Papers and Proceedings of the Royal Society of Tasmania, Volume 121, 1987 159

MID-MESOZOIC LEAVES FROM NEAR IDA BAY, SOUTHERN TASMANIA, AUSTRALIA

by William D. Tidwell, Jong-Heon Kim and Tatsuaki Kimura

(with three text-figures and three plates)

TIDWELL, W.D., KIM, 1.H. & KIMURA, T., 1987 (30:vi): Mid-Mesozoic leaves from near Ida Bay, southern Tasmania, Australia. Pap. Proc. R. Soc. Tasm., 121: 159-170,3 pI. ISSN 0080-4703. Department of Botany and Range Science, Brigham Young University, Provo, Utah 84602, U. S. A.; Department of Astronomy and Earth Sciences, Tokyo Gakugei University, Koganei, Tokyo, 184 Japan (l.R. Kim is a Korean student studying with Professor Kimura). Several leaf specimens from a locality in southern Tasmania are described. They are assigned to Coniopteris webster;; sp.nov., Cladophlebis indica (Oldham & Morris) Sahni & Rao, Pachypteris sp. cf. indica (Oldham & Morris) Bose & Roy, Otozamites sp., Pterophyllum? sp. and Conites sp. The specimen of Olozamiles sp. is also partially petrified and, thus, gives some indication of its internal tissues. These genera and species suggest a mid-Mesozoic, rather than a Tertiary, age for this locality. Key Words: Mid-Mesozoic, Tasmania, Coniopteris, Cladophlebis, Pachypteris, Otozamites, Ptilophyllum, Ptero­ phyllum, Conites, Osmundacaulis.

INTRODUCTION SYSTEMATIC DESCRIPTION A locality south of Ida Bay in southern Filicales Tasmania has yielded numerous specimens offossil Dicksoniaceae plants (pI. I, fig. 1). This locality lies a few hundred metres to the east and west of the Lune River Road Genus Coniopteris Brongniart, 1849 immediately north of its junction with the Leprena Coniopteris websterii sp.nov. Track (text-fig. 1). The plant remains include two PI. II, figs 1,2; pI. III; text-figs 2B, 3C(i & ii) osmundaceous species, Osmundacaulis neTii Tid­ Material well and 10nes (1987) and O. jonesii Tidwell (1987), a fossil tree fern Cibotium tasmanense Gould Holotype - Z2296, Paratype - Z229S. (1972), unidentified permineralized conifer wood Diagnosis and the fossilized leaf material of this report. The Leaves mostly fertile; main rachis, 1.5 mm permineralized plant materials and the leaf im­ thick bearing pinnae at 6 mm intervals; main rachis pressions were collected together but were not in and pinna rachis rounded; pinnae, thick, leathery, situ. The rock matrix having the leaf impressions about 15 mm long, arising alternately from main also contains petrified fern petioles assignable to rachis; first pinnules of pinna arising opposite, later Cibotium tasmanense Gould and an, as yet, pinnules alternate on the two sides; pinnules of one unnamed tree fern. This suggests that the fossil pinna "commonly" overlapping those of the next plants from this locality came from the same and adjacent pinnules of the same pinna; first horizon. Gould (1972) considered the strata from pinnule consists of short stalk, forking after whence these plant fossils were derived as possibly approximately 0.2 mm into an upward pointing Tertiary. fertile branch and forward and outward pointing Although the fossil leaves described in this sterile branches forked into two to four, lobes report are preserved in three dimension, they lack projecting over main rachis; upper fertile appen­ cuticle. In some, however, the outline of the dages alternate, consist of long stalk, 1 mm long epidermal cells is impressed upon the specimens. and terminal sorus; sorus and indusium circular­ These leaf specimens were collected by N eri Webster cordate in lateral compression, about 2 mm wide of Bicheno, Tasmania and Ross Jones of Blackmans by 1-1.6 mm high; basal sori generally smaller and Bay, Tasmania. The specimens are deposited in the hemispherical; sori lying in flat plane or slightly Tasmanian Museum and Art Gallery. tilted; indusium thick, 200Mm wide, marked 160 W.D. Tidwell, J-H. Kim and T. Kimura

resembles Gonatosorus sphenopteroides Brick (1937) and G. nathorsti Raciborski(l894). Shorter sterile lobes and a much smaller sorus of G. sphenopteroides separates it from C. websterii and broader, undivided sterile lobes and a smaller sorus separates G. nathorsti from C. websterii. Halle (1913) described four species of Coniopteris from the Jurassic or Early Cretaceous (Stipanicic & Bonetti 1970) of Graham Land, and Douglas (1973) described fertile leaves or sterile parts of four species of this genus from the Early Cretaceous of Victoria, Australia, but none is similar to C. websterii. Based on sterile foliage, Coniopteris nanopinnata Douglas has been trans­ ferred to Gleichenites (Drinnan & Chambers 1986). Incertae Sedis Genus Cladophlebis Brongniart, 1849 Cladophlebis indica (Oldham & Morris) Sahni & Rao 1933 PI. I, figs 4, 7; pI. II, figs 5, 6; text-fig. 2A Material FIG. 1 - Map of southern Tasmania showing Z2297, Z2298. collecting sites for fossil plants. Road names in Diagnosis parenthesis are those used on Gould's (1972) index map. 0 = collection locality for Otozamites sp. Five sterile pinna fragments with rather stout (Modified from Leprena 4818, edition 1, 1985; pinna axis up to 1 mm wide; pinnules falcate­ Tasmania 1:25 000 Landsmap.) deltoid in shape, tapering strongly towards apex, typically 11 mm long by 5 mm wide at the base; externally by differential preservation; margins apices obtusely or subacutely pointed; close, entire; sporangia numerous, not projecting beyond suboppositely or alternately attached by entire the edge of the indusium, rather small, about base at an angle of 50°; margins entire, slightly 200Mm wide, annulus oblique, nearly complete; no recurved; laminae contiguous near base; midrib is distinct, extending to tip; lateral veins katadromic, spores were observed. 8-9 pairs, forking once; basiscopic basal vein Etymology occasionally forked twice. The specific epithet honours Neri Webster Comparisons who collected the specimens of this fern and who The leaves of only a few reported Cladophle­ has contributed so much to this study. bis species resemble Cladophlebis indica in having Remarks entire, falcate pinnules and singly-forked lateral The holotype is fragmental and consists of a veins (Retallack 1980). Among these species similar single fertile pinna with seven fertile appendages. to C. indica are C. nebbensis (Brgn.) Nathorst Sterile leaves are similar to the sterile branches of (1878), C. acuta Fontaine (1889), C. grahamii the first fertile pinnules. Frenguelli (1947), and C. oblonga Halle (1913). The latter two species were reported from Jurassic Coniopteris websterU is similar to C. or Early Cretaceous strata of Graham Land. margaretae Harris (1961) and C. simplex(L. & H.) Cladophlebis nebbensis, although poorly preserved, Harris (1961). It is distinguished in having larger has pinnules that appear less falcate, less tapered sori than C. simplex and having alternate rather and more obtusely pointed, whereas, the pinnule than opposite pinnules. The upward pointing fertile apices in C. acuta are more acute than in C. indica. branch of the first pinnules in C. websterU differs Cladophlebis grahamii differs from C. indica by from the forward pointing fertile branch of these the former species having its basal lateral vein pinnules in C. margaretae. Coniopteris websterii is occasionally forking twice and the basal acroscopic unlike other species of Coniopteris. It, however, margins of its -Pinnules being subauriculate. Mid-Mesozoic leaves from southern Tasmania, Australia 161

A(i)

D(iv) • FIG. 2A(i-v) - Cladophlebis indica (Oldham & Morris) Sahni & Rao showing various forms ofpinnules (i); from Z2298. (ii-v)from Z2297; x 2; 2B - Coniopteris websterii sp. nov. from Z2296; x 2; C(i-v)­ Pachypteris sp. cf indica (Oldham & Morris) Bose & Roy showing venation (i,vi,v),from Z2301; (ii,iii) from Z2300; x 2; 2 D(i-iv) - Otozamites sp., showing pinna-form and venation from Z2302; x 2. 162 W.D. Tidwell, J-H. Kim and T. Kimura

Cladophlebis oblonga is closest to C. indica and is Diagnosis distinguished from this latter species by its less­ Leaf at least bipinnate with a rather thick tapering pinnule apices (Retallack 1980). Based rachis, I mm wide with a median groove, ultimate upon some fertile specimens, Bose & Sah (1967) pinnae arise alternately at an angle of about 50°; transferred C. indica to Todites indicus. Retallack pinnae elongate-oblong in outline, 15 mm long and (1980), however, considered them separate taxa up to 8 mm wide, 5-8 pairs of pinnules arising and, noted that when fertile fronds of C. indica are katadromically together with a terminal one. collected, they will probably be osmundaceous. Pinnules generally short-oblanceolate or ovate in The present pinna fragments are similar to outline; apices obtusely pointed or blunt, pinnules those of Cladophlebis australis (Morris) Seward attached to pinna axis at a narrow angle, decurrent, (1904) and C. bi/ormis Drinnan and Chambers. forming a narrow web, 0.5 mm wide along both Cladophlebis australis is known extensively from sides of pinna axis; typically 5 mm long and up to Triassic strata in the Southern Hemisphere (see 105 mm wide; veins Sphenopteris-type; three pairs Herbst 1978, Retallack 1980, 1983). In fact, the of simple or once forked lateral veins arise from species name australis was given originally by midrib; basal pinnules give rise to a pair of deeply Morris (1845) to Tasmanian Triassic leaves. dissected lobes; these pinnules attached at a wide Cladophlebis australis, in a strict sense, has angle or perpendicular; veins invisible; a single relatively erect, falcate pinnules with entire or shallow lobe present only on the acroscopic side in slightly undulating margins (Retallack \983). some pinnules (text-fig. 2C(ii»; no pinnule at very Cladophlebis bi/ormis was reported from the Lower base of pinna or on rachis between adjacent pinnae; Cretaceous Koonwarra Fossil Bed on the South cuticle not preserved. Gippsland Highway south of Leongatha, Victoria (Drinnan & Chambers 1986). In C. bi/ormis, the Remarks entire margined, falcate pinnules have acute apices Specimens consist of five fragments of iso­ and are distichously arranged. Both of these species lated leaves. Although detailed cuticular features have lateral doubly-forking veins throughout their of these leaves are unknown, they appear to be pinnules which contrasts with the single-forking similar to the seed fern Pachypteris. Epidermal veins of C. indica. Cladophlebis australis, as loosely cells of the upper epidermis were observed in the identified in the past, has been incorrectly regarded pinnule impressions of these specimens (text­ as the most widespread, long-ranging and common fig.3B). These cells appear to have had straight fern of the Gondwanan Mesozoic (Retallack 1983). walls and are almost isodiometric in shape. How­ However, as pointed out by Retallack (1983), if ever, they are slightly elongated over the veins. No Morris' original illustrations are followed then C. stomata were observed. Because of their incom­ australis would be restricted to Middle and Late pleteness, it is difficult to make a definite identifi­ Triassic rocks of eastern Australia and Argentina. cation to species of these leaves. However, they are Cladophlebis indica is a long-ranging taxon sufficiently similar to Pachypteris indica to tenta­ that has been found in both New Zealand and India tively assign them as Pachypteris sp. cf. indica. (Retallack \980). Pachypteris indica has been reported from the Some species of Cladophlebis have been Middle to Upper Jurassic Kachchh localities (Bose regarded as members of the Osmundaceae (Town­ & Banerji 1984) and the Lower Cretaceous JabaJpur row 1963, Harris 1961, Andrews 1970). Cladoph­ Series (Bose & Roy 196') of India. lebis indica may also belong to this family. Petrified fern stems assignable to Osmundacaulis fossilized Comparisons with this foliage (Tidwell 1987, Tidwell & Jones Pachypteris specifica Feistmantel known from 1987), suggest such assignment may be correct the Lower Cretaceous of India (Bose & Banerji (MedwellI954, Douglas 1969). 1984) ana P. lanceolata Brongniart reported from the Middle Juras~ic of Yorkshire (Harris 1964) are Pteridospermales similar to the Tasmanian Pachypteris specimens. Genus Pachypteris Brongniart, 1828 However, they are distinguished from each other Pachypteris sp. cf. indica (Oldham & Morris) Bose by their size, the presence of pinnules on the main & Roy 1967 rachis and their respective venation. PI. II, fig. 4; text-figs 2C(i-v) Pachypteris sp. cf. indica is also similar to P. crassa Townrow, and P. austropapillosa Douglas. Material Pachypteris crass a was collected from Upper Z2299, Z2300, Z2301. Triassic sediments at the Langloh coal mine, Mid-Mesozoic leaves from southern Tasmania, Australia 163

C(n

FIG. 3A(i-ii) - Otozamites sp. (i) cross-section of petiole and lamina. Black dots represent vascular bundles and veins, 22302; x 20. (it) epidermal cells of the upper lamina and a vein (arrow); x 10; 3 B - Pachypteris sp. cf indica (Oldham & Morris) Bose & Roy. Epidermal cells of the upper lamina, 22301; x 100; 3 C(i-ii) - Coniopteris websterii sp. nov. (i) enlargement of sorus containing sporangia; x 10. (ii) showing distribution of sori, 22296; x 2.5.

Hamilton, Tasmania (Townrow 1965). Pachypteris Bennettitales crassa differs from Pachypteris sp. cf. indica by Genus Otozamites Braun, 1842 being larger, having slightly contracted pinnule Otozamites sp. bases and having pinnules that are joined together PI. II, figs 2, 3; text-figs 2D(i-iv), 3A(i-ii) by a wing of tissue along the pinna rachis. This wing gives a bilateral symmetry to the rachis of P. Material crassa that does not appear in P. sp. cf. indica. Z2302 Pachypteris austropapillosa, described from the Lower Cretaceous of Victoria (Douglas 1969), Diagnosis differs from P. sp. cf. indica by being larger, having Leaves nearly parallel-sided for the most undulate pinnule margins, rounded apices and part, elongated, up to 15 mm wide; apex and base more perpendicular aspects in its pinnule attach­ unknown; pinnae falcate-deltoid with rather ment. expanded acroscopic basal angle; acroscopic margin Sterile and fertile pinnules in Coniopteris concave, basiscopic margin mostly convex, con­ frutiformis Douglas (1973) are attached to each tracted and slightly decurrent near base; typically other. The sterile pinnules of this species are similar 7.5 mm long, 2.5 mm wide at middle; pinna apex to P. sp. cf. indica and its fertile pinnules are like C. obtusely or subacutely pointed; pinnae close, over­ websterii. However, the sterile pinnules of C. lapping, sub oppositely attached to upper edge of frutiformis lack a midrib which is present in P. sp. slender, round rachis (up to 2.5 mm wide) by lower cf. indica and the sterile pinnules of C. websterii are portion of inner edge of pinna base at a right angle; dissected, unlike those of C.frutiformis. The fertile upper half of base free; expanded acroscopic basal pinnules of C. websterii are continuous to the apex angle often overlapping succeeding basiscopic pinna of the pinna (pI. IlI) whereas upper portions of the base; veins originated from the attachment of fertile pinna of C. frutiformis are sterile. pinna, five in number at base, upper three fork 164 W. D. Tidwell, J-H. Kim and T. Kimura

twice and radiate to the acroscopic angle and appear to be arched. No hypodermal layer was margin; lower two fork twice, each ending at the detected. distal basiscopic margin; cuticle not preserved. The mesophyll, although not well preserved, appears to have been well developed and consisted Remarks of one, possibly two, layers of closely packed This specimen of Otozamites sp. is charac­ cylindrical cells 12/-1m wide and 50/-lm long. Their terized by its small size and elongated leaf composed walls appear thin. Throughout the remainder ofthe offalcate-deltoid pinnae with strongly marked and leaf, the mesophylJ cells are round to irregularly free acroscopic angle. shaped. The fossil is preserved in an unusual manner. Outlines of polygonal epidermal cells are It is intermediate between a cast and a petrification. preserved in impressions of the leaves (text-fig. The specimen has the external impressions of the 3A(ii». The cells over the rachis are rectangular to Otozamites leaves, but the bulk of the foliage is still occasionally rectangular spindle-shaped (30-40}1m embedded in is matrix which was sectioned. long by 15}1m wide) and round hair bases appear to In cross-section, Otozamites sp. consists of be present. Stomates were not observed. round rachides with attached "gull-wing" shaped Impression of the upper epidermis shows the pinnae (text-fig. 3A(i». Although poorly preserved, epidermal cells between the veins are rectangular to various components of the rachides can be observed. polygonal and are parallel to subparallel to the The epidermal and hypodermal tissues of the veins (pI. I, fig. 6). The cells over the veins are rachides cannot be separated and the ground tissue elongated in longitudinal rows and are of a more inside these structures appears to be parenchy­ uniform rectangular shape. Cell walls are moder­ matous. However, the outline of these cells is not ately marked, rather broad and straight. No evi­ often seen and most of it is disorganized into a dence oftrichome bases is present. Lower epidermis brown humus-like substance. Complete layers of was not observed. thick-walled mechanical tissue similar to scleren­ chyma surround the vascular bundles. Up to forty Comparisons vascular bundles occur in two rather vague circular Specimens from the Boola Boola forest bands. The inner circle is situated within an outer (Lower Cretaceous) that were first described as circle which is near the epidermal-hypodermal Otozamites by Douglas (1962, 1963) and later region. The other bundles are randomly distributed. transferred to Ptilophyllum (Douglas 1969), No resin cavities appear in the rachides or in the resemble this Tasmanian specimen in pinna size, mesophyll of the pinna. but they are distinguished from each other by the Bundle configuration is not known in other strongly developed acroscopic angle of the pinna in species of Otozamites, but it is known in some the Otozamites specimen from Tasmania. species of the Bennettitales such as Ptilophyllum This Tasmanian Otozamites species resembles amarjolense Bose, P. sahnii Gupta and Sharma most closely in leaf form and size the species O. and Dictyozamites sp. (Bose & Kazat 1970a, bengalensis (Oldham & Morris 1863) Seward (1917) 1970b). In these, the bundles are arranged in which is, according to Bose & Kasat (l970a), double U's which are unlike the more or less synonymous with Ptilophyllum cutchense Morris. scattered bundle configuration in the rachides of Otozamites sp. can be distinguished from the above Otozamites sp. mentioned Indian species by its very different Vascular bundles can be observed in the pinna form and by having venation which radiates lamina as black dots but their anatomy is difficult from the base of the pinnae rather than more or less to determine. The upper and lower cuticles appear parallel veins arising from the entire pinna base as to be thin and about the same thickness. It is in Ptilophyllum. difficult to distinguish the cuticle from the epider­ The small leaves of Ptilophyllum bengalense mis. No attached trichomes were noted. Stomata (Oldham) Schimper from the Liassic plant-bed of were not recognized on the upper surface of the North Iran (Kilpper 1968) and Otozamites walka­ lamina, which seems to be covered by an unbroken motaensis Bose & Zeba-Bano (1978) from Kachchh, epidermis. In some areas, stomata can be seen in India resemble in pinna form those of Otozamites the lower epidermis, lying in the lamina between sp. Ptilophyllum bengalense and O. walkamo­ bundles. Although not perfectly preserved, the taensis are distinguished from O. sp. by their guard cells seem to be at an oblique angle usual in pinnae not being distinctly auriculated, by the gymnosperms (Stopes 1910). Surfaces of the guard venation in P. bengalense being coarser, simple or cells are thinly cutinized. The apertures do not once forked and the higher number of veins (8-14) Mid-Mesozoic leaves/rom southern Tasmania, Australia 165 m O. walkamotaensis which fork one to three they most closely approximate the scales of Conites times. sp. B from India (Bose & Banerji 1984). Ptilophyllum spinosum described by Douglas (1969) from the Lower Cretaceous of Victoria, DISCUSSION Australia is similar in leaf form, size and pinna form to Otozamites sp., but is distinguished from The leaf specimens constituting this report, this latter taxon by its densely crowded veins (ten in various osmundaceous species and two genera of number at the base) that fork only once. The tree ferns were collected on the slope and flat areas Tasmanian leaves were placed in Otozamites instead (PI. I, fig. 1) east and west of the Lune River Road, of Ptilophyllum because of the presence of strongly north of the junction with the Leprena Track. A marked and free acroscopic basal angle in the basalt, containing siliceous amygdules and xenoliths pinna of the Tasmanian specimens. Based upon of silicified sediment, crops out on the ridge above this feature, it is likely that Ptilophyllum spinosum and west of the locality and extends southwards. Douglas is closer to the genus Otozamites than to Gould (1972) thought the fossil material may have Ptilophyllum. It is difficult to distinguish Otoza­ been derived from weathering of the pale, leached mites from Ptilophyllum by their cuticular charac­ sediments interbedded with or just beneath the ters alone. basalt, or from the xenoliths in the basalt itself. There is also a possibility that these Tas­ However, the thick vegetational cover obscures the manian leaves might belong to Ctenozamites, but actual original site(s) of the fossil plant remains. As the slight radial divergence of the leaves of O. sp. pointed out by Gould (1972), there is not any direct makes it unlikely that they belong to the former evidence for the age of these sediments. However, genus. based upon a suggested Tertiary age for the basalt, he thought them to be Tertiary. Upper Triassic coal measures containing silicified wood occur north of Genus Pterophyllum Brongniart, 1828 this fossil locality. Gould (1972) considered the Pterophyllum? sp. preservation of this Triassic wood to be different PI. I, figs. 3, 5 from that of the petrified fern he described from the Material Lune River Road-Leprena Track site (fig. I). The Z2303 fossils from this latter locality appear to be too young to be Triassic and are certainly older than Remarks Tertiary. The only specimen in this flora that can be Based upon the occurrence of the pterido­ assigned to this genus is a single pinna fragment sperm Pachypteris and bennettitalean Otozamites, that unfortunately is only represented by its basal there is little doubt that the geological age of the part. The pinna is 9 mm wide at its base and is collecting site for Cibotium tasmanense Gould, attached to the side of a slender tachis. It is slightly Osmundacaulis nerii, O. jonesii and the fossil leaf expanded at its base. The parallel veins arise from materials comprising this report is Mesozoic. It is the rachis at an angle and are simple and nine in difficult to make a precise age determination of the number. The basiscopic veins are bent slightly plant bed within the Mesozoic because of the downwards at their point of origin. Cuticle was not insufficiency of the fossil material. preserved. Because of its incompleteness, this In comparing this Tasmanian flora with the fragment is only provisionally assigned to Ptero­ flora from the Middle Jurassic Walloon Coal phyllum. Measures of Queensland, none of the fossil plant species reported by Gould (1974, 1981) from this strata are the same as those from the Tasmanian Gymnosperm -~ Incertae Sedis locality. Anatomically, Osmundacaulis hoskingii Genus Conites Sternberg, 1823 Gould (1973), is the closest, but it differs in various Conites sp. ways from the Tasmanian species (Tidwell 1987, PI. I, fig. 2 Tidwell & Jones 1987). The other species of petrified osmundaceous stems from the Walloon Remarks Coal Measures are also different from the Tas­ Very small cone scales resembling Conites manian axes. These latter species have been trans­ occur in the matrix of the specimen containing ferred to Millerocaulis (Tidwell 1986). Otozamites sp. (Z2302). None of these scales are It is most likely that the strata of this Tas­ attached to a cone axis. Based upon size and shape, manian plant bed is Late J urassic-Early Cretaceous 166 W.D. Tidwell, J-H. Kim and T. Kimura

in age. As presently known, Pachypteris species or REFERENCES pteridospermous plants with small pinnae and pinnules as in the Tasmanian specimens are pre­ ANDERSON, J.M. & ANDERSON, H.M., 1985: PALAEOFLORA OF SOUTHERN AFRICA. dominant in the Upper Jurassic and Lower Creta­ PRODROMUS OF SOUTH AFRICAN ceous plant beds in India (Bose & Banerji 1984), MEGAFLORAS DEVONIAN TO LOWER Argentina (Archangelsky 1963a, 1963b, 1964) and CRETACEOUS. A.A. Balkema, Rotterdam: South Africa (Anderson & Anderson 1985; their 423 pp. gen. A. sp. A and gen. B. sp. A), and smaller ANDREWS, H.N., 1970: Famille des Osmundaceae R. Otozamites (or Ptilophyllum) species such as these Brown. In Boureau, E. (Ed.): TRAITE DE from Tasmania are predominant in the Lower PALEOBOTANIQUE, Tome IV, fase, I: 250. Cretaceous plant beds in Australia (Douglas 1969). ARCHANGELSKY, S., 1963a: Notas sobre la flora fosil Because Cladophlebis indica has been recorded de la zona de Tico, Provincia de Santa Cruz. 2 Tres neuvas especies de Mososingeria. Ameghin­ from the Upper Triassic to Lower Cretaceous plant iana, 3(4): 113-120. beds (Retallack 1980), the sterile fern leaves of this ARCHANGELSKY, S., 1963b: A new Mesozoic flora species are useless for precise age determination. from Tico, Santa Cruz Province, Argentina. The Osmundacaulis species (0. nerii and O. Bull. Brit. Mus. (Nat. Hisl.} Geol., London, 8(2): jonesii) from this locality are like O. hoskingii, but 45-92. are most similar to O. natalensis (Schelpe 1955) ARCHANGELSKY, S., 1964: Notas sobre la flora fosil Miller and O. atherstonei (ScheJpe 1956) Miller de la zona de Tieo, Provincia de Santa Cruz. 3. from the Lower Cretaceous of South Africa. Ruflorina pilifera n.sp. 4. Equisetites sp. Erasmus (1978) identified two new species of Ameghiniana, 3(8): 221-226. BOSE, M.N. & BANERJI, J., 1984: The fossil floras of Osmundacaulis from the Lower Cretaceous of Kachchh. I. Mesozoic megafossils. Palaeo­ South Africa. The anatomy of these two species is botanist, 33: 1-189. exceptionally close to O. nerii, O.jonesii and other, BOSE, M.N. & KASAT, M.L., 1970a: The genus as yet undescribed, osmundaceous species from Ptilophyllum in India. Palaeobotanist (1972), this Tasmanian locality. The inner and outer 19(2): 115-145. sclerotic rings which are found surrounding the­ BOSE, M.N. & KASAT, M.L., 1970b: On petrified petiolar vascular strands in O. nerii also occur in specimen of Dictyozamites from the Rajmahal one of Erasmus' species. These two species, as well Hills, India. Palaeobotanist (1972), 19(3): as another Osmundacaulis species (0. lemonii 248-250. BOSE, M.N. & ROY, S.K., 1967: On the occurrence of Tidwell).from the Late Jurassic of Utah, are the Pachypteris in the Jabulpur Series of India. only taxa with these unique tissues. Palaeobotanist (1968),16(1): 1-9. Based upon the fossil evidence, this flora BOSE, M.N. & SHAH, S.C.D., 1967: Some pteridophy­ appears to be either Late Jurassic or Early Creta­ tic remains from the Rajmahal Hills, Bihar, ceous and, most likely, Early Cretaceous in age. Palaeobotanist (1968), 16(1): 12-28. Because of the lack of angiospermous fossils, this BOSE, M.N. & ZEBA-BANO, 1978: On the new species flora is probably from older Lower Cretaceous of Otozamites from Kachchh, western India. strata. Palaeobotanist (1981), 27: 227-231. BRICK, M.I., 1937: La flore Mesozoique du Ferghana Meridoional, II. Fougeres (fin) Equisetinees. ACKNOWLEDGEMENTS Trans. geol. Trust Centro Asia, Tashkent, 3: 1-75. The authors wish to express appreciation to DOUGLAS, J,G., 1962: The occurrence of Otozamites Neri Webster and Ross Jones for their donation of in south-eastern Victoria, Proc. R. Soc. Vict., specimens which made this study posible, to Dr 75(1): 4143. M.R. Banks of the University of Tasmania for his DOUGLAS, J.G., 1963: Newbennettitalean leaves from helpful assistance and patience during this study, the Mesozoic of eastern Australia. Proc. R. Soc. and to Dr S.R. Ash of Weber State College and Dr Vict., 77(1): 197-206. G. Retallack of the University of Oregon for DOUGLAS, J.G., 1969: The Mesozoic floras of Victoria. Pts 1-2. Geol. Surv. Viet. Mem. 28: 310 pp. reviewing the manuscript. DOUGLAS, J.G., 1973: The Mesozoic floras of Victoria. Pt 3. Geol. Surv. Vict. Mem. 29: 185 pp. DRINNAN, A.N. & CHAMBERS, T.e., 1986: Flora of the Lower Cretaceous Koonwarra Fossil Bed (Korumburra Gtoup), South Gippsland, Vic­ toria. Mem. Assoc. Australas. Palaeontols, 3: 1-77. Mid-Mesozoic leaves/rom southern Tasmania, Australia 167

ERASMUS, T., 1978: THEANATOMY AND EVOLU­ RAClBORSKI, M., 1894: Flora Kopalna ogniotrwalych TION OF OSMUNDACAULlS MILLER glinck Krakowskich. Czesc. 1. Rodniowce EMEND. WITH NOTES ON THE GEO­ (Archaeginiatae). Pam. Mat.-przyr. Akad. Urn., METRY OF THE XYLEM FRAMEWORK Krakow, 18: 143-243. OF THE OSMUNDA CEO US STELE. Unpub\. RET ALLACK, G.J., 1980: Middle Triassic megafossil Ph.D. thesis, Univ. Pretoria: 155 pp. plants and trace fossils from Tank Gully, FONT AINE, W.M., 1889: The Potomac or younger Canterbury, New Zealand. J. R. Soc. New Mesozoic flora. u.s. Geol. Surv. Mon. 15: Zealand, 10(1): 31-63. 1-377. RETALLACK, G.J., 1983: Middle Triassic megafossil FRENGUELLl, J., 1947: E1 genero Cladophlebis y sus marine algae and land plants from near Benmore representante en la Argentina. An. Mus. La Dam, southern Canterbury, New Zealand. J. R. Plata n.s. Paleont., sec. B. Paleobot., 2: 1-'74. Soc. New Zealand, 13: 129-154. GOOLD, R.E., 1972: Cibotium tasmanense sp. nov., a SAHNI, B. & RAO, AR., 1933: On some Jurassic plants fossil tree-fern from the Tertiary of Tasmania. from the Rajmahal Hills. J. Asiat. Soc. Beng. Aust. J. Bot., 20: 119-126. NS., 27(2): 183-208. GOOLD, R.E., 1973: A new species of Osmundacaulis SCHELPE, E.A.C.L.E., 1955: Osmundites natalensis­ from the Jurassic of Queensland. Proc. Linn. a new fossil fern from the Cretaceous of Zululand. Soc. NS. w., 98(2): 86-94. Ann. Mag. Nat. Hist., 8(12): 654-656. GOULD, R.E., 1974: The fossil flora of the Walloon SCHELPE E.A.C.L.E., 1956: Osmundites atherstonei Coal Measures: a survey. Proc. R. Soc. Queens­ - new Cretaceous fern from Cape Province, land, 85(3): 33-41. South Africa. Ann. Mag. Nat. Hist., 9: 330-332. GOULD, R.E., 1981: The coal-forming flora of the SEW ARD, AG., 1904: On a collection of Jurassic Walloon Coal Measures. Coal Geology, 1(3): plants from Victoria. Rec. geol. Surv. Vict., 1: 83-105. 155-210. HALLE, T.G., 1913: The Mesozoic flora of Graham SEWARD, A.G., 1917: FOSSIL PLANTS. III. Cam­ Land. Wiss. Ergb. Schwed. Sub-polar Exp. bridge 0 niversity Press, Cambridge: 656 pp. 1901-1903,3(14): 1-123. STIPANICIC, P.N. & BONETTI, M.I.R., 1970: Posi­ HARRIS, T.M., 1961: THE YORKSHIRE JURASSIC ciones estratigraficas y edads de las principales FLORA. I. THALLOPHYTA-PTERIDO­ floras jurasicas argentinas; I, Floras liasicas. PHYTA. British Museum (Natural History), Ameghiniana, 7(1): 57-78. London: 212 pp. STOPES, M.C., 1910: The internal anatomy of HARRIS, T.M., 1964: THE YORKSHIRE JURASSIC "Nilssonia orientalis". Ann. Bot., 24: 389-393. FLORA. II. CA YTONIALES, CYCADALES TIDWELL, W.D., 1986: Millerocaulis, a new genus with & PTERIDOSPERMS. British Museum species formerly in Osmundacaulis Miller (Natural History), London: 191 pp. (Fossils: Osmundaceae). Sida, 11(4): 401-405. HERBST, R., 1978: Revision de las especies australasicas TIDWELL, W.D., 1987: A new species of Osmunda­ de Cladophlebis (Bgt.). (I). Facena, Corrientes, caulis (0. jonesii sp. nov.) from Tasmania, 2: 3-28. Australia. Rev. Palaeobot. Palynol., in press. KILPPER, K., 1968: Einige Bennettiteen-Blatter aus TIDWELL, W.D. & JONES, R., 1987: Osmundacau/is dem Lias von Karmozd-Zirab (N-Iran). J. Linn. nerii, a new osmundaceous species from Tas­ Soc. (Bot.), 61 (384): 129-135. mania, Australia. Palaeontographica B, 204: MEDWELL, L.M., 1954: A review and revision of the 181-191. flora of the Victorian Lower Jurassic. Proc. R. TOWNROW, J.A, 1963: A speculation of the Rhaeto­ Soc. Vict., 65: 63-111. Liassic climate of Tasmania. Pap. Proc. R. Soc. MORRIS, J., 1845: In PHYSICAL DESCRIPTION Tasm., 98: 113-118. OF NS. W. AND VAN DIEMEN'S LAND. TOWNROW, J.A, 1965: A new member of the Cory­ NAT HORST, A.G., 1876(1878?): BEITRAGE ZUR stospermaceae Thomas. Ann. Bot. NS., 29: FOSSILEN FLORA SCHWEDENS. UBER 495-511. EINIGE RHATISCHE PFLANZEN VON PALSJO IN SCHONEN. Stuttgart: 34 pp. (accepted 3 March 1987) OLDHAM, T. & MORRIS, J., 1863: Fossil flora of the Rajmahal Series in the Rajmahal Hills. Mem. geol. Surv. India Palaeont. indica, Ser.2, 1(1): 1-52. 168 W.D. Tidwell, J-H. Kim and T. Kimura

PLATE!

1. View of collecting locality. Collecting sites are 4. Pinnules of Cladophlebis indica (Oldham & in the flattened area in the foreground and on Morris) Sahni & Rao, Z2298; x 2.5. the slopes in the background (photo by Ross Jones). 5. Pterophyllum? sp. pinna base (arrow), Z23030; xl. 2. View of separate cone scales (arrow) of Conites , sp. in the matrix of specimen Z2302: x 10. 6. Cuticle of Otozamites sp., Z2302; x 10. 3. Portion of Pterophyllum? sp. showing pinna 7. Pinnules of secondary pinnae of Cladophlebis fragment (arrow), Z2302; x 1.8. indica (Oldham & Morris) Sahni & Rao, Z2297; x2. Mid-Mesozoic leaves from southern Tasmania, Australia 169

PLATE II

1-2 Coniopteris websterii sp. nov. 1, afertile pinna 4. Pachypteris sp. cf indica (Oldham & Morris) fragment, Z2296; x 2.2, enlargement of 1; x 4. Bose & Roy, a bipinnate leaffragment, Z2300. 3. Otozamites sp. A bundle of leaves consisting 5-6 Cladophlebis indica (Oldham & Morris) Sahni of more than 30 leaves preserved in parallel in & Rao. 5, a pinna fragment, Z2298; x 2.6, a the area of 50 mm x 30 mm, Z2302; x 2. bipinnate leaffragment, Z2297; x 2. 170 w.n. Tidwell, J-H. Kim and T. Kimura

PLATE III

1. Coniopteris websterii sp. nov. (Specimen in the collection of Neri Webster); x 4.