藏南舌羊齿植物群的再研究

采自西藏南部定日县和定结县的舌羊齿植物群已由徐仁进行过研究(1973,1974,1976)。考虑当时文献资料的欠缺和近年来南半球舌羊齿类植物研究的新进展,作者重新审查了全部化石,对某些属种作了修正,并相应地对含该植物化石的地层时代也提出了新认识。

REVISION OF GLOSSOPTERIS FLORA FROM SOUTHERN XIZANG

Hsu reported and described (1973, 1974, 1976)a collection of typical Permian Lower Gond-wana plants from localities in Southern Xizang, Dingri and Dingjie districts. Since then there have been a number of advances on the study of Lower Gondwana floras. This prompted us to re-examine the collection and revise some of Hsu’s original determinations. The Qubu flora in-cludes: Trizygia specoisa Royle (pl. 1, figs 2, 2a), Sphenophyllum minor (Sterzel) Gu et Zhi (pl. 1, figs 1, la), Glossopteris communis Feistmantel, G. dingriensis Rigby sp. nov. (pl. 1, figs 5-8), G. sp. cf. G. indica Schimper, G. spp., Austroonnualaria qubuensis (Hsu) Rigby, Paraca-lamites anstralis Rigby, Pecopteris sp. cf. P. unita Brongniart, Cladophlebis qubuensis (Hsu) Li, Asterotheca sp. (pl. 1, 4, 4a), scale leaf (pl. 1, 3, 3a), and an unidentifiable stem. Some taxa need further explanation:(1) The distinction between Trizygia and Sphenophyllum: Trizygia is for bilaterally symmetrical laef whorls, in which the leaflet form and size are different; Sphenophyllum is used for axially symmetrical leaf whorls, in which the leaf form and size are uniform.(2) Diagnosis of Glossopteris dingriensis Rigby sp. nov. Long leat, parallel sides for a major part of its length, tapering very gradually, almost imperceptibly towards the apex over a distance of about four times the maximum leaf width; apex acute; base also gradually tapering; venation moderately arched, parallel, of the communis type, one or two dichotomies and up to three anastomoses along each vein. The leaf shape and thinner veins distingush it from G. angustifolia and G. linearis. Holotype is No. 4738b (pl. 1, fig. 8), paratypes are No. 4776 and 4759 (pl. 1, figs 6, 7), housed at the Institute of Botany, Academia Sinica, Xiangshan, Beijing.(3) Austroannularia Rigby 1988 was founded on specimens from Australia, Annularia stellata (?) Walkom 1916, and from Xizang, Ramganjia qubuensis Hsu (1976, pl. I, figs 4, 5, text-fig. 1). Diagnosis of the genus: Annul aria-like to Lobatannularia-like leaf whorls, asymmetrical, with shorter leaves on one side of the whorls giving the whorl a reniform outline; longer leaves tending also to be broader, fused for half their length or less, shorter leaves fused to a lesser extent if at all; leaves linear-obovate, with an acute apex, possible mucronate, leaf tips or apical regions tending to droop or be lax, sheath probably absent. Type species is A. subcircu-laris Rigby (Australian specimens). The leaves of A. qubuensis (Hsu) Rigby are much narrower than those of A. subcircularis, being approximately ten times as long as broad, decidely linear instead of linear-obovate, and with an acute apex. The leaf tips are more erect with no apparent drooping, a feature that separates A. qubuensis from Lobatannularia as well as from A. subcircularis. The specimen 4738a figured by Hsu, 1976, pl. I, fig. 4, is selected as lectotype.(4) Paracalamites australis Rigby is a form genus and species used for sphenopsid stems devoid of foliage, from the Lower Gondwanas. We redetermined stem specimens figured by Hsu as Raniganjia qubuensis (1976, pl. I, fig. 6) as Paracalamites australis.(5) Another stem figured by Hsii as Raniganjia qubuensis (1976, pl. 1, fig. 7) is different from Paracalamites australis as it lacks longitudinal ribs; it may be a fern stem.(6) Specimens of Dizeugotheca qubuensis Hsii (1976, pl. II, figs 8, 9, 11, 12, textfig. 2) comprise only sterile pinnae. We revise these as Pecopteris sp. cf. P. unita Brongniart for the similar pinna shape and venation.(7) Scale leaves are common in the Gondwanas often connected with the reproductive organs of Glossopteris,Geological age of the Qubu flora.The Qubu flora is a Gondwana flora not a mixed flora (Li et al., in press). Most species are endemic so cannot indicate age more precisely than Permian based on the distribution of the particular genera elsewhere. Glossopteris communis is the most reliable indicator in the flora based on the species present. According to Chandra and Surange (1979) G. communis is restricted to the Karharba.