it abuts on and is continuous with the upturned end of the sporophyll. It is clearly established by Scott that the tissue which invests the sporangia is not the upturned margins of the sporophyll, but a new formation fully entitled to the designation integument. It is noteworthy that the integument is not developed until a late stage in the ontogeny of the strobilus; it is not formed until after the production of the prothallus[667]. The diagrammatic sketch, [fig. 218], B, shows the relation of the integument to the sporophyll and sporangium, the outline of the latter being indicated by a broken line. The columnar wall of the sporangium ([fig. 218], A, b) forms a closed beak within the micropylar crevice, and in the interior of the sporangial cavity the slightly shrivelled membrane, a, represents the single megaspore; traces of the aborted sister-cells of the megaspore are occasionally met with. Scott describes a specimen in which the megaspore is filled with tissue agreeing in appearance with the prothallus in a megaspore of Isoetes or Selaginella; no undoubted archegonia or female organs have been discovered, nor has any spore been found containing an embryo.

Fig. 218. Lepidocarpon Lomaxi, Scott.

The axis of L. Lomaxi has a medullated stele constructed on the same plan as that of some species of Lepidodendron and Lepidostrobus; the vascular bundles supplying the sporophylls pass obliquely upwards and outwards from the stele, St, fig. 218, B, and bend slightly downward just before entering the pedicel of a sporophyll.

Dr Scott has also described a strobilus containing microsporangia partially enclosed by a rudimentary integument. It is, however, of considerable interest to find a partial development in the case of a male flower of an integumentary outgrowth, which it would seem could only be of real functional importance in the female shoot.

It is important to notice that specimens of a second species of Lepidocarpon, L. Wildianum, are recorded from Lower Carboniferous beds of Scotland, a fact which points to a considerable antiquity for this seed-bearing Lycopodiaceous type[668].

The most important question to consider in regard to Lepidocarpon is—are we justified in applying to the integumented sporangia the term seed? The megaspore was not set free as it is in recent Pteridophytes, such as Azolla and other genera with which Lepidocarpon may be compared; it was on the other hand retained in the sporangium, as may sometimes happen even in recent species of Selaginella (cf. [fig. 131], D). Moreover, the megaspore is characterised by a thin enclosing membrane in contrast to the thick coat of a spore which is destined to be shed. The peculiar slit-like form of the micropyle is a distinguishing feature, but this may be readily explained as a convenient form in the case of a radially elongated sporangium. The absence of an embryo, though a distinguishing feature of Lepidocarpon, cannot be held to be a serious obstacle to the use of the term seed; in recent Cycads the embryo, as Scott points out, may not begin to develope until the seed has been shed. It is possible that the seeds of Lepidocarpon were not pollinated on the parent plant.

The lesson which this extinct type teaches, is that certain Lycopodiaceous plants of the Palaeozoic era had reached an important stage in the evolution of a seed. The morphological essentials of true seeds had been acquired; but we do not know the biological conditions under which pollination and fertilisation were effected. Another point of considerable interest is the value of this discovery as an argument in favour of the view that some Gymnosperms are derived from Lycopod ancestors. Leaving the general question until later, it may at any rate be stated that in Lepidocarpon we have a demonstration of the fact that the Lycopodiales were not always distinguished from Gymnosperms by the absence of seeds. There are certain features in Lepidocarpon shared by the seeds of Araucarieae[669] which may well mean something more than mere parallel development in two distinct phyla of the plant-kingdom[670].

ii. Miadesmia.

In 1894 Prof. Bertrand[671] published an account of certain fragments of petrified leaves and twigs of a small herbaceous Lycopodiaceous plant, under the name Miadesmia membranacea, which he discovered in English material in association with Lepidodendron Harcourtii. Subsequently Scott recognised the megasporophylls of the same plant, and microsporophylls have also been discovered. The most complete account of Miadesmia so far published we owe to Dr Benson[672], whose description is based on specimens from several sources.

Miadesmia membranacea, Bertrand. Fig. 219, A–D.

Fig. 219.

The slender stem, characterised by unequal dichotomy, has a single protostele composed of scalariform tracheids with 3–6 peripheral protoxylem groups. A zone of delicate tissue surrounds the xylem; this is described as phloem, but it is not clear whether the designation is based on histological characters or primarily on its position. The cortex consists of an inner lacunar tissue and an outer region limited by a small-celled superficial layer sharply contrasted with the underlying layers of larger cells. The stem of Miadesmia is not uncommon in sections of the Lancashire calcareous nodules, and may be recognised by the delicate crushed tissue of which it mainly consists and by large hypodermal parenchyma. The spirally disposed leaves bear a conspicuous and relatively large ligule, 3 mm. long, in a deep pit ([fig. 219], B and C) roofed over by a few layers of tissue corresponding to the velum in Isoetes (cf. [fig. 133], E, v). The fairly thick central region of the lamina is expanded laterally into thin wings, which in the living state probably bore delicate hairs. These delicate leaves, apparently without stomata, were attached to the stem at an acute angle, and Miss Benson suggests that their form and arrangement may have enabled them to hold water by surface-tension. As seen in [fig. 219], B, C, which represents part of a transverse section near the leaf-base, the ligule is a very characteristic feature, and the size of the single vein is in keeping with the almost filmy nature of the lamina.

In addition to the sections in British collections, I have been enabled by the kindness of Prof. Bertrand to see photomicrographs of the sections on which he founded the genus. One of these sections, transverse to the stem and leaves, illustrates in a striking manner the relatively large size of the leaves and ligules in proportion to the delicate axis of the shoot.

The megasporangiate cone has an axis which agrees in its structure with that of the vegetative stem and bears several megasporophylls approximately at right-angles. As in the foliage leaves, the ligule is prominent and large, and lies in a groove which contains also the megasporangium; both ligule, l, and sporangium, s, as seen in the transverse section represented in [fig. 219], D, are covered by an integument or velum which arises in the proximal part of the leaf and leaves a circular micropylar opening at the beak-like apex of the sporangium. The circular micropyle is surrounded by numerous hairs borne on the integument and which presumably played the part of a feathery stigma. A single megaspore with a thin membrane, m, abuts on the fairly strong sporangial wall, s; in some cases the sporangium and megaspore walls may be indistinguishable, a feature suggesting comparison with seed-structure. Some megaspores have been found filled with a prothallus. The longitudinal section shown in [fig. 219], A, illustrates the characteristic horizontal position of the megasporophyll, as also the relation of the ligule, l, to the sporophyll with its single vascular bundle, and to the hairy integument, which overarches both sporangium and ligule; the line m shows the position of the megaspore-membrane, detached from the sporangial wall on the upper side but in contact with it below. The microsporophyll shown in 219, E, was originally referred to Miadesmia but has since been recognised by Watson[673] as that of a Bothrostrobus.

Miadesmia affords an example of a Palaeozoic plant comparable with Isoetes and Selaginella; it agrees also with Lepidocarpon in possessing true seeds, and with Watson’s Bothrodendron cone in the shape of the sporangia, which are more like those of Selaginella than the radially elongated sporangia of Lepidostrobus. Miadesmia agrees with Selaginella, e.g. S. spinosa, in its stelar structure, in the form of the sporangia, and in the presence of a ligule. It is distinguished by having only one instead of four megaspores in a sporangium, in the possession of an integument which formed a close investment to the spore and served as a stigma (comparable with the stigma-like integument of the male flower of Welwitschia), and in the shedding of the megasporophylls, which have been aptly compared with winged seeds.

LEPIDOCARPON

On the ground of their general anatomical features Lepidocarpon and Miadesmia are clearly entitled to be included among extinct representatives of the Pteridophyta. These plants had, however, crossed what it has been customary to regard as the boundary between Pteridophytes and Phanerogams: they possessed megasporangia with the attributes of seeds. It has been suggested by Lester Ward[674] that Pteridophytic seed-bearing plants shall be recognised as a distinct phylum for which he proposes the name Pteridospermaphyta, a designation implying exclusion from the Spermatophyta as usually understood. For seed-bearing Lycopodiaceous genera he suggests the name Lepidospermae. As knowledge of the Palaeozoic seed-plants increases revision of existing classifications and group names will become necessary, but as yet we are hardly in a position to draw up a satisfactory scheme of grouping; we know little of Lepidocarpon as a whole and it would be premature to commit ourselves, even provisionally, to a classification which is based on such meagre evidence as we possess. Moreover the value to be attached to the seed-habit as a basis of classification can hardly be estimated until fuller information is obtained.