FLORISTICS OF THE PERMIAN AND TRIASSIC GONDWANAS OF INDIA HARI K. MAHESHWARI Birbal Salmi Institute of Palaeobotany, Lucknow-226007 ABSTRACT Indian Peninsula The boulder bed is over• A synthesis of the floristics through the Upper lain by green laminated shales and sand• Palaeozoic and Lower Mesozoic Gondwanas of stones showing cold climate. The climate India is presented. Definite lycopsids are known gradually warmed up as is seen from the only from Middle Permian but could have been present in other ages as well.. The solitary spheno• diver~ification of the flora. The succeeding phyll SpeClOSranges from Middle Permian to basal PermIan strata contain thick coal seams Triassic. Equisetales are known by vegetative which were laid down in a warm humid shoots only, but a change in form through the time is noticed. Fertile filicinean remains known arC! climate. The early Triassic, rocks contain mostly referable to the family Asterothecaceac of undecomposed felspar which supposedly the Marattiales. The most important element of indicates arid conditions (Wadia, 1975). the period, i.e. Glossopteris-complex reprosents at The younger Triassic has thick red sandstone least two orders, viz., Ottokariales and Lidgetto• niales, as is evident from the fertile organs. This deposits which are supposed to denote complex continues into the basal Triassic and is arid desertic conditions. Robinson (1967, gradually replaced by Mosozoic 'Pteridosperm' p. 230) suggests a monsoon type of climate families Corystospermaceae and Peltaspermaceae. for the Indian Triassic. A classification Cordaltalea:l type of leJ.ves are more common in the Lower Permian; their affinity, however, is of the Permian and Triassic Gondwanas uncertain. Cycadales, Ginkgoales and Coniferales of India is shown in the table (slightly though present are meagrely represented. modified from Shah, Singh & Sastry, 1971). Plant fossils are quite abundant in the INTRODUCTION Permian and Triassic systems of the Indian Gondwana Sequence. These are of great biological interest as they comprise certain plant groups which are not represented is foundPalaeophyte-Mesophytepreserved in a thick insequenceIndia THEof mostly continental deposits known in the contemporary floras of the Northern as the Gondwana Sequence. This enormous Hemisphere. In the period under review, sequence of rocks ranges in age from the two distinct floras have been found. The Lower Permian (may be Upper Carbon,i• older, Glossopteris flora is characteristic ferous) up t.o the Lower Cretaceous (in• of the Permian though it extends into the clusive). Parallel sequences of continent.al basal Triassic. The younger, the Dicroidium deposit.s are also kno~n from the Southern flora is confined to the Triassic System only. Hemisphere cont.inents, viz., Australia, The plant groups represented in the two Africa, South America and Antarctica. It is floras are: ?Bryophyta, Lycopodophyta, supposed that the rocks of t.he Gondwana Arthrophyta, Pterophyta, Glossopterid 0• Sequence in India were deposited in slowly psida, ' Pteridospermophyta " Cycadophyta, sinking faulted troughs fed by rivers of the Ginkgophyta and Coniferophyta. In addition Gondwana country (Wadia, 1975, p. 165). there are certain plaJl,ts whose taxonomic In a few instances the deposits could have position is uncertain. been laid down in lakes. The sequence starts with a boulder bed, containing the BRYOPHYTA characteristically glaciated, striated and facetted blocks of rock embedded in a fine This group is probably represented by a silt-like matrix, suggesting a fluvio-glacial solitary capsule-like structure, Capsulites agency of transport. and deposition. The gondwanensis Saksena, obtained from a boulder bed forms a conspicuous and charac• Lower Permian shale. According to Saksena teristic datum line in the geology of the (1958), the distinct division of this structure 146 THE PALAEOBOTANIST into foot, seta and capsule, the possibility dendron by Hislop, and later as " Stigmaria? of thE' presence of an annulus and the (Portion of the Rhizome of a Fern?)" by indication of a spore-sac extending along Bunbury (1861, pI. 12, fig. 2). The species the entire length of the capsule places it is heterosporous, its sporophylls being distri• near the Muscineae. The exact affinities buted on normal leafy shoots, without any of the specimen, however, still remain organized cone (Krausel, 1961). In the uncertain. There is a distinct possibility latter character it resembles the Lower that this structure represents a seed. Carboniferous Lepidodendropsis Lutz, the Upper Carboniferous Pinakodendron \iVeiss and probably Omphalophloios White. This LYCOPODOPHYTA could probably represent an unique line of evolution from the Lower Carboniferous Lycopsid megafossils in the Permian• to the Permian or may be that the four Triassic of India are extremely rare. The genera evolved independently of each other oldest records of the lycopsids are from the (Chaloner & Boureau, 1967, p. 510). Lower Permian Gangamopteris beds of Zewan From the Triassic of Beli, South Rewa Spur, Srinagar District, Kashmir. A frag• Gondwana Basin, another lycopsid species, mentary specimen showing crescentic ligular Lycopodites sahnii Lele (1962, pI. 1, figs. scars inside spirally disposed rhomboid leaf 1, 2, text-figs. 1, 2) is known. The frag• cushions has been tentatively identified with mentary axis bears spirally disposed linear Lepidodendron (Kapoor, 1969, pI. 2, fig. 1). delicate leaves almost at right angles to the Another specimen shows a cone-like struc• axis. The leaf-scars show a centrally situated ture probably attached to a lycopsid stem minute depression probably representing thE: (Kapoor, 1969, pI. 2, fig. 2). The presence vascular trace. The genus Lycopodites con• of a ligule in these specimens has been tinues into the Jurassic where it is repre• doubted by Surange (1971, p. 65) who opines sented by the species L. gracilis (Oldham & that "The scars described and figured as Morris) Seward & Sahni. ligules seem to correspond to the elevati?ns While the lycopsid megaplant remains are which are left on cushions when leaves dned scarce in the Permian-Triassic Gondwana without abscission like those in Sublepido• Sequence of India, a large number of dendron." A better preserved and almost microspores and megaspores, probably of complete lycopsid-like cone was found in the selaginelloid affinity, have been recorded tuffaceous shales of Liddar Valley near throughout the sequence (Bharadwaj & Pahlgam, Anantnag District, Kashmir (Sri• Tiwari, 1970; Maheshwari & Banerji, 1975). vastava & Kapoor, 1969, pI. 5, figs. 1-3, This does indicate the possibility that many text-fig. 1). The cone Lepidostrobus kash• more species of lycopsids flourished than IS mirensis is cylindrical and pedunculate. evident from the megafossil records. The The sporophylls are closely placed a~d paucity of lycopsid megaplant remains can spirally arranged. Detailed structure Of.t~IS probably be attributed to their delicate cone is not known and hence its affin1tIes nature and their suppression by an ove~• are as yet uncertain. whelming preponderance of gymnosperm1c In the peninsula the oldest lycopsid mega• megaplant remains (Maheshwari, 1974, p. 52). fossil record is from the MiddJe Permian Ironstone Shale Formation. The solitary species, Cyclodendron leslii (Seward) Krausel ARTHROHYTA is represented only by fragments of stems, which bear a number of spirally arranged, EQUISET ALES eye-shaped leaf-scars. Each leaf-scar has a more or less circular boss which perhaps The equisetalean remains, represented by indicates the position of the vascular supply. vegetative shoots only, belong to the genera: The ligular pit or the parichnos scars have Schizoneura, Phyllotheca, Lelstotheca, Bara• not been found (Kar, 1968, pI. 1, figs. 1, 2). karia and Raniganjia. In the absence of This species probably persisted into the fructifications, the relationships of the above Lower Triassic Mangali beds, from where a genera with one another or their affinities similar specimen was described as ?Lepido- within the group Equisetales are as yet MAHESHWARI - FLORISTICS OF PERMIAN AND TRIASSIC GONDWANAS 147 obscure. As far as their stems are con• shows that most of them had a similar cerned, they are all similar in being jointed epidermal pattern. In Raniganjia benga• and ribbed, the ribs continuing through lensis the guard cells are reported to possess successive nodes. In thp latter character obscure radiating striations like those seen they resemble the stems of Asterocalamites. in the modern Equisetum (Pant & Nautiyal, In that genus, however, the leaves are 1968). This character is not known either dichotomously forked many times. Such in Schizoneura gOt~dwanensis (Srivastava, forking of leaves is known only in the genus 1955) or in Phyllotheca angusta (Surange & Barakaria, and there is a report of a solitary Kulkarni, 1968). forked leaf in Phyllotheca australis (Pant & O£ the two species of the genus Schizo• Kidwai, 1968, pI. 30, fig. 6£). Further in neura, S. wardii Zeiller is restricted to the Asterocalamites the pelt ate sporangiophores Lower Permian formations only. The other are not subtencl.ed by bracts whereas species, S. gondwanensis Feistmantel ranges in P. australis the strobili have alternating from the Lower Permian to the Upper whorls of bracts and branched sporangio• Triassic but is more common in the Upper phores. The strobili of the Calamitaceae Permian and Lower Triassic. Phyllotheca have sporangiophores subtencl.ecl.by bracts makes it