CHAPTER XXII.

Marattiales (Fossil).

The discovery of Pteridosperms has necessarily led to a considerable modification of the views formerly held that existing genera of Marattiaceae represent survivors of a group which occupied a dominant position in the forests of the Coal age. Mr Arber writes:—“The evidence, formerly regarded as beyond suspicion, that the eusporangiate ferns formed a dominant feature of the vegetation of the Palaeozoic period, has been undermined, more especially by the remarkable discovery of the male organs of Lyginodendron by Mr Kidston. At best we can only now regard them as a subsidiary group in that epoch in the past history of the vegetable kingdom[983].” Dr Scott expresses himself in terms slightly more favourable to the view that the Marattiaceae represent the aristocracy among the Filicales. He says:—“We now have to seek laboriously for evidence, which formerly seemed to lie open to us on all hands. I believe, however, that such careful investigation will result in the resuscitation of the Palaeozoic ferns as a considerable, though not as a dominant group[984].” Zeiller’s faith[985] in the prospect of Marattiaceous ferns retaining their position as prominent members of Palaeozoic floras, though shaken, is not extinguished: he recognises that they played a subordinate part.

Reference has already been made to the impossibility of determining whether Palaeozoic fern-like fronds may be legitimately retained in the Filicales, or whether they must be removed into the ever widening territory of the Pteridosperms. The difficulty is that the evidence of reproductive organs is very far from decisive. In the absence of the female reproductive organs, the seeds, we cannot in most cases be certain whether the small sporangium-like bodies on fertile pinnules are true fern sporangia or the microsporangia of a heterosporous pteridosperm. What is usually called an exannulate fern sporangium, such as we have in Angiopteris and in many Palaeozoic plants, has no distinguishing features which can be used as a decisive test. The microsporophylls of the Mesozoic Bennettitales produced their spores in sporangial compartments grouped in synangia like those of recent Marattiaceae; and in the case of Crossotheca, a type of frond always regarded as Marattiaceous until Kidston[986] proved it to be the microsporophyll of Lyginodendron, we have a striking instance of the futility of making dogmatic assertions as to the filicinean nature of what look like true fern sporangia. In all probability Dr Kidston’s surmise that the supposed fern sporangia known as Dactylotheca, Renaultia, Urnatopteris are the microsporangia of Pteridosperms will be proved correct[987]. The question is how many of the supposed Marattiaceous sporangia must be assigned to Pteridosperms? There is, however, no reasonable doubt that true Marattiaceae formed a part of the Upper Carboniferous flora. All that can be attempted in the following pages is to describe briefly some of the numerous types of sporangia recognised on Palaeozoic fern-like foliage, leaving to the future the task of deciding how many of them can be accepted as those of ferns. It is impossible to avoid overlapping and some repetition in the sections dealing with true Ferns and with Pteridosperms. The filicinean nature of the stem known as Psaronius (see [page 415]) has not as yet been questioned.

The nomenclature of supposed Marattiaceous species from Carboniferous and Permian rocks is in a state of some confusion owing to a lack of satisfactory distinguishing features between certain types to which different generic names have been assigned. As we have already seen in the case of supposed leptosporangiate sporangia, the interpretation of structural features in petrified or carbonised sporangia does not afford an example of unanimity among palaeobotanical experts.

Ptychocarpus.

This generic name, proposed by the late Professor Weiss[988], is applied to a type of fructification illustrated by the plant which Brongniart named Pecopteris unita, a species common in the Upper Coal-Measures of England[989]. It is adopted by Kidston for fertile specimens from Radstock which he describes as Ptychocarpus oblongus[990], but the precise nature of the fertile pinnules of this species cannot be determined.

Ptychocarpus unita (Brongn.[991]). Fig. 291, A, B. (= Goniopteris unita, Grand’Eury.)

This species has tripinnate fronds with linear pinnae bearing contiguous pinnules of the Pecopteris type ([fig. 291], B), 4–5 mm. long, confluent at the base or for the greater part of their length. On the under surface of the fertile segments, which are identical with the sterile, occur circular synangia ([fig. 291], A) consisting of seven sporangia embedded in a common parenchymatous tissue and radially disposed round a receptacle supplied with vascular tissue. The synangium is described as shortly stalked like those of Marattia Kaulfussii ([fig. 245], B′, p. 320). In shape, in the complete union of the sporangia, and presumably in the apical dehiscence, Ptychocarpus agrees very closely with Kaulfussia ([fig. 245]); but we cannot be certain that we have not a collection of microsporangia simulating a fern synangium.

A synangium closely resembling Ptychocarpus has been described by Mr Watson[992] from the Lower Coal-Measures of Lancashire as Cyathotrachus altus, but there is no convincing evidence as to the nature of the plant on which it was borne.

Danaeites.

This generic name, instituted by Goeppert[993], has been used by authors without due regard to the nature of the evidence of affinity to Danaea. The type named by Stur Danaeites sarepontanus[994] ([fig. 291], E) bears small pecopteroid pinnules with ovoid sporangia in groups of 8–16 in two contiguous series on the lower face of the lamina. The sporangia dehisce by an apical pore and are more or less embedded in the mesophyll of the segments. No figures have been published showing any detailed sporangial structure, and such evidence as we have is insufficient to warrant the conclusion that the resemblance to Danaea is more than an analogy.

Parapecopteris.
Parapecopteris neuropteroides, Grand’Eury. Fig. 290, D.

The plant described by Grand’Eury[995] from the Coal-fields of Gard and St Étienne, and made the type of a new genus, is characterised by pinnules intermediate between those of Pecopteris and Neuropteris[996] and by the presence of two rows of united sporangia along the lateral veins, as in Danaea and Danaeites.

Asterotheca.

Certain species of Pecopteris fronds from Carboniferous strata are characterised by circular sori or synangia consisting of a small number (3–8) of exannulate sporangia attached to a central receptacle and free only at their apices. Strasburger[997] suggested a Marattiaceous affinity for Asterotheca and Stur[998] describes the species Asterotheca Sternbergii Goepp. ([fig. 291], C, D) as an example of a Marattiaceous fern. The latter author retains Corda’s genus Hawlea[999] for the fertile fronds of the common Coal-Measures species Pecopteris Miltoni, while on the other hand Kidston[1000] includes this type in Asterotheca.

Pecopteris (Asterotheca) Miltoni (Artis).
1825.Filicites Miltoni, Artis, Antedil. Phyt. Pl. XIV.
1828.Pecopteris Miltoni, Brongniart, Prodrome, p. 58.
1828.Pecopteris abbreviata, Brongniart, Hist. vég. foss. p. 337, Pl. CXV. figs. 1–4; Lindley and Hutton, Foss. Flor. Vol. III. Pl. 184.
1845.Hawlea pulcherrima, Corda, Flor. Vorwelt, p. 90, Pl. LVII. figs. 7, 8.
1877–1888.Hawlea Miltoni, Stur, Culm Flora, p. 293; Farne Carbon. Flora, p. 108, Pls. LIX. LX.
1888.Pecopteris (Asterotheca) abbreviata, Zeiller, Flor. Valenc. p. 186, Pl. XXIV. figs. 1–4.

Fig. 290.

  1. Alethopteris lonchitica. × 2½. | For description
  2. Lonchopteris rugosa. × 2.| see; Chap. XXVII.
  3. Sphenopteris Hoeninghausi. × 4.|
  4. Parapecopteris neuropteroides.
  5. Pecopteris (Dactylotheca) plumosa [= P. (Dactylotheca) dentata Zeiller (88)]. × 4.

(A–C, E, after Zeiller; D, after Grand’Eury.)

The fronds of this species reached a length of more than 3 metres and a breadth of 2 metres. They are characterised by the presence of aphlebiae[1001] appressed to the rachis and by circular sori composed of a small number (3–6) of sporangia. In habit and in the form of the pinnules this type is similar to Dactylotheca plumosa.

Fig. 291.

(A, B, after Renault; C–G, after Stur; H, I, after Strasburger; K, after Sterzel.)

Hawlea.

Stur[1002] retains this generic name for sori in which the sporangia are free and united only by the proximal end to a central receptacle ([fig. 291], F, G). He describes the individual sporangia as possessing a rudimentary annulus, a comparatively strong wall, and terminating in a pointed distal end. He emphasises the greater degree of cohesion between the sporangia of Asterotheca as the distinguishing feature of that genus; but this is a character difficult to recognise in some cases, and from the analogy of recent ferns one is disposed to attach little importance to the greater or less extent to which sporangia are united, at least in such cases as Asterotheca and Hawlea when the cohesion is never complete.

Scolecopteris.

Zenker[1003] gave this name to detached fertile pinnules from the Lower Permian of Saxony, which he described as Scolecopteris elegans. He recognised the fern nature of the sori and suggested that the pinnules might belong to the fronds of one of the “Staarsteinen” (Psaronius), a view which subsequent investigations render far from improbable. The sori, which occur in two rows on the lower surface of the small pecopteroid segments with strongly revolute margins ([fig. 291], H–K), contain 4–5 sporangia attached to a stalked receptacle comparable with that of Marattia Kaulfussii. These pedicellate synangia were fully described by Strasburger[1004], who decided in favour of a Marattiaceous alliance. The lower portions of the distally tapered sporangia are concrescent, the distal ends being free ([fig. 291], H). Stur includes in Scolecopteris the common species Pecopteris arborescens ([fig. 376]), but Kidston[1005] states that the British example of Scolecopteris is S. polymorpha, Brongn. from the Upper Coal-Measures.

Scolecopteris elegans Zenk. furnishes an example of a plant, or plant fragment, which has been assigned to the animal kingdom. Geinitz[1006] described silicified pinnules as Palaeojulus dyadicus, the generic name being chosen because of the resemblance to Millipedes such as the genus Julus. The mistake is not surprising to anyone who has seen a block of siliceous rock from Chemnitz crowded with the small pinnules with their concave surfaces formed by the infolding of the edges. Sterzel[1007], who pointed out the confusion between Myriapods and Filices, has published figures which illustrate the deceptive resemblance of the pinnules, with their curved lamina divided by lateral veins into segments, to the body of a Millipede (fig. 291, K). He points out that Geinitz searched in vain for the head and legs of Palaeojulus and expressed the hope that further examination would lead to fresh discoveries: the examination of sections revealed the presence of sporangia and demonstrated the identity of Palaeojulus and Scolecopteris.

Discopteris.

Stur[1008] instituted this genus for fertile fronds from the Upper Carboniferous Schatzlarer beds, including two species Discopteris karwinensis and D. Schumanni. He described the small Sphenopteroid pinnules as characterised by disc-shaped sori made up of 70–100 sporangia attached to a hemispherical receptacle: the absence of a true annulus led him to refer the genus to the Marattiaceae. In his memoir on the coal-basin of Heraclea (Asia Minor), Zeiller[1009] instituted the species Sphenopteris (Discopteris) Rallii and figured sporangia resembling those described by Stur in the possession of a rudimentary “apical annulus.” He compared the sporangia with those of recent Osmundaceae and Marattiaceae. In the later memoir on the Upper Carboniferous and Permian plants of Blanzy and Creusot, Zeiller[1010] gives a very full and careful description of fertile specimens of Sphenopteris (Discopteris) cristata, a fern originally described by Brongniart as Pecopteris cristata[1011]. Many of the Sphenopteroid pinnules of this quadripinnate fern frond show the form and structure of the sori with remarkable clearness in the admirable photographs reproduced in Plates I.–III. of Zeiller’s Blanzy memoir. The lobed pinnules of this species are of oval-triangular form, 5–15 mm. long and 2·5–6 mm. broad[1012]. An examination of the type-specimens of Discopteris from Vienna enabled Zeiller to correct Stur’s original description of the sori: he found that the Austrian and French specimens, though specifically distinct, undoubtedly belong to one genus. The sori in Discopteris cristata are globular, as in the recent genera Cyathea and Alsophila, and frequently cover the whole face of the lamina. The individual sporangia are 0·4–0·5 mm. long and 0·15–0·2 mm. in diameter; they are exannulate, but for the annulus is substituted a group of thicker-walled and larger cells in the apical and dorsal region. The description by Stur of a hemispherical receptacle seemed to indicate an important difference between the Austrian and French species; but Zeiller found that this feature does not actually exist and that it was so described as the result of misinterpretation. Zeiller succeeded in isolating spores, 40–50 μ in diameter, from some of the sporangia of D. cristata and found that they exhibited the three-rayed pattern characteristic of fern-spores and which is indicative of their formation in tetrads. The conclusion arrived at is that the genus Discopteris, as represented by D. karwinensis, D. cristata etc., may be regarded as a true fern and included in the Marattiaceae. As Zeiller points out, the sori of Discopteris differ from those of recent Marattiaceae in their pluriseriate construction and agree in this respect with those of the Cyatheaceae. The comparison already made[1013] between the sporangia of D. Rallii and those of recent Osmundaceae holds good: the genus affords another example of a generalised type, in this case probably a fern, combining features which are now distributed among the Marattiaceae, Osmundaceae and Cyatheaceae.

•••••

In addition to genera founded on true synangia or groups of free or partially united sporangia, the literature of Palaeozoic ferns contains several generic names applied to sporangia which occur singly on Sphenopteroid or Pecopteroid pinnules. The following may serve as examples; but it should be stated that these will probably be transferred eventually to the Pteridosperms. It is, however, immaterial whether they are dealt with here or in the chapter devoted to the seed-bearing “ferns.”

Dactylotheca.

Zeiller[1014] created this genus for fertile fronds of Pecopteris dentata Brongn. (= P. plumosa Artis[1015]), a common British species in the Upper and Middle Coal-Measures. Stur[1016] included P. dentata in his list of species of Senftenbergia, the genus to which reference was made under the Schizaeaceae.

Pecopteris (Dactylotheca) plumosa (Artis). Figs. [290], E, [292], [293].
1825.Filicites plumosus, Artis, Antedil. Phyt. p. 17, Pl. XVII.
1828.Pecopteris plumosa, Brongniart, Hist. vég. foss. p. 348, Pls. CXXI. CXXII.
—P. dentata, Brongniart, ibid. Pls. CXXIII. CXXIV.
—P. delicatulus, Brongniart, ibid. Pl. CXVI. fig. 6.
1832.Sphenopteris caudata, Lindley and Hutton, Foss. Flor. Vol. I. Pl. XLVIII.; Vol. II. Pl. CXXXVIII.
1834.Pecopteris serra, Lindley and Hutton, ibid. Vol. II. Pl. CVII.
1834.Schizopteris adnascens, Lindley and Hutton, ibid. Vol. I. Pls. C. CI.
1836.Aspidites caudatus, Goeppert, Syst. fil. foss. p. 363.
1838.Steffensia silesiaca, Presl, in Sternberg, Flor. Vorwelt, Vers. II. p. 122.
1869.Pecopteris silesiacus, Schimper, Trait. pal. vég. Vol. I. p. 517.
—Cyathocarpus dentatus, Weiss, Flora der jüngst. Stk. und Roth. p. 86.
1877.Senftenbergia plumosa, Stur, Culm Flora, II. p. 187 (293).
—S. dentata, ibid.
1886.Dactylotheca plumosa, Kidston, Cat. Palaeozoic Plants, p. 128.
1888.Dactylotheca dentata, Zeiller, Flor. Valenc. Pls. XXVI.–XXVIII.

For a fuller synonymy reference should be made to Kidston’s account of this species[1017], from which the above list is compiled. The large fronds of this species are tri- or quadripinnate. The pinnules vary much in shape and size and in degree of lobing, according to their position on the frond ([fig. 293]). The primary pinnae are subtended by two Aphlebiae ([fig. 293], A) appressed to the rachis, like the delicate leaves of the recent fern Teratophyllum aculeatum (see [page 301]). The sporangia (0·5–0·65) are oval and exannulate and are attached parallel to the lateral veins; they may occupy the whole of the space between the midrib and the edge of the pinnules. This species occurs in the Upper, Middle, and Lower Coal-Measures of Britain, reaching its maximum in the Upper Coal-Measures. The aphlebiae undoubtedly served to protect the young fronds, as shown by a specimen figured by Kidston ([fig. 293], B); they may also have served other purposes, as suggested by the above comparison with Teratophyllum, in the mature frond. Lindley and Hutton regarded the aphlebiae as leaves of a fern climbing up the rachis; which they named Schizopteris adnascens, a confusion similar to that already mentioned in the description of Hemitelia capensis (see p. 304).

Fig. 292. Dactylotheca plumosa. (After Kidston. Slightly reduced.)

Fig. 293. Dactylotheca plumosa: A. Rachis with Aphlebiae. B, a, young pinnae circinately folded. (After Kidston. A, B, ⅘ nat. size.)

Renaultia.

This name was proposed by Zeiller[1018] for Upper Carboniferous fertile pinnae of the Sphenopteroid type, bearing ovoid sporangia either singly or in marginal groups of 2 to 5 at the ends of the veins. The appearance of the apical cells occasionally suggests the presence of a rudimentary annulus. Kidston has recorded this type of fructification in Britain[1019]. Stur describes fertile pinnules of the same type under the generic name Hapalopteris[1020].

Zeilleria.

This genus was founded by Kidston[1021] for fertile pinnae of a very delicate fern, Zeilleria delicatula (Sternb.) characterised by filiform ultimate segments bearing an indusium-like body, spherical when immature and splitting at maturity into four small valves. Kidston, in his earlier paper, compared the species with recent Hymenophyllaceae. In the same genus he includes Z. avoldensis[1022] (Stur) assigned by Stur to Calymmatotheca, a genus described by some authors as characterised by groups of radially elongated sporangia at the tips of the pinnules; these supposed sporangia are now known to be the valves of an indusium-like organ or cupule, as Stur asserted. There can be little doubt that the fertile fronds placed in Calymmatotheca and in Zeilleria were borne by Pteridosperms.

Urnatopteris.

The Upper Carboniferous fronds of a delicate Sphenopteris habit, to which this name was assigned by Kidston[1023], were described by him as Eusphenopteris tenella (Brongn.)[1024] and compared with Hymenophyllaceae; subsequently Kidston expressed the opinion that Urnatopteris may be a Marattiaceous fern, as Williamson[1025] believed; he has since suggested that the sporangia are the microsporangia of a Pteridosperm[1026]. The sterile and fertile pinnae differ in the absence of a lamina in the latter. The sporangia (or microsporangia) are characterised by a poricidal dehiscence.

The records from strata higher in the geological series than the Permian, disregarding many of doubtful value, afford ample testimony to the existence of Marattiaceae in Upper Triassic and Rhaetic floras.

Marattiopsis.

The generic name Danaeopsis was applied by Heer[1027] to an Upper Triassic fern, previously described by Presl as Taeniopteris marantacea. A splendid specimen from the Keuper of Stuttgart is figured in Schimper’s Atlas[1028] showing the pinnate habit of the frond and the broadly linear segments, 25 cm. × 3·5 cm., bearing rows of contiguous sporangia. The large pinnules have a strong midrib giving off curved and forked lateral veins. Presl’s species may most appropriately be included in the genus Marattiopsis. A specimen of M. marantacea described by Leuthardt[1029] as Danaeopsis marantacea from the Upper Trias of Basel shows a peculiarity in the venation; the lateral veins often fork near their origin, as noticed by other authors, but each vein forks a second time near the edge of the lamina and the two arms converge, forming a series of intramarginal loops ([fig. 265], B).

Marattiopsis Muensteri (Goepp.). [Fig. 245], D, E.

This widely spread Rhaetic plant affords the best example of a post-Permian species which may be accepted as an authentic record of fossil Marattiaceae. Various generic names have been used for this species; Goeppert originally described the plant as Taeniopteris Muensteri[1030]; Schimper[1031] proposed the name Marattiopsis, and Schenk[1032] substituted Angiopteris on the ground that the fertile pinnules resemble that genus rather than Marattia. Marattiopsis, if interpreted as indicating a family resemblance rather than special affinity to the genus Marattia, would seem to be the more appropriate designation.

This species has been figured by several authors and in many instances with fertile pinnules; the best illustrations are those published by Zeiller[1033] in his monograph of Tonkin plants.

The pinnate fronds are characterised by a broad rachis bearing sessile broadly linear pinnules rounded at the base, obtusely pointed at the apex, reaching a length of 15–20 cm. and a breadth of 12–35 mm. From a well-marked midrib are given off secondary veins dichotomously branched close to their origin. The linear synangia near the ends of the veins contain two rows of sporangial compartments and open as two valves as in Marattia. (Cf. [fig. 245], A, p. 320.)

This species occurs in the Rhaetic beds of Scania, Franconia, and Tonkin. A similar type is figured by Fontaine from Jurassic beds in California as Angiopteridium californicum[1034], and Bartholin[1035] and Moeller[1036] record M. Muensteri from the Lias of Bornholm. Schenk’s species from China[1037], Angiopteris Richthofeni, is a closely allied species, and a similar form is recorded from Jurassic and Caucasian strata[1038]. The microscopical examination by Nathorst[1039] of a group of spores from a synangium of M. Muensteri shows that they resemble those of recent Marattiaceae.

•••••

From the Upper Triassic plant beds of Lunz, Stur has included several species of ferns in the Marattiaceae, and of these Krasser[1040] has recently published full diagnoses but unfortunately without illustrations. In addition to Marattiopsis marantacea (Presl) the list includes species referred to Coniopteris, to Speirocarpus, a genus founded by Stur, to Oligocarpia, Asterotheca, and Bernouillia (Heer).

As already pointed out, some at least of these Austrian ferns are more probably Osmundaceous than Marattiaceous.

Danaeopsis
Danaeopsis Hughesi, Feistmantel.

The pinnate fronds described by Feistmantel[1041] from the Middle Gondwana rocks of India and recorded from Rhaetic strata in South Africa[1042], China[1043], and Tonkin[1044], may belong to a member of the Marattiaceae, but no fertile specimens have been described. The close agreement between the sterile leaves from India and South Africa and the fertile fronds of Marattiopsis marantacea suggests generic identity.

The Upper Triassic ferns described by Heer, Krasser[1045], and Leuthardt[1046] as Bernouillia have been referred to the Marattiaceae, but without trustworthy evidence in favour of this affinity.

The large leaves, 70 cm. long and 7 cm. broad, described by Zigno[1047] from the Jurassic of Italy as Danaeites Heeri, are probably Cycadean. The Polish Jurassic species Danaea microphylla[1048] is a more satisfactory record.

Fig. 294.

Nathorstia.

This name was instituted by Heer[1049] for pieces of pinnate fronds from Lower Cretaceous rocks of Greenland. The resemblance of the long pinnules to the fertile segments of Laccopteris is so close that generic identity might well be assumed, but it has recently been shown by Nathorst[1050] that the soral characters justify Heer’s use of a distinctive name for the Arctic fern. The circular sori arranged in two rows ([fig. 294], A, B) are superficially identical with those of Laccopteris, but consist of concrescent sporangia forming a circular synangium ([fig. 294], C, D) like those of Kaulfussia and Ptychocarpus. The lighter areas in [fig. 294], D, represent the sporangia: fig. C shows the radial disposition of the numerous sporangial compartments round a central receptacle. From a stout midrib lateral veins arise at right angles, but their distal terminations are not preserved. It is probable, as Nathorst suggests, that Bayer’s[1051] species Drynaria fascia from the Lower Cretaceous rocks of Bohemia should be referred to Heer’s genus. In the absence of well-preserved sori it would be exceedingly difficult, or even impossible, to distinguish between pinnules of Laccopteris and Nathorstia.

A Tertiary species, Marattia Hookeri ([fig. 261], C, p. 350), described by Gardner and Ettingshausen[1052] from the Eocene beds of the Isle of Wight is referred by them to the Marattiaceae because of a resemblance of the sterile pinnae to those of M. Kaulfussii; but this is insufficient evidence of relationship.