Mesozoic gymnosperms: megafloral changes through the Triassic–Jurassic of Eastern Gondwana

Abstract

Pteridosperms are preserved abundantly in the Gondwanan Triassic; in particular, the umkomasiaceans are the most commonly recorded floral element but they are not represented subsequently in that region or beyond. Indeed, nearly all umkomasiacean fronds are unknown above the Norian. However, representatives of one genus, Zuberia Frenguelli 1943 emend. Artabe 1990, persisted almost to the close of the Triassic. Umkomasiacean fronds probably belong to at least four genera (Dicroidium Gothan 1912 emend. Townrow 1957, Johnstonia Walkom 1924b, Xylopteris Frenguelli 1943 emend. Pattemore 2016b, Zuberia). Several commonly occurring umkomasiacean species are re-assessed with particular attention to their type specimens. The view that a morpho-continuum links umkomasiacean fronds is shown to be erroneous, being based on greatly enlarged circumscriptions of some species and misidentification of numerous specimens. Many species exhibit considerable morphological variation that has been attributed to a hybridization model of speciation. This is an improbable explanation given that hybridization is very rare among gymnosperms. Allopatric speciation resulting from geographic and climatic provincialism is a more likely explanation for the morphological diversity represented in Middle and Late Triassic floras of Australasia and elsewhere in Gondwana. Representatives of Umkomasia Thomas 1933 and Pteruchus Thomas 1933 emend. Townrow 1962 from the Carnian–lower Norian Ipswich Basin of Queensland are confirmed to have a bipinnate structure. The latter genus is also recorded for the first time from the Carnian Tarong Basin, Queensland. The holotype of Stachyopitys simmondsii Shirley 1898 from the Ipswich Basin, having been regarded as a male fructification for over 100 years, is shown to be ovuliferous. The species is included in Umkomasia and may be synonymous with the type species of that genus. Genuine male fructifications previously identified as S. simmondsii or P. simmondsii (Shirley 1898) Jones & de Jersey 1947a and a number of other Gondwanan species that were originally included in Townrovia Retallack 1981 or Stachyopitys Schenk 1867 belong to P. minor Thomas 1933. Townrovia was erected with an inaccurate diagnosis and the genus has not been convincingly distinguished from Pteruchus. Furthermore, Stachyopitys may not be genuinely represented in the Gondwanan Triassic. No reproductive organ has been confidently allied with Linguifolium Arber 1917 emend. Pattemore & Rigby in Pattemore et al. 2015b, its pteridospermous affiliation being based on limited cuticular evidence. The genus is known chiefly from the Middle and Upper Triassic of Gondwana but its representatives apparently extend into the lowermost Jurassic beyond Gondwana and possibly in Western Gondwana. Speciation in the genus has been largely determined statistically using leaf size; however, a number of other characters identified herein provide a basis for a more convincing separation of species. Consequently, the spatio-temporal distributions of some species have been more precisely defined. Australasian post-Triassic strata contain strikingly different floral assemblages to those of the Triassic. Eastern Gondwanan Lower and Middle Jurassic floras largely comprise ferns, lycopods, conifers, cycads and bennettitaleans. Ginkgoaleans were present in Australia until near the end-Triassic but were absent below ca 60oS palaeolatitude during Early and Middle Jurassic time. No Eastern Gondwanan fructifications from that interval are clearly pteridospermous and no remains show any obvious relationship with pteridosperms of the Gondwanan Triassic. Fern-like fronds from the Lower Jurassic through Eocene of Eastern Gondwana, claimed to be pteridospermous, lack supporting evidence of such affiliation. Bona fide caytonialean fructifications are unrecorded from Australasia and their representation elsewhere in the Eastern Gondwanan Triassic and Jurassic remains speculative. Caytonialean leaves have been reported from the Eastern Gondwanan Upper Triassic–Middle Jurassic, albeit rarely. However, many specimens are attributable to Scoresbya Harris 1932 or to other genera exhibiting reticulate venation. Australasian and Antarctic assignments to Palissya Endlicher 1847 emend. Florin 1958 are structurally distinct from those of the northern hemisphere and may belong to the Knezourocarponaceae Pattemore in Pattemore et al. 2014. The higher taxonomic placement of the family is uncertain; however, affinity with ginkgoaleans or pteridosperms is improbable.