Psilophytites HØEG , 1952

Cleal, Christopher J. & Thomas, Barry A., 2021, Naming Of Parts: The Use Of Fossil-Taxa In Palaeobotany, Fossil Imprint 77 (1), pp. 166-186 : 177-178

Psilophytites HØEG, 1952

Høeg (1952) created this fossil-genus to be similar to Hostinella as he had defined it, but for axes with “… spreading undivided spines.” Most of the specimens that Høeg listed as typifying the fossil-genus had either not been published with a species name or had been assigned (incorrectly, according to Høeg) to other, better-known species such as Psilophyton princeps DAWSON. However , one was the holotype of Psilophyton rectissimum HØEG ( Høeg 1945) , which consisted of a spiny axis lacking reproductive structures or anatomy, exactly as Høeg had defined Psilophytites . Since it is unlikely that reproductive structures or cell structure will ever be found for the P. rectissimum holotype, it is unlikely to be ever transferred to another, more natural taxon. It therefore makes complete sense also to designate the Høeg (1945) specimen the generic lectotype for Psilophytites , thereby effectively stabilising the name for the fossil-genus envisaged by Høeg (1952).

Pecopteris (BRONG.) STERNB., 1825 (≡ Filicites (Pecopteris) BRONGN., 1822 )

Fern-like frond fragments with small dentate pinnules are common in many Carboniferous and Permian adpression floras. When evidence of reproductive structures has been available, they have been assigned to one of various fossilgenera (e.g., Acitheca SCHIMP. , Cyathocarpus C.E.WEISS and Seftenbergia CORDA ; Corda 1845, Weiss 1869, Schimper 1879) but for many species these structures are unknown (many such species can be found in the monographs by Corsin (1951) and Dalinval (1960)). Traditionally these fossils have been assigned to the fossil-genus Pecopteris , adopting the original diagnosis given by Brongniart (1822), but this overlooks the fact that the reproductive structures of the type of Pecopteris ( P. penniformis ) are well documented: they compare with the sori of the tedeleacean fern Senftenbergia ( Bertrand 1912, Bek and Pšenička 2001, Frojdová et al. 2020).

There are two potential solutions to this problem. The original Brongniart (1822) taxon could be retained (with the change of rank suggested by Sternberg 1825), and all fernlike fossil fronds with small, linguaeform to dentate pinnules assigned to Pecopteris , irrespective of affinities. This has the advantage of inclusivity, as no fossils of this type need be excluded, but has the disadvantage of lacking any taxonomic resolution. Since it is now accepted that at least two distinct orders of ferns ( Marattiales and Filicales) have fronds with such dentate pinnules, and they even occur in some seedplants (e.g., Callistophytales ‒ e.g., Crookall 1930), such a lack of resolution would be clearly disadvantageous.

It was for this reason that a more natural classification was developed that takes into account the reproductive structures (summarised by Cleal 2015). But what happens to the species for which the reproductive structures are unknown? It would be possible to create a new fossil-genus for the fossil-species for which reproductive structures are unknown. But for this to be a stable concept, the type would have to be chosen such that reproductive structures would be impossible or unlikely ever to be found – otherwise, as soon as those structures were discovered, the type would be excluded from the fossil-genus and a new name would need to be created. The only real solution is to make comparisons with the framework species of the classification (i.e., those for which sori/synangia are known) based on morphological features of the sterile pinnae, and named maybe using open nomenclature ‒ not a totally satisfactory solution but maybe the only practical one.

Lepidostrobus variabilis LINDL. et HUTTON emend. by Chaloner (1953)

Brongniart (1828b) named and briefly described as the holotype of Lepidostrobus a fossilised cone figured by Parkinson (1804). From then on, all cones that even vaguely fitted Brongniart’s morphological criteria were included in this genus. Sporophyll morphology later came to be regarded as the most important character for taxonomic purposes. Lepidostrobus was generally accepted to be a heterogenous group including both bisporangiate cones with megaspores and microspores, and microsporangiate cones with just microspores.

However, it eventually became obvious that, wherever possible, the in-situ spores in the cones need to be thoroughly investigated for an accurate determination ( Chaloner 1953, Felix 1954). The earlier taxonomic efforts emphasized the characteristics of the megaspores rather than those of the microspores. Although microspores from Lepidostrobus can be correlated with the genus Lycospora J.M.SCHOPF, L.R.WILSON et BENTALL ( Schopf et al. 1944) , it is often difficult to relate them to the dispersed spore species. Nevertheless, studies of in-situ spores have suggested that there are at least two groups of Lycospora : those with distinctive proximal ornamentation and broad equatorial flange occur within microsporangiate Lepidostrobus strobili, whilst those with smooth proximal surfaces lacking broad equatorial flanges occur within bisporangiate strobili ( Thomas 1970 b, 1978, Thomas and Dytko l980). Since the type of Lepidostrobus contained microspores of the type found in microsporangiate cones, Brack-Hanes and Thomas (1983) redefined Lepidostrobus as a fossil-genus of microsporangiate strobili; bisporangiate cones formerly included in Lepidostrobus were transferred to Flemingites CARRUTH. ( Carruthers 1865) .

So, what should the strobili that lack evidence of their spore content be called? They originally tended to be referred to as Lepidostrobus sp. but there was also a tendency to give them specific names based on morphological features. This could imply they were microsporangiate cones, which might not necessarily be correct and could be misleading. Chaloner (1953) suggested that the fossil-species Lepidostrobus variabilis as emended by Arber (1922) should be retained as a designation for such barren cones (the revision of this species by Sen (1958) should probably be rejected as having been based on contaminated samples ‒ Bek 2012). This seems to us as an admirable suggestion and would leave no doubt about it use when applied to figured cones or identifications in species lists.