1. Introduction
The Noeggerathiales is a puzzling extinct group of late Palaeozoic plants known by fossils of foliage associated/disassociated with strobilli, isolated strobilous or isolated foliage (e.g. Wang et al. Reference Wang, Zhang, Bek and Pfefferkorn2004, Reference Wang, Pfefferkorn and Bek2009; Pfefferkorn & Wang, Reference Pfefferkorn and Wang2016). Due to a lack of anatomically preserved specimens, these plants until recently had an uncertain systematic position, being historically considered as related to ferns, cycads, sphenopsids or to progymnosperms (e.g. Leary, Reference Leary1980; Wang et al. Reference Wang, Zhang, Bek and Pfefferkorn2004; Taylor et al. Reference Taylor, Taylor and Krings2009). Recently, the finding and description of whole specimens found within a Permian ash-tuff with articulated foliage, (pseudo-like)strobili and stem with anatomical preservation, allowed the elucidation of the Noeggerathiales as members of the progymnosperms clade, a sister group to the seed plants (Wang et al. Reference Wang, Hilton, Pfefferkorn, Wang, Zhang, Bek, Pšenička, Seyfullah and Dilcher2021; Yang et al. Reference Yang, Wang and Wang2023).
Palaeobiogeographically, the Noeggerathiales are restricted to two floral realms, the Cathaysian floral realm (i.e. China) where these plant fossils are relatively abundant and in the Euramerican floral realm where these are considered rare (Pfefferkorn & Wang, Reference Pfefferkorn and Wang2016). As pointed by some authors (Leary & Pfefferkorn, Reference Leary and Pfefferkorn1977; Pfefferkorn & Wang, Reference Pfefferkorn and Wang2016) despite extensive exploration and description of several palaeofloras, noeggerathialean fossils are seldom found in Euramerica Palaeozoic floras. This bias is most probably due to their palaeoecological preferences to extra-basinal ecosystems or specific edaphic/nutrient conditions (but see Pfefferkorn & Wang, Reference Pfefferkorn and Wang2016). Despite its rarity, within Iberian Peninsula, at least ten noeggerathialean taxa were recognized mainly in Carboniferous basins from Spain: Palaeopteridium reussii (Ettingshausen) Kinston, Palaeopteridium michiganense (Arnold) Álvarez-Vázquez, cf. Saaropteris guthoerlii Hirmer, Rhacopteris bipinnata Němejc, Rhacopteris elegans (Ettinghausen) Kidston, Rhacopteris cf. robusta Kidston, cf. Rhacopteris paniculifera Stur, Rhacopteris inaequilatera Göppert, cf. Rhacopteris dichotoma Kidston; in Portugal, only one record is known, Rhacopteris gomesiana Teixeira (Teixeira, Reference Teixeira1943; Wagner, Reference Wagner, Lemos de Sousa and Oliveira1983; Wagner & Álvarez-Vázquez, Reference Wagner and Álvarez-Vázquez2010; Correia et al. Reference Correia, Šimůnek, Sá and Flores2018). However, this number can be reduced to four, as the fossil taxa Rhacopteris is possibly not a noeggerathialean (see Pfefferkorn & Wang, Reference Pfefferkorn and Wang2016).
In recent years, efforts were made to identify old and new fossiliferous outcrops and to collect and describe the palaeobiodiversity of the Portuguese Carboniferous basins. These resulted in the description of several new fossil plant taxa, many endemics to the Portuguese territory (see Correia et al. Reference Correia, Šimůnek, Pšenička, Sá, Domingos, Carneiro and Flores2014 a; Correia et al. Reference Correia, Sá, Murphy, Šimůnek and Flores2016; Pšenička et al. Reference Pšenička, Correia, Šimůnek, Sá, Murphy and Flores2017; Correia et al. Reference Correia, Šimůnek, Sá and Flores2018; Correia et al. Reference Correia, Šimůnek, Cleal and Sá2019; Correia et al. Reference Correia, Bashforth, Šimůnek, Cleal, Sá and Labandeira2020; Correia et al. Reference Correia, Šimůnek and Sá2021 a; Correia et al. Reference Correia, Šimůnek, Cleal, Vallois, Domingos and Sa2021 b; Barbosa et al. Reference Barbosa, Correia, Muchagata, Domingos and Sá2022; Correia et al. Reference Correia, Barbosa, Šimůnek, Muchagata and Sá2023 a; Correia et al. Reference Correia, Pereira, Šimůnek and Cleal2023 b). Despite this effort, since the 1940s, no new noeggerathialean fossils from Portugal were described.
In this paper, we describe and illustrate a new noeggerathialean species of the fossil-genus Palaeopteridium and dissociated reproductive structures possibly related to macrosporangia for the uppermost Carboniferous of Portugal and discuss its implications for the palaeoecology and palaeodiversity of Noeggerathiales.
2. Material and methods
2.a. Fossil locality and material
The fossil material of the new noeggerathialean species was collected in an excavation during fieldwork in 2006, in upper Westphalian D (upper Asturian/upper Moscovian, Middle Pennsylvanian) outcrops of Tanjarro locality, found along the roadside of Dom Miguel (known as ‘Estrada Dom Miguel’), located around Seixo (Fânzeres) region, in Gondomar, northwestern Portugal (Fig. 1). This locality of late Westphalian D plants has been described as an integral part of the Ervedosa flora (Teixeira, Reference Teixeira1944). The outcrops of ‘Estrada Dom Miguel’ fossil site represent a narrow geological-stratigraphic strip that contains a diverse and well-preserved fossil macroflora. This narrow strip occurs on the geographical alignment of the classical Westphalian outcrops of Ervedosa, which also include the localities of Montalto and Tanjarro.
The palaeobotanical material (Plates I−II) consists of a single specimen (holotype) which is also represented by a counterpart (isotype). Both plant fossil samples are preserved as adpressions (compressions and impressions; Shute & Cleal, Reference Shute and Cleal1986) in laminated grey shales.
2.b. Sample preparation and repository
The fossils of the new species were prepared either manually, using small chisels, hammers and stiff needles, or mechanically with a compressed air and an electrically powered vibro-tool. The specimens were later drawn using a camera lucida setup attached to a stereomicroscope to elucidate the morphology and help with the taxonomic description in Systematics part. Both part and counterpart are stored in the palaeontological collections of the Geological Museum Fernando Real of the University of Trás-os-Montes e Alto Douro (MGUTAD; Vila Real, Portugal).
3. Geological, palaeobotanical and biostratigraphic background
The Ervedosa flora occurs in upper Westphalian D (upper Asturian/upper Moscovian, Middle Pennsylvanian) strata in a separate thrust slice, into two thin strips which surroundings to the NE lower Gzhelian (Upper Pennsylvanian) rocks of Douro Carboniferous Basin (e.g. Teixeira, Reference Teixeira1944; Wagner & Lemos de Sousa, Reference Wagner and Lemos de Sousa1982 a; Wagner, Reference Wagner, Lemos de Sousa and Oliveira1983; Wagner & Álvarez-Vázquez, Reference Wagner and Álvarez-Vázquez2010). In the geological-structural and sedimentological context, Ervedosa has a genesis not very different from the Douro Carboniferous Basin (Pinto de Jesus, Reference Pinto de Jesus2001, Reference Pinto de Jesus2003).
The first studies on the Ervedosa (Westphalian D) flora date back to the 1940s and 1950s, with works by Teixeira (Reference Teixeira1941, Reference Teixeira1942 a, Reference Teixeira1942 b, Reference Teixeira1942 c, Reference Teixeira1943, Reference Teixeira1944, Reference Teixeira1945, Reference Teixeira1951, Reference Teixeira1954) that make reference to a set of plant fossils harvested by this author in the localities of Ervedosa, Tanjarro and Montalto (Wagner & Lemos de Sousa, Reference Wagner, Lemos de Sousa, Lemos de Sousa and Oliveira1983). The original materials are stored in the collections of the Geological Surveys of Portugal in Lisbon and at the Museu de História Natural e da Ciência da Universidade do Porto (MHNCUP; Portugal). The lists initially published by Teixeira (Reference Teixeira1943, Reference Teixeira1944, Reference Teixeira1945, Reference Teixeira1954) for the Ervedosa flora were commented by Wagner (1959, p. 405), and afterwards, revised by Wagner (Reference Wagner, Lemos de Sousa and Oliveira1983, p. 156) and Wagner & Lemos de Sousa (Reference Wagner, Lemos de Sousa, Lemos de Sousa and Oliveira1983, p. 132).
The stratigraphic age of the Ervedosa flora was first discussed by Laveine et al. (Reference Laveine, Coquel and Loboziak1977), making reference to the early-late Westphalian D (Asturian) age with the comparison between the Callipteridium (Praecallipteridium) jongmansii (Bertrand) Wagner and Callipteridium (Praecallipteridium) armasii (Zeiller) Wagner biozones from the Santa Susana Basin (upper Westphalian D) of the Portuguese Carboniferous and Asturian Central Massif in northwestern Spain, respectively (e.g. Wagner, Reference Wagner1966, Reference Wagner, Lemos de Sousa and Oliveira1983; Wagner & Lemos de Sousa, Reference Wagner, Lemos de Sousa, Lemos de Sousa and Oliveira1983; Correia et al. Reference Correia, Šimůnek, Cleal and Sá2019). Wagner (Reference Wagner1971) recorded Callipteridium jongmansii under name of Callipteridium armasii (see also Laveine et al. Reference Laveine, Coquel and Loboziak1977; Wagner, Reference Wagner, Lemos de Sousa and Oliveira1983; Wagner & Lemos de Sousa, Reference Wagner, Lemos de Sousa, Lemos de Sousa and Oliveira1983). This author later described both species as a single group, grouped into a ‘Callipteridium jongmansii-armasii’ complex, which appears in ages from late Westphalian to mid-to-late Cantabrian (Wagner, Reference Wagner, Sutherland and Manger1984, see chart 3). More recently, Wagner and Álvarez-Vázquez (Reference Wagner and Álvarez-Vázquez2010) described Callipteridium jongmansii as only species which appears in the upper Asturian (upper Westphalian D) of Santa Susana Basin and Ervedosa (Portugal) and in the lower Asturian (lower Westphalian D) and lower? Cantabrian of Spain.
Also, in the Laveine (Reference Laveine, Holub and Wagner1977)’s work, there is the reference to the Dicksoniites plueckenetii Sterzel as equivalent to the late Westphalian D age, and in correspondence with the South Wales horizon (Cleal, Reference Cleal1978). However, D. plueckenetii has been recorded in the Carboniferous of Iberian Peninsula (Iberia) in ages from Westphalian D to Stephanian C (early Gzhelian) (e.g. Wagner, Reference Wagner, Lemos de Sousa and Oliveira1983, Reference Wagner, Sutherland and Manger1984; see also Correia et al. Reference Correia, Šimůnek, Sá and Flores2018, Table 1). Other macrofloral elements present, Alethopteris corsinii Buisine is a species that occurs in the Westphalian C and Westphalian D horizons/units of the northeastern France (Buisine, Reference Buisine1961), and the species Linopteris obliqua (Bunbury) Zeiller and cf. Lobatopteris vestita (Lesquerex) Wagner, two biozones which are correspond to the lower and upper units of the Asturian substage of Western European regional scale (equivalent to the middle Westphalian D and upper Westphalian D units, according to Heckel & Clayton, Reference Heckel and Clayton2006), respectively (e.g. Wagner & Lemos de Sousa, Reference Wagner, Lemos de Sousa, Lemos de Sousa and Oliveira1983; Wagner, Reference Wagner, Sutherland and Manger1984; Correia et al. Reference Correia, Šimůnek, Sá and Flores2018, Figure 10). Asterotheca cf. nyranensis Nĕmejc reported in the Ervedosa flora is an another Asturian (Westphalian D) floral record, which is associated with the obliqua and vestita Zones (macrofloral zones) in the floral massifs of Spain and Czech Republic (e.g. Pšenička et al. Reference Pšenička, Zodrow, Mastalerz and Bek2005; Wagner & Álvarez-Vázquez, Reference Wagner and Álvarez-Vázquez2010, pp. 270, 273). In Portugal, this species has been reported in upper Westphalian D (upper Asturian) strata of the Santa Susana Basin (e.g. Wagner, Reference Wagner, Lemos de Sousa and Oliveira1983; Wagner & Lemos de Sousa, Reference Wagner, Lemos de Sousa, Lemos de Sousa and Oliveira1983).
Ervedosa flora is also characterized by the presence of some native or endemic species, such as Linopteris florini Teixeira and Neuropteris (Mixoneura) ervedosensis Teixeira described by Teixeira (Reference Teixeira1941, Reference Teixeira1944, 1Reference Teixeira942 a; Reference Teixeira1942 c; see also Wagner & Lemos de Sousa, Reference Wagner and Lemos de Sousa1982 a; Wagner & Lemos de Sousa, Reference Wagner and Lemos de Sousa1982 b; Wagner, Reference Wagner, Lemos de Sousa and Oliveira1983). L. florini is a long ranging form, which is generally mid-Westphalian (Westphalian B) and early Stephanian (late Cantabrian/early Barruelian), and has been found in Portugal, Spain and Germany (Wagner & Lemos de Sousa, Reference Wagner and Lemos de Sousa1982 a; Wagner & Lemos de Sousa, Reference Wagner and Lemos de Sousa1982 b; Wagner, Reference Wagner, Lemos de Sousa and Oliveira1983). N. ervedosensis has been later redescribed by Teixeira (Reference Teixeira1944) as a probable form of Neuropteris ovata Hoffmann, Neuropteris ovata Hoffmann forma ervedosensis Teixeira. It is noted that according to Wagner and Lemos de Sousa (Reference Wagner, Lemos de Sousa, Lemos de Sousa and Oliveira1983), Neuropteris ovata forma ervedosensis is a species in its own rights and probably unrelated to Neuropteris ovata. Other floral elements such as Lobatopteris micromiltonii (P. Bertrand) Wagner and Macroneuropteris scheuchzeri (Hoffmann) Cleal, Shute & Zodrow compose the Ervedosa flora, which tend to suggest mid-Westphalian D (Asturian) to late Cantabrian (early Stephanian) ages attributed to the floral massifs of Iberian Peninsula (e.g. Wagner & Lemos de Sousa, Reference Wagner, Lemos de Sousa, Lemos de Sousa and Oliveira1983; Wagner, Reference Wagner, Sutherland and Manger1984; Wagner & Álvarez-Vázquez, Reference Wagner and Álvarez-Vázquez2010, see comments in p. 276). These latter species suggest a higher age than that previously established, for a Cantabrian age, but the set of species above-mentioned places flora of Ervedosa with age most likely to the late Westphalian D.
4. Systematic description
The taxonomic classification system is according to Taylor et al. (Reference Taylor, Taylor and Krings2009) and Novikoff and Barabasz-Krasny (Reference Novikoff, Barabasz-Krasny, Novikoff and Barabasz-Krasny2015).
Division: †Progymnospermophyta Bold et al.
Class: †Noeggerathiopsida Kryshtofovich
Order: †Noeggerathiales Nĕmejc emend. Wang et al. (Reference Wang, Hilton, Pfefferkorn, Wang, Zhang, Bek, Pšenička, Seyfullah and Dilcher2021)
Fossil-family: Discinitaceae Zhifeng & Thomas
Fossil-genus: Palaeopteridium Kidston (Reference Kidston1923)
Type species: Palaeopteridium reussii (Ettingshausen) Kidston (Reference Kidston1923) from Westphalian Series of the Carboniferous Rocks of Great Britain.
Basionym: Asplenites reussii Ettingshausen (Reference Ettingshausen1852), ‘Steinkohlenflora von Stradonitz in Böhmen’, Ablandl. K.K. geol. Reichsanst., Band I, Abth. 3, No. 4, p. 16, pl. I, Figures 8, 9 (Kidston, Reference Kidston1923, p. 201)
Palaeopteridium andrenelii sp. nov.
Etymology. The specific name ‘andrenelii’ honours André Nel from Muséum National d’Histoire Naturelle (Paris), a world expert in palaeoentomology and honourable colleague who has been cooperating with us on the systematic study of new insect fossils recently described in the Douro and Buçaco Basins (Loureiro et al. Reference Loureiro, Correia, Nel and Pinto de Jesus2010; Correia et al. Reference Correia, Nel, Sá, Domingos, Carneiro and Flores2014 b; Correia et al. Reference Correia, Schubnel and Nel2021 c; Correia & Nel, Reference Correia and Nel2023; Correia et al. Reference Correia, Pereira, Cavaleiro, Correia, Sá and Nel2023 c).
Holotype. Specimen MGUTAD-1121 (Plate I, 1−3) is composed of a counterpart – isotype MGUTAD-1122 (Plate II, 1).
Type locality. Outcrops located in the northeast banks of road Dom Miguel, Seixo (Fânzeres) region, Gondomar, northwestern Portugal (Fig. 1).
Type horizon and age. Horizon composed of laminated and compact grey shales; late Westphalian D (late Asturian/late Moscovian, Middle Pennsylvanian).
Diagnosis. Bipinnate frond with subopposite or alternate pinnules, relatively symmetrical and spaced, sometimes touching or slightly overlapping, obliquely attached to an ultimate rachis by a very narrow base (1.5–2.0 mm long and 0.8–1.0 mm wide) − petiolate attachment. Ultimate rachis very thin, straight or slightly flexuous, detached from a possible penultimate rachis. Fan-shaped (flabelliform or semiflabelliform type) pinnules, of 7–8 mm long and a maximum fan width of about 5−6 mm, with entire lateral and crenate/lobed distal margins. Distal margins bearing irregular and asymmetrical 12−14 crenate-shaped teeth/lobes of about 0.3−1.0 mm long. Venation pattern of open dichotomous showing a single vein entering each pinnule, giving rise to a series of radiating veinlets of equal strength which in their course to the margin divide three or four times. There are 14–20 veins on distal margin of pinnule.
Description of the holotype. The frond is at least up to 37 mm long, with subopposite or alternate pinnules, relatively symmetrical and spaced, sometimes touching or slightly overlapping (Plate I, 3), and obliquely attached on an ultimate (last order) rachis by a very narrow base (1.5–2.0 mm long and 0.8–1.0 wide) − petiolate attachment (Plate I, 2, 3, white arrows; Plate II, 2A). The ultimate rachis is very thin, of 0.2−0.5 mm wide, which is detached from a possible penultimate rachis (petiole-like) up to 15 mm long and 1.0−1.5 mm wide (Plate I, 1, white arrows). The petiole displays a prominent central vein. The pinnules are 7–8 mm long and 5−6 mm wide and fan-shaped (flabelliform or semiflabelliform type), having entire lateral and crenate/lobed distal margins. The crenate/lobed margins have 12−14 crenate-shaped teeth/lobes, relatively irregular and asymmetrical, of about 0.3−1.0 mm long. The venation is dichotomous displaying veinlets of equal strength, well visible, and which in their course to the margin are divided in three to four times (Plate II, 1, 2). The veins are very few in number − there are 14–20 veins on distal margin of pinnule.
Remarks. Some upper pinnules of the frond appear to display slightly cleft or slotted distal margins (see Plate I, 2), but it is likely this is damage on the pinnule lamina due to the taphonomy (see holotype counterpart and restoration of foliage in Plate II, 1, 2A).
5. Discussion
5.a. Comparisons
In comparison with other noeggerathialean species, Palaeopteridium andrenelii sp. nov. (Plates I, 1–3; Plate II, 1) bears a close resemblance to Palaeopteridium reussii (Ettingshausen) Kidston, Reference Kidston1923. This species occurs in the Duckmantian and Bolsovian (Westphalian B, C) of Bohemia (Czech Republic), lower Asturian of Spain and Westphalian Series of South Staffordshire Coalfield (Britain) and Namurian (Serpukhovian−Bashkirian) from West-Central of Illinois (USA) (Kidston, Reference Kidston1923; Leary & Pfefferkorn, Reference Leary and Pfefferkorn1977; Wagner & Álvarez-Vázquez, Reference Wagner and Álvarez-Vázquez2010). P. reussii bears semiflabelliform to flabelliform (fan-shaped) pinnules with dentate distal margins in which, showing 10−20 dentate-teeth and having 10 to 20 veins which are divided up to three times (e.g. Kidston, Reference Kidston1923, pp. 201, 202, pl. LV; Leary & Pfefferkorn, Reference Leary and Pfefferkorn1977, p. 16−19, Figure 7A−C, pl. 4, Figures 1−5; Wagner & Álvarez-Vázquez, Reference Wagner and Álvarez-Vázquez2010, p. 274, pl. XVII, Figures 5, 6). The major difference between P. reussii and P. andrenelii sp. nov. is in the form of the distal margins of the pinnules. The distal pinnule margins of P. reussii are notably dentate to serrate (e.g. Kidston, Reference Kidston1923, pl. LV, Figure 2a; Leary & Pfefferkorn, Reference Leary and Pfefferkorn1977, Figure 7A−C; Leary & Trask, Reference Leary and Trask1985, pl. 4, Figure 1), while those of P. andrenelii sp. nov. are relatively crenate or lobed (Plate II, 2). Other differences occur in the lower and upper parts of the pinnules. The pinnules of P. reussii have an apex shape more circular, while pinnules of P. andrenelii sp. nov. show irregular distal margins. On the other hand, the pinnules of the Portuguese species shows a more petiolate attachment (Plate II, 1, 2) when compared to the pinnule attachment of P. reussii pinnules which is stronger and shorter or less developed and, sometimes, it is highlighted by the absence of a petiole (e.g. Kidston, Reference Kidston1923, pl. LV; Leary & Pfefferkorn, Reference Leary and Pfefferkorn1977, Figure 7A−C; pl. 4, Figures 1−5; Wagner & Álvarez-Vázquez, Reference Wagner and Álvarez-Vázquez2010, pl. XVII, Figures 5, 6; see summary of the main characteristics of foliage for comparison in Table 1).
P. andrenelii sp. nov. shares with Rhacopteris lindseaeformis Bunbury sp. (= Rhacopteris [Anisopteris] paniculifera Stur, Reference Stur1875) the semiflabelliform-flabelliform pinnule shape with crenulate distal margin, but it differs from the latter in the symmetry of pinnules. The pinnules of R. lindseaeformis are notably asymmetrical, having one of lateral margins more developed and distal margins cleft into segments (Kidston, Reference Kidston1923, pp. 204, 205, text-Figure 9, pl. XLIX, Figures 1−4, pl. LI, Figure 2; Boureau & Doubinger, Reference Boureau and Doubinger1975, pp. 624, 627, Figures 530, 534; Taylor et al. Reference Taylor, Taylor and Krings2009, pp. 555, 562, Figure 14.79). In addition, P. andrenelii sp. nov. differs from R. lindseaeformis in the venation pattern. R. lindseaeformis pinnules show radiating veinlets of equal strength which, in their course to the pinnule margin, are divided two or three times. In contrast, the pinnules of P. andrenelii sp. nov. display veinlets of equal strength, well visible, and which in their course to the margin are divided in three to four times (Plate II, 1, 2; see Table 1). R. lindseaeformis is described from the Carboniferous Limestone Series and is there restricted to the Lower Limestone Group but is widely distributed in the Oil-Shale Group of the Carciferous Series (Mississippian) of Britain (Kidston, Reference Kidston1923, p. 207; Taylor et al. Reference Taylor, Taylor and Krings2009, pp. 555, 562).
P. andrenelii sp. nov. also is comparable to the Rhacopteris inaequilatera (Göppert) Stur, Reference Stur1875 (= Anisopteris inaequilatera Göppert) which has semiflabelliform to flabelliform (fan-shaped) pinnules with finely crenulate distal margins. Nevertheless, P. andrenelii sp. nov. differs from R. inaequilatera in the apex shape of pinnules. R. inaequilatera pinnules show notably truncate-shaped apices (Kidston, Reference Kidston1923, p. 208, pl. XLIX, Figures 5, 6, pl. LI, Figure 3; see also Boureau & Doubinger, Reference Boureau and Doubinger1975, p. 621, Figure 528). There is also some dissimilarity in the venation pattern between two species. In R. inaequilatera, the veins dichotomize two or three times when they reach the pinnule margin (detailed description in Kidston, Reference Kidston1923, p. 208), while those of P. andrenelii sp. nov. are divided in three to four times (Plate II, 1, 2; see Table 1). R. inaequilatera occurs in lower Carboniferous/Mississippian (Calciferous Sandstone Series – Oil-Shale Group) of Britain (Kidston, Reference Kidston1923; Boureau & Doubinger, Reference Boureau and Doubinger1975).
5.b. Generic placement
The new fossil species described herein (Plates I, 1–3; Plate II, 1) shares macro-morphological features with Palaeopteridium Kidston, Reference Kidston1923 and Rhacopteris Schimper, Reference Schimper1869. Rhacopteris, similarly as Palaeopteridium belongs to the noeggerathialeans. Both fossil-genera are typical floristic elements of the Euramerican floral realm and bear very similar morphological traits in the pinnule general shape, bearing semiflabelliform to flabelliform (fan-shaped) pinnules, of which are compared favourably with the new Portuguese species. However, Palaeopteridium and Rhacopteris differ in the apical margin of pinnules. Rhacopteris has usually more divided pinnules (i.e. strongly slotted leaflets) than Palaeopteridium (e.g. Kidston, Reference Kidston1923; Boureau & Doubinger, Reference Boureau and Doubinger1975) [some species with non-slotted leaflets, such as ‘Rhacopteris circularis’ Walton, 1926 (see Boureau & Doubinger, Reference Boureau and Doubinger1975, p. 622), have been classified as belonging to the genus Rhacopteris; however, this taxonomic misunderstanding needs to be revised and the species should be transferred to the genus Palaeopteridium]. Palaeopteridium has usually only teeth in distal pinnule margin (e.g. Leary & Pfefferkorn, Reference Leary and Pfefferkorn1977; Pfefferkorn & Wang, Reference Pfefferkorn and Wang2016, Figure 4). The pinnules of the new Portuguese species are not slotted, having dentate or lobed distal margins (Plate I, 2, 3; Plate II, 2A).
Palaeopteridium andrenelii sp. nov. also shares some basic features (fan-shaped pinnules) with the fossil-genera Anisopteris (Oberste-Brink) Hirmer, Reference Hirmer, Hirmer and Gothan1940 (see Boureau & Doubinger, Reference Boureau and Doubinger1975, p. 621) and Nothorhacopteris Archangelsky, Reference Archangelsky1983 (see Taylor et al. Reference Taylor, Taylor and Krings2009, pp. 651, 677). Anisopteris belongs to quite different group [all the ‘Anisopteris’ species described in literature are different by stratigraphy and by systematic classification and the legitimacy of genus as valid name is questionable (various species classified to the genus Anisopteris were previously attributed to the genus Rhacopteris; see Kidston, Reference Kidston1923, pp. 203−9; Boureau & Doubinger, Reference Boureau and Doubinger1975, pp. 623−7)]. The South American Carboniferous−early Permian Nothorhacopteris is of a different floristic province, and it belongs to different taxonomic group (Taylor et al. Reference Taylor, Taylor and Krings2009, p. 677).
Therefore, on the basis of these generic diagnostic morphological features and the phytogeographical differences, the new Portuguese species fits well into the fossil-genus Palaeopteridium.
5.c. Reproductive structures associated with Palaeopteridium-type fronds and its parental relationship
Leary (Reference Leary1980) described a specimen of Lacoea Read (Reference Read1946), a spore- or pollen-bearing organ (sporophyll type) with affinities to early gymnosperms (progymnosperms), from the Lower Pennsylvanian deposits of Rock Island County, Illinois (USA). The Leary’s specimen is found closely associated with Archaeopteris-like foliage known as Palaeopteridium, which has previously been suggested as the foliage of the Lacoea-bearing parent plant (Leary, Reference Leary1980). Lacoea has been compared with Discinites, one of the several fossil-genera of fructifications described by Leary (Reference Leary1973) (see also Leary & Pfefferkorn, Reference Leary and Pfefferkorn1977, p. 16). Discinites was established by K. Feistmantel (Reference Feistmantel1879) and revised by Bek and Šimůnek (Reference Bek and Šimůnek2005). Feistmantel (Reference Feistmantel1879) believed that Discinites showed affinity to sphenopsids (see also Wang et al. Reference Wang, Zhang, Bek and Pfefferkorn2004, p. 243). Nĕmejc (Reference Nĕmejc1937, Reference Nĕmejc1941) added several fossil species to the genus and found that these strobili appear together with fronds of Palaeopteridium and Rhacopteris. Several other authors (Nĕmejc, Reference Nĕmejc1937, Reference Nĕmejc1950; Hirmer, Reference Hirmer, Hirmer and Gothan1940; Zimmermann, Reference Zimmermann1959; Leary & Pfefferkorn, Reference Leary and Pfefferkorn1977; Remy & Remy, Reference Remy and Remy1977; Meyen, Reference Meyen1987; Gao & Thomas, Reference Gao and Thomas1994; Wang & Shen, Reference Wang and Shen2000) referred Discinites to noeggerathialean plants (Wang et al. Reference Wang, Zhang, Bek and Pfefferkorn2004). Leary & Pfefferkorn (Reference Leary and Pfefferkorn1977) included a tentative suggestion of a possible evolutionary sequence from an Archaeopteris-like form through forms similar to Noeggerathiales and Lacoea to Discinites (Leary, Reference Leary1980). Wang (Reference Wang2000) described Discinites sinensis as the first noeggerathialean strobilus that has been found as a petrifaction from the Permian of Shizuishan, Ningxia (China), yielding information on the structure of the Discinites spores. Bek (Reference Bek2001) and Bek and Šimůnek (Reference Bek and Šimůnek2005) provided a new knowledge about the morphology and phylogeny of Discinites spores based on adpression materials (Wang et al. Reference Wang, Zhang, Bek and Pfefferkorn2004). In both cases, the reproductive structures were found not to be biologically connected to frond, which is why it is not possible to conclude which of the examples represents a clear affinity. Perhaps even more interesting, both Lacoea and Discinites produce Calamospora-type spores (see Leary, Reference Leary1980; Wang et al. Reference Wang, Zhang, Bek and Pfefferkorn2004, Table 2). This last fact means that these reproductive structures may actually have belonged to the same noeggerathialean parent plant.
In the new noeggerathialean fossil described here, two very small reproductive structures are in close association with the frond fragment (Plate I, 4, 5; Plate II, 1, 2). They have lengths of about 4 to 5 mm and a width of about 2−3 mm. These structures most likely correspond to macrosporangia. One of the ‘macrosporangia’ shows a multicelled gametophyte extending from ruptured spore wall (Plate I, 5, white arrows; Plate II, 2B). Although not organically linked, both structures could belong to parent plant (noeggerathialean frond) described above. However, better preserved and articulated specimens are necessary to confirm the identity of these structures.
6. Biostratigraphic and palaeoecological considerations
A diverse macroflora was found together with Palaeopteridium andrenelii sp. nov., which is composed of several groups of plants, consisting in a different rate of marattialean ferns (e.g. Asterotheca, Acitheca, Cyathocarpus, Lobatopteris), sphenopsids (e.g. Sphenophyllum, Annularia, Calamites, Asterophyllites, Calamostachys), pteridosperms (e.g. Linopteris, Callipteridium, Neuropteris, Odontopteris, Macroneuropteris, Sphenopteris), cordaitopsids (Cordaites), and ginkgoopsids (Dicranophyllum). The presence of species stratigraphically important such as Sphenophyllum cf. majus Bronn (cf. Sphenophyllum emarginatum (Brongniart) Koenig, Macroneuropteris scheuchzeri (Hoffmann) Cleal, Shute & Zodrow, Neuropteris cf. flexuosa Sternberg (cf. Neuropteris machadicostai Teixeira), and Callipteridium jongmansii (Bertrand) Wagner confirms the late Westphalian D/Cantabrian age previously established to the Ervedosa flora.
From a palaeoecological point of view, the dominance of taxa of pteridopsids (ferns), sphenopsids, and pteridosperms indicate that the resident floras were essentially composed by mesophytes to hygrophytes, of which grown in clastic wetlands during intervals of moist subhumid climate. This scenario is also based on the absence or rarity of ‘walchian’ conifers and other derived taxa (primarily gymnosperms) typical of dry subhumid and semiarid climate regimes (see in Bashforth et al. Reference Bashforth, Dimichele, Eble, Falcon-Lang, Looy and Lucas2021, Table 1, the segregation of habitats, floral types, and main plant groups that existed in tropical Euramerican Pangaea during the Pennsylvanian, according to seasonality of precipitation and climate).
7. Conclusions
This study describes a new noeggerathialean species, Palaeopteridium andrenelii sp. nov., based on adpression (compressions-impressions) fossils from the classical Westphalian outcrops of Ervedosa (Gondomar), in northwestern Portugal. This new endemic species is the first record of the fossil-genus Palaeopteridium in the Portuguese Carboniferous. Putative reproductive structures, possibly isolated macrosporangia, found in close association with the frond fragment, might also represent the first record of these structures for this genus.
Palaeopteridium andrenelii sp. nov. has palaeoecological affinities with the noeggerathialean genus Rhacopteris, with which shared the same clastic wetland environments in Ervedosa, in western Iberia, during intervals of moist subhumid climate in the late Moscovian (Middle Pennsylvanian).
Acknowledgements
The authors thank two anonymous reviewers for their constructive comments that helped improve the submitted manuscript. This study was supported by Portuguese funds by Fundação para a Ciência e a Tecnologia, I.P. (Portugal) in the frame of the UIDB/00073/2020 (DOI 10.54499/UIDB/00073/2020; https://doi.org/10.54499/UIDB/00073/2020), UIDP/00073/2020 project of the I&D unit Geosciences Center (CGeo). CAG-M would like to acknowledge financial support from Fundação para a Ciência e Tecnologia (FCT, Portugal) Individual Call to Scientific Employment Stimulus contract number 2022.00017.CEECIND/CP1730/CT0001 (https://doi.org/10.54499/2022.00017.CEECIND/CP1730/CT0001) and the project UIDB/50027/2020.
Competing interests
The authors declare that they have no known conflicts of interest.