APPROACH TO THE CLASSIFICATION OF MESOZOIC " GINKGOALEAN" PLANTS FROM SIBERIA
V. A. KRASSILOV Far Eastern Institute of Geology, USSR Academy of Sciences, Vladivostok
ABSTRACT plant with the Czekanowskia leaves. The New findings from the Mesozoic of Bureja basin structure of Leptostrobtts is unique among confirm Harris' interpretation of Leptostrobus. Its female flowers of Gymnosperms. The plants capsule is analogous to the ovary of Angiospermae. with such flowers must be removed from the contact surface of valves is verrucose and papil• Ginkgoales. Probably they composed a late and corresponds to the stigma. The leaves of Czekanowskia. Phoenicopsis and allied genera are separate taxa. At the same time Harris assigned to the plants with Leptostrobus type of showed some difficulties in dividing the fructification and mu t be excluded from the Czekanowskia group from Ginkgos: the male Ginkgoales. Three types of female fructifications: reproductive structures are completely un• megastrobili of Ginl~go, Karkenia and Umaltolepis Aretobaiera (a new organ-genus referred to Pseudotorellia) are known and "the position of distinguished within the Ginkgoales, each of them and Sphenobaiera would be doubtful and representing a distinct family. this would be a real disadvantage" (HARRIS, 1951, p. 505). INTRODUCTION Our study of Ginkgoales and allied plants from the Upper J urassic- Lower Cretaceous of the Bureja River basin confirms the Mesozoic(1876)floradescribedas dominatedthe bySiberianGinkgo statements of Oishi, and Harris on the HEERfamily. In this family he included heterogenous nature of fossil "Ginkgoales" besides Ginkgo itself such extinct genera as sensu Heer and others and some new details Baiera, Triehopitys, Czekanowskia, Phoeni• come to light. eopsis, Dieranophyllum, Feildenia and Rhipi• dopsis. Heer's (1876) remarks on the affi• MATERIAL AND METHODS nity of Phoenicopsis are noticeable: he conside red this genus as some sort of a link The material was collected at the outcrops between the Ginkgoalean plants and Paleo• of Upper J urassic- Lower Cretaceous conti• zoic Cordaites. Seward and Gowan (1900) nental beds within the valley of Bureja were perhaps the first to doubt Heer's opi• River and its tributaries. Cleaned leaves, nion on the systematic posit.ion of Phoeni• brachyblasts, capsules, seeds and sporangia copsis-Czekanowskia group. Later Oishi were obtained by means of bulk maceration (1933) separated this group from Ginkgoales and transfer techniques. Some of them on the grounds of cuticular analysis: he were then subj ected to oxidation and mace• emphasized the striking difference in the ration with alkaline solution. epidermal structure of leaves of Phoenicopsis• Czekanowskia on one side and true Ginkgoa• DESCRIPTION les on the other. Oishi also rej ected a coniferous affinity of this group and supposed CZEKANOWSKIALES it to be the order of its own. These very exquisitive ideas were somehow oversha• The name" Czekanowskiales " apparently dowed by the well known works of Florin arrived for the first time in Pant's (1958) (1936a, b) on the Ginkgoales from the Franz classification of Gymnospermous plants. Jozef Land. Florin thoroughly investigated FC(lnale organs - Two species of Leptos• the favourably preserved plants from the trobus and one of Staphidiophora were inves• Phoenieopsis-Czekanowskia group and ex• tigated. The structure of Siberian Leptos• pressed no doubt in their Ginkgoalean trobus ex gr. laxiflorus Heer, is as a whole affinity. identical with those described by Harris Harris' (1951) study of Leptostrobus has from Greenland and Yorkshire: these fructi• thrown a new light on the problem. He fications consist of a long axis bearing fertile found a strong evidence of Leptostrobus capsules composed of two valves. Some being the female reproductive organ of a additional details were observed: 12 KRASSILOV - MESOZOIC" GINKGOALEN •. PLANTS FROM SIBERIA 13
1. A short stalk of the capsule ,,,hich is inflorescence. On this shortened axis only sometimes almost completely reduced• two fertile complexes (simple strobiles or capsules being sessile, and sometimes proli• carpels) had been retained which were united ferated in a comparatively solid structure in a two-valved capsule. The capsule is resembling a cylindrical short shoot, bears analogous to the ovary of Angiospermae somewhat ill defined marks or scars of and the contact surface of valves (carpc1s) is scale-leaves. Identical scars were found on specialized for.pollination (the present author the flattened basal part of the capsule. has no alternative interpretation of the 2. Convex central part of each valve is verrucose and papillate sculpture of this bordered by a flat marginal flange. The surface) and is analogous to the stigma. inner face of the flange is verrucose, with Pollination there ",as perhaps of gymnos• pointed or rounded outgrowths. The cuticle permous nature, but it is worth mentioning of inner side of the flange is densely papillate. that in ~ome tropical flowering plants (some The marginal flange gives inside very thin Anonaceae and others) pollen falls through lamellae overhanging the seeds. The cuticle the stigma and does not germinate on it. of lamellae is also papillate. M ale Organs - I consider the organ-genus 3. The inner concave side of the valve has 1xostrobus Raciborski as a male cone of very thin cuticle which differs in its struc• Czekanowskiales. In two localities in the ture from the outer cuticle. Seeds were Bureja basin these cones were found in not embedded in the "fibrous layer" as association with Leptostrotl1s. But in both Harris (1951) supposed them to be. cases 1xostroGus is a rare fossil while Leptos• Another type of LeptostroGus - L. ex troGus is abundant. The association with gr. crassipes Heer differs from the above the leaves attributed to Czekanowskiales described L. ex gr. laxiflorus in its large (see below) is more impressive. Still Raci• and flattened capsules with more dense borski and Heer emphasized the association and occasionally anastomo~ing venation. of I xostrobus (A nthohthcs sdmidtianus in Seeds are more numerous (about eight on Reer's monograph) with Czekanowsln'a or each valve). Inner cuticle of the valve Phoenicopsis. Prynada (1962) analizing the is more robust and clearly differentiated extensive Siberian material had come to on zones of very narrow cells under the the conclusion that I xostrocus belongs to veins and broader cells with slightly sinous Czehanowshia or Sphenobaicra. But male or undulating walls between them. The fructifications of the later are known and axis is pilose - the long unicellulJ.r tri• their structure is different (KR.:\usEL, 1943). chomes are rather frequent. One cone is I found two species of 1xostrobus -I. still attached to the cylindrical short shoot schmidtianus (Heer) Krassilov and I. heeri covered with scale leaves. Prynada. The first occurs in association The fructification closely allied to Sta• with Stephencphylhm in three locdities phidiophora secunda Harris was found in and the second - with both Czekancwskt'a the upper layer of coal-bearing rocks of and Phoenicopsis (Stcphcncphyllum) in four Bureja basin. According to the first inter• localities. Other a~sociated leaves belong to pretation given by Harris (1935) lateral Ctenis, Nilsscnia, Pteropl,yllum and Sphe• appendages of this fructification were seeds. nobaiera. It seems quite unlikely that Later Harris (1951) realized the other possi• one of them is a foliage of I xostrobus. bility - that appendages are fertile capsules. Harris provisionally assigned I xostrobus to This new interpretation is confirmed now Podozamites but the evidence is not very and some resistant to maceration megaspore strong. membranes were found within the capsules. I xostrobus consists of an axis and spirally Megaspore membranes are elliptical, folded attached microsporophylls composed of a and with few small apical protuberances thin stalk and more or less expanded ter• (archegonial necks ?). Staphidiophora is minal cup. This cup is built of five appen• very close to Leptostrobus and differs from dages. The basiscopic appendage is sterile it in much smaller dimensions of capsules and backwardly curved while the others and of fructification as a whole, in more are more or less fused and form a sinangium. regular distichous arrangement of capsules, Cutiele of sinangium is very thin and sto• etc. mata was not observed. Pollen grains are According to the authors' interpretation small, wingless, badly preserved. Micro• the stalk of a capsule is a reduced axis of sporophylls of I xostrobus differ from all 14 THE PALAEOBOTANIST
corresponding structures of other gymno• but some abnormal cylindrical shoots with sperms and some points of resemblance to a cluster of Stephenopkyllu1n leaves were a stamen of Angiospermae can be traced. found. The scale leaves on the fertile The Leaves - Harris (1951) put forward and sterile shoots are identical and their a convincing evidence of attribution of cuticles have frequent trichomes which Leptostrob~ts to Czekanowshia. In the Bureja are closc1y similar to trichomes on the cone basin L. ex gr. laxiflor us occurs in three axis of L. ex gr. erassipes. localities and in all of them Czehanowshia I obtained many hand-specimens on is a mostly abundant leaves. In two of which Staphidiophora and Hartzia are asso• them leaves of Pseudotorellia were found in ciated. Other plants in this locality are great number, but in the third (in the lower Nilssonia, Phoenieopsis, Ginkgoites, Pit yo• part of the Azanowshi section) Czehanowshia phyllum and Ixostrobus but none of them and isolated capsules of Leptostrobus ex gr. resembles Staphidiopltora in structure. laxiflorus completely cover the blocks sur• We may presume that all genera of Czeka• face and only some fragments of fern pinnu• nowskia group (Czehanowshia, Solenites, les associated with them. Similarity in Hartzia) as well as Stephenophyllum (and epidermal structure of Czehanowshia and probably other closely allied genera of Leptostrob~ts was fully discussed by Harris Phoenieopsis-group - Windwardia and C~tl• (1935, 1951) and Vachrameev and Doludenko goweria) belong to the Czekanowskiales. (1961). I found some additional evidence It would not be too impudent to suppose in the close agreement between the cuticles that Arctobaiera is also a member of the of capsule of L. ex gr. laxiflorus and basal new order Czekanowskiales because morpho• scale leaves of Czellanowshia, where the logically and anatomically this genus is epidermal cells are shorter and broader and identical with Stephenophyllum (see FLORIN, stomata are scattered and amphycyclic. 1936a). The only difference make some Leptostrobus ex gr. erassipes occurs in (but not all) leaves of Arctobaiera which four localities: (1) in the middle part of are split at the apices - not a great diffe• Umaltinski section with Stephenophyllum rence indeed. Florin regarded Aretobaiera (abundan t), Pseudotorellia, 5phenobaiera, to be closely allied to Sphenobaiera but the pterophyllum and I xostrobus; (2) in the upper true Sphenobaieras (allied to the type• part of Umaltinski section with a mass of species S. speetabilis) have two vascular Stephenophyllum leaves and one specimen bundles entering the leaf base (this was of Ixostrobus; (3) in the Tchegdomyn coal demonstrated on our Siberian material) mine on one block with Stephenophyll1l1n and numerous secretory cavities between and Nilssonia; (4) in the upper part of Aza• the veins. S. horniana - the only species nowski section with Stepltenopltyllu1n and which Florin had taken in consideration, Czekanowshia. Besides the evidence of asso• differs from typical Sphenobaiera and other ciation agreement in structure between Ginkgoales in its epidermal structure, secre• L. ex gr. erassipes and Stephenophyllum tory system and vascularization. I feel is very impressive. Stomata of the Burejean that it must be removed from the genus species of Stephenophyllum are of two types: Sphenobaiera and included in Stephe110• one with strongly thickened subsidiary phyllum or Aretobaiera. It seems that cells without a distinct papillae and the specimens from Ust-Baley described by other with rather thinly cutinized subsidiary Heel' as the leaves of Leptostrobus (L. rigida cells with prominent papillae bordering Heel' and L. angust1folia Heer) and redes• the stomatal pit. Stomata are distributed cribed later by Prynada as a new genus in narrow somewhat sunken bands or single Angariella had nothing to do with Czeka• rows. Epidermal cells are elongated, some nowskiales and had been "rather the Isoetes• of them with strong papillae and others like plants. with slight median elevation. Epidermis of As a whole leaves of Czekanowskiales the capsule of L. ex gr. eras sipes has close• are linear or ribbon-shaped, simple or forked ly similar cells and stomata (both types with few or numerous parallel veins. They are present), the distribution of stomata form a definite morphological range: is also similar, but more sparse. The shoot Czehanowskia (leaves narrow and repeatedly bearing Leptostrobus fructification is much forked) - Solenites (leaves narrow and longer than short shoots of Stephenophyl• forked once) -Hartzia (leaves narrow and lum which are as a rule almost spherical, simple or forked at the apices) - Areto- KRASSILOV - MESOZOIC" GINKGOALEAN " PLANTS FROM SIBERIA 15
baiera (leaves broader, with numerous veins, type (HARRIS, 1935), which are similar to simple or forked at the apices) - Phoeni• the ovules of Ginkgo, were found in the copsis (including Stephenophylhtm, Wind• Mesozoic and Cainozoic strata. These seeds wardia and Culgoweria -leaves ribbon• are provided with the thick and resistant shaped, unforked, with numerous veins). outer cuticle of integument. It is papillate Leaves are borne on the short shoots of and has amphicyclic stomata and conspi• limited growth and only one vascular bundle cuous openings of secretory cavities. Inner enters the leaf base. Epidermal cells are cuticle of integument and cuticle of nucellus elongated and make definite rows. Stomata are joined in a thin and colourless double arranged in narrow bands or single rows, cuticle. Megaspore membrane is resistant, longitudinally oriented, typically with strong• thick and tough. Associated leaves in ly thickened subsidiary cells. Mesophyll several cases belong to the genus Ginkgoites: among veins is devoted from secretory Ginkgoites taeniata Harris from the Rhaeto• cavities. They occur in sclerenchymatous Liassic of East Greenland, Ginkgoites ex gr. tissue under the vascular bundles. adiantoides (Unger) Sew. from the Lower Distribution - Czekanowskiales flourished Cretaceous of the Bureja basin and others. in the temperate Siberian flora where they However, such species as Ginkgoites lun• arrived in Late Triassic and disappeared zensis (Stur) Krausel, G. tigrensis. Archan• only in Late Cretaceous (Cenomanian• gelsky were attributed to the fructifications Turonian time). In Rhaeto-Liassic time (A ntholithes wettsteinii, Karkenia incurva) Czekanowskiales penetrated into Europe, which differ fundamentally from the cor• Greenland, Caucasus, Central Asia, China responding organs of Ginkgo. So we may and Japan, but in Late Jurassic they be• conclude that Ginkgoites is an artificial came rather rare in all these countries with genus and comprises the species of true subtropical and dry climate, and completely Ginkgoaceae and of other families of vanished from Europe, Central Asia, Mari• Ginkgoales. Probably this is also true for time Territory of the U.S.S.R. and Japan Baiera leaves which were recorded in the at the beginning of the Early Cretaceous association with Ginkgo-like seeds (TRALAU, (records of CZekanowskia from the Lower 1965). Cretaceous of the U.S.S.R. Maritime Terri• 2. The fructifications of Karkenia-type tory and Japan are not trustworthy). It is which are known from two localities• interesting that in the Mesozoic of North one from the Lower Cretaceous of Argentina America Czekanowsllia occurs but Phoeni• and the second from the Upper Jurassic copsis-group is completely absent (" Phoeni• of the Bureja basin. These fructifications copsis" from Cape Lisbourne belongs to consist of an axis bearing more than a hund• Sphenobaiera: Cahoon, 1960; "Phoenicop• red spirally attached ovules which form more sis ?" recorded by \Y. Bell from Canada is or less compact cylindrical or rounded an uncbssifiable fragment of linear leaf). strobilus (PL. 1, FIG. 1). Ovules are born There are no convincing evidence of exist• on the short stalk and are straight or in• ence of Czekanowskiales in the Southern curved but not inverted. No indications Hemisphere (as "convincing" I regard of "collar" were noticed. Ovules are the records of fructifications, short shoots oval with slightly projecting micropyle and with clusters of leaves but not the remains rounded or somewhat truncated chalazal of linear leaves with unknown epidermal end, the surface is striated. The integu• structure which are assigned to Czellanowsllia ment breaks into two halves. Outer cuticle or Phoenicopsis,- for instance, Phoenicopsis is thin and filmy, without stomata and elongata from the Jurassic of New Zealand: nonpapillate. The stone layer is well deve• JONES & JERSEY, 1947). loped consisting of the large oval cells with pitted surface (PL. 1, FIG. 9). Inner cuticle GINKGOALES of integument lining the comparatively long micropylar canal is equally thin. The Three types of female Ginkgoalean fructi• nucellus is free to the base and its cuticle fications are known: is prepared intact (PL. 1, FIGS. 6-8). Only 1. The megastrobili of living Ginkgo biloba. this nucellar cuticle was preserved under We have no fossil fructification analogous the strong oxidizing procedure. Nucellar to the biovulate Ginkgo megastrobilus, but beak is cylindrical with a rim of cuticle numerous ovules of Allicospermum xystum bordering a small rounded opening on its 16 THE PALAEOBOTANIST
top (PL. 1, FIGS. 3, 11). Nearly a half of epidermal structure, secretory cavities and the preparated nucellar cuticles enveloped venation). The Siberian species Eretmophyl• the megaspore membranes. At the micro• lum glandulosum (Samyl.) Krassil. (it had been pylar end of megagametophyte two small described earlier as Ginkgodium glandulosum) cavities occur and two pollen grains (or was found in the Bureja basin in association pollen tubes) are situated exactly opposite with abundant small seeds of Karkenia-type. these cavities (PL. 1, FIG. 2; TE'XT-FIG.1). 3. Supposed female organs of Pseudotorel• The cavities are interpreted as archegonia. lia. In many localities Pseudotorellia is Large egg cell is visible in one of them and accompanied by peculiar organs called two small neck cells above it. The space Umaltolepis (generic name is derived from between the archegonial neck and the pollen the Umalta River) which consist of a stalk tube is bordered by the curved dark lines bearing a single large terminal bract (PL. 2, and is somewhat darker than the rest of FIG. 12; PL. 3, FIGS. 22-25). The base of pollen chamber. Small oval body next to a stalk is surrounded with scale leaves. the neck may be the penetrating spermato• The bract is elongated, entire or sometimes zoid head. No central column between divided into two lobes, concave abaxially the archegonia was observed. and probably secured a single seed (though Karkenia incurva was referred by Archan• seeds were never found intact). The evi• gelsky to Ginkgoites tigrensis. I have rather dence of attribution Umaltolepis to Pseudo• convincing evidence of association for refer• torellia is: (1) Repeated association; (2) Scale ring Siberian species of Karkenia to Spheno• leaves at the base of Umaltolepis stalk are baiera (new species, PL. 1, FIGS. 4, 5, 10). structurally identical to the scale leaves The hand-specimens of Karkenia are in covering the short shoots of Pseudotorellia. most cases covered and penetrated with (3) Bracts of Umaltolepis and leaves of Sphenobaiera leaves. Other fossils in the Pseudotorellia are similar in their cuticle Karkenia bed are pterophyllum and Stepheno• though bracts have stomata on both sides phyllum. And there is no reason for attri• and their stomata are somewhat larger than buting any of them to Karkenia. If the in the leaves. Karkeniaceous affinity is true for Spheno• Isolated seeds resembling the seeds of baiera it might be true also for the closely Ginkgo were found in the localities where allied genus Eretmophyllum (leaves of Spheno• Pseudotorellia and Umaltolepis occurred. baiera and Eretmophyllum are similar in the These seeds (PL. 3, FIGS. 26-30) called
TEXT-FIG.I-Fertilization in Karkenia, X 175. PT-pollen tube, NC-neck cell, EC-egg cell. KRASSILOV - MESOZOIC" GINKGOALEAN •• PLANTS FROM SIBERIA 17
Burejospermum are oval, up to 10 mm. long (DORF, 1958), among them such Palaeozoic with fairly thick testa. Epidermal cells plants as Psygmophyllum, Gin1?gophyllum, are polygonal with thick walls and with• Saportaea, Burriadia, Phylladoderma, etc. out papillae. Stomata were not observed. Recently a new division of plant kingdom Nucellar cuticle with elongated cells form - Progymnospermae with the leaves of ing indistinct files. Resin canals (PL. 3, Psygmophyllmn - Ginl?gophyllum type was FIGS. 26, 31) are narrow and linear resemb• proposed and in light of this discovery the ling those of the leaves of Pseudotorellia. Ginkgoalean affinity of Palaeozoic leaf• A new material from the Bureja basin genera became rather doubtful. It seems gives some idea about the mode of growth that Palaeozoic genus Trichopitys has more of Pseudotorellia leaves. They were borne in common with Progymnospermopsida than on the short-shoots (PL. 2, FIGS. 13-18, 21) with true Ginkgoalean plants. The list which had a strong resemblance to the of Mesozoic Gin1?goales must be reduced by short-shoots of Ginkgo: they are cylindrical, excluding Czekanows1?ia, Phoenicopsis and not more than a 80 mm. long and 10 mm. allied genera which belong to the Czekanows• thick, few of them branched, covered with kiales. Ginkgoalean affinity is more or less small scales and leaf scars. The latter being certain for the following leaf-genera: Gink• slightly raised and bear two scars of vascular goites Seward, Baiera Braun, Sphenobaiera bundles (PL. 2, FIG. 16) and minute openings Florin, Baierella Potonie, Eretmophyllum of secretory canals. The apical bud was Thomas, Glossophyllum Krausel, Pseudo• protected with bud scales (PL. 2, FIG. 14) torellia Florin, Torellia Heer. as in the shoots of living Ginl?go. The These leaf-genera form four morphological leaves, leaving along their mode of vascu• groups: larization, resemble the foliage of some Arau• 1. Leaves fan-shaped, semicircular, trian• cariaceae more than the typical Ginkgoa• gular, entire orlobed with more than one vein lean leaves: they are entire, linear or oblan• in each segment, hypostomous or (rare) am• ceolate, without a definite stalk, gradually phistomous, petiole distinct, longer than leaf tapering to the base and more abruptly to blade. Resin bodies rounded or oval - the apex. Veins in the middle part of a Ginkgoites Seward leaf are unbranched and parallel. The Baiera Braun emend. stomata are longitudinally oriented and Florin arranged in distinct files. Stomatal pit 2. Leaves fan-shaped, divided, with one bordered with a prominent rim of cuticle vem in each ultimate division- (PL. 2, FIG. 20). Absence of papillae (be• Baierella Potonie sides the subsidiary cells) and trichomes 3. Leaves wedge-shaped, tongue-shaped, is unusual feature in Ginkgoalean leaves, oblong, obovate, entire or lobed, with more but the most extraordinary is the structure than one vein in each segment, amphistomous of secretory system: it consists of the straight or (rare) hypostomous, petiole absent or not and ± evenly thickened canals extending distinctly cut, shorter than leaf blade. Resin from the base of a leaf to the apex between bodies elongated, needle-shaped- each pair of veins (PL. 2, FIG. 19). These Sphenobaiera Florin peculiarities show that Pseudotorellia repre• Eretmophyllum Thomas sents an isolated group among Ginkgoales. Glossophyllum Krausel 4. Leaves linear, lanceolate, hypostomous, DISCUSSION AND CONCLUSION without petioles, resin canals continuous from the base to the apex of blade• Engler and Prantl included in the family Pseudotorellia Florin Ginkgoaceae the following genera: Ginkgo, ? Torrellia Heer emend. Baiera, Phoenicopsis, Czekanowskia and Ves• Florin quia. Besides Vesquia (which is regarded There are two types of female organs now as a seed of Taxaceae) all of them are of extinct plants - Karkenia and Umalto• the leaf-genera. Some Ginl?go-like fructi• lepis, resembling the Ginkgo megastrobilus fications were assigned to Baiera and Czel?a• sufficiently to justify their inclusion in the nowskia on the basis of indirect evidence. same order, but differing from it (and one A lot of the genera of leaves approaching from another) enough to distinguish each Ginkgo in their form, venation and/or mode of them as a representative of the family of growth were supposed to be Ginkgo allies of its own. Fertilization in Karkenia and 18 THE PALAEOBOTANIST
Ginkgo biloba is similar but the morphology Karkeniaceae is perhaps the most ancient of megastrobilus and cutinized seed mem• family of Ginkgoales known from the Nor• branes are different. If my interpretation thern as well as Southern hemisphere. The of Umaltolepis is true it might be analogous true Ginkgos (Ginkgoaceae) have started with the female strobilus of Ginkgo, but it from the Late Triassic of Northern hemis• has one terminal fertile structure instead phere (ovules of the Ginkgo type, entire of two. Its bract may correspond to the Ginkgoites leaves) and the Pseudotorelliaceae so called collar of Ginkgo ovules. are not known from any country besides It is supposed that the Ginkgoites• Eurasia and their range in time is from group of leaf-genera roughly corresponds Rhaetic to the Lower Cretaceous. to the Ginkgoaceae sensu stricto (though I believe that a new classification of some Ginkgoites and Baiera belong to the Ginkgoales proposed now would improve other family or families) and the Spheno• our knowledge of the geological history baiera group - to Karkeniaceae. Pseudo• and evolution of these plants, though toreUia and Umaltolepis form the third its preliminary nature must be accen• family Pseudotorelliaceae. tuated.
REFERENCES
ARCHANGELSKY,S. (1965). Fossil Ginkgoales from JONES O. A. & JERSEY, N. J. DE (1947). The the Tico flora, Santa Cruz province, Argentina. flora of the Ipswich coal measures - morpho• Bull. Brit. Mus. (nat. Hist.) Ceol. 10 (5): 121-137. logy and floral succession. Univ. Qld. Papers CAHOON, E. J. (1960). Sphenobaiera ikorfatensis Ceol. (n.s.) 3: f. pap illata from the Lakota formation of the KRAUSEL, R. (1943). Die Ginkgophyten der Trias Black Hills. Bull. Torrey bot. Cl. 87 (4): 247- von Lunz in Nieder-Osterreich und von Neue 257. . Welt bei Basel. Palaeontographica. 87 B: DORF, E. (1958). The geological distribution of 59-93. the Ginkgo family. Bull. Wagner Free Inst. OISHI, S. (1933). A study on the cuticles of some Sci. 33 (1): 1-10. mesozoic gymnospermous plants from China FLORIN, R. (1936a). Die fossil en Ginkgophyten and Manchurica. Sci. Rep. T{jhoku Imp. Univ., von Franz-Joseph Land nebst Erorterungen ser. 2, 12 (2): 239-252. tiber vermeitliche Cordaitales mesozoischen PANT, D. D. (1959). The classification of gymnos• Alters. I. Spezieller Teil. Palaeontographica. permous plants. Palaeobotanist. 6 (2): 65-70. 81, Abt. B. (3-6): 1-173. PRYNADA, V. D. (1962). Mesozoic flora from the Idem (1936b). Die fossilen Ginkgophyten von East Siberia and Zabaikalie. Moscow (in Franz- Joseph Land nebst Erorterungen tiber Russian). vermeitliche Cordaitales mesozoischen Alters. SEWARD,A. C. & GOWAN, J. (1900). The Maiden• 2. Allgemeiner Teil ibid. 82 B. (1-4): 1-72. hair Tree (Ginkgo biloba L.) Ann. Bot. Lond. 14: HARRIS, T. M. (1935). The fossil flora of Scoresby 109-154. Sound, East Greenland. Medd. om Cronland TRALAu, H. (1966). Botanical investigations in 112 (1): 1-176. the fossil flora of Eriksdal in Fyledalen, Scania. Idem (1951). The fructification of Czekanowskia Sveriges Ceol. Undersokning. ser. C, (611): and its allies. Philos. Trans. R. Soc. Lond. Ser. 1-36. B. 235 (628): 483-508. VACHRAMEEV,V. A. & DOLUDENKO, M. P. (1961). HEER, O. (1876). Beitrage zur Jura - Flora Ost• Late Jurassic and Lower Cretaceous flora of the sibiriens und des Amurlandes. lVIem. l'A cad. Bureja basin and its stratigraphical significance. Imp. Sci. St.-Pet. 22 (12): 1-122. Akad. Sci. U.S.S.R., Moscow (in Russian).
EXPLANATION OF PLATES
PLATE 1 5. Basal part of Sphenobaiera leaf with two vas• cular bundles entering the leaf base, with secretory 1. I I 2 3 5 6 I I KRASSILO' r - PLATE 2 THE PAL.\ EOBOTANIST, VOL. 18 {' ' . , • 16 ••-17 -- ~. • • __.tI •• THE PALAEOBOTANIST, VOL. 18 I\:RASSILOV - PLATE 3 22 25 26 27 , 28 29 30 31. KRASSILOV - MESOZOIC" GINKGOALEAN •• PLANTS FROM SIBERIA 19 11. Karkenia sp., upper part of the nucellar 19. Pseudolorellia, fragment of foliage leaf with cuticle, same specimen as in figure 8. Note the resin canals (dark longitudinal lines) (specimen rim of cuticle bordering the opening of nucellar No. 501-522) X 8. chamber and fragments of the inner cuticle of 20. Pseudolorellia, papillate stoma (specimen integument lining the micropylar canal (specimen No. 515-258) X 474. No. 516-387) X 69. 21. Pseudolorellia, abaxial cuticle of ~caleleafwith occasional stoma (specimen No. 515-284) X 175. PLATE 2 PLATE 3 12. Four leaves of Pseudolorellia and two scales 22. Two Umaltolepis structures consisting of a of Umaltolepis (specimen No. 515-224) X 1. stalk with scale leaves surrounding its base and 13. Branching short shoot of Pseudolorellia (speci• terminal bract (specimen No. 515-271) X 2. men No. 515-279) X 1,5. 23. Umaltolepis, outer surface of two-lobed 14. Pseudolorellia, short shoot with apical bud bract (specimen No. 515-125) X 2,5. protected with bud scales (specimen No. 515-284) 24. Umaltolepis, inner surface of entire bract X 2. (specimen No. 515-121) X 2,5. 15. Pseudolorell-ia, short shoot bearing two foliage 25. Cuticle of the bract (specimen No. 515-271) leaves intact (specimen No. 515-289) X 2. X 90. 16. Part of the short shoot same as in the figure 3 26-29. Seeds attributed to Pseudolorellia. Seed showing scale leaves and leaf-scars with two rounded coat in figure 26 with clearly marked resin canals traces of vascular bundles (specimen No. 515-289) (specimens No. 501-532, 533, 534, 535, 536, 537, X 5. 538, 544, 539) X 2,5. 17. Pseudolorellia, resin bodies of the scale leaves 30. Seed coat with resin canals (specimen No. (specimen No. 515-284) X 10. 509-499) X 8. 18. Pseudolorellia, adaxial cuticle of scale leaf 31. Cells of outer cuticle of the seed and resin with stomata (specimen No. 515-284) X 175. canal (specimen No. 501-540) X 175.