| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Paleobotany |
2Paleobotany and Palynology Laboratory, Florida Museum of Natural History, University of Florida, Gainesville, Florida 32611-7800 USA; 3Laboratório de Geociências of the Universidade Guarulhos and Instituto de Geociências of the Universidade de São Paulo, São Paulo, SP, Brazil; 4Laboratoire de Paléobotanique et Paléoécologie, UMR 5143, Université Pierre et Marie Curie, Paris, France
Received for publication June 24, 2004. Accepted for publication April 21, 2005.
ABSTRACT
Welwitschiaceae, a family in the Gnetales, is known today from only one extant species, Welwitschia mirabilis. This species is distributed in the Namibian desert, along the western coast of southern Africa, about 10 km inland from the coast. Very little is known about the fossil record of this family. Lower Cretaceous megafossils of various organs, assigned to Welwitschiaceae, are presented here. These fossils include young stems with paired cotyledons attached (Welwitschiella austroamericana n. gen. et sp.), isolated leaves (Welwitschiophyllum brasiliense n. gen. et sp.), and axes bearing male cones (Welwitschiostrobus murili n. gen. et sp.). They were collected in the Crato Formation, which is dated by palynomorphs and ostracods as Late Aptian (114 to 112 million years ago). These sediments are exposed in the Araripe Basin of northeastern Brazil. This study brings together new information of the megafossil record of Welwitschia-like plants and also reports of pollen said to be similar to that of Welwitschia from Lower Cretaceous sediments.
Key Words: Aptian • Brazil • Crato Formation • Cretaceous • Gnetales • Welwitschia
The Gnetales is a group of seed plants that is poorly documented in the fossil record (Crane, 1988
, 1996
). Recent studies of the Lower Cretaceous sediments in Brazil provide important new data about the occurrence of the Gnetales (Osborn et al., 1993
) and in particular the fossil record of Welwitschiaceae (Rydin et al., 2003
). Welwitschiaceae is a family of Gnetales, represented today only by the unique species Welwitschia mirabilis (Hooker, 1863
). This species lives scattered along a restricted strip of land 1200 km long and 140 km wide along the southwestern coast of Africa from the Nicolau River (north Mossamedes or Namibe, Angola) to the Kuiseb River (Swakopmund, Namibia). The western boundary occurs about 10 km inland from coastal South Africa, and its easternmost boundary may extend up to 150 km inland. This area corresponds to the northern and central part of the Namibian Desert, reaching eastward to the Mopane Savanna (Kers, 1967
; von Willert, 1985
; fig. 4 in Crane and Hult, 1988
).
|
The megafossils presented in this paper including leaves, seedlings, and pollen cones provide evidence that the family Welwitschiaceae, currently endemic to southwestern Africa, was present in South America and probably was even much more widespread during the Lower Cretaceous. In addition to these megafossils, polyplicate pollen grains belonging to five genera with gnetalean affinities have been identified in the Lower Cretaceous sediments of the Crato Formation. They include Equisetosporites, Gnetaceaepollenites, Singhia, Steevesipollenites, and Regalipollenites distributed in 52 palynomorph species (Table 1). These species, and others related to gnetaleans, are widespread throughout Northern Gondwana in the Early Cretaceous (Tables 1, 2).
|
|
). It has been suggested that some of these pollen species are related to Welwitschiaceae (Lima, 1983
; Osborn et al., 1993
) because of their elliptical shapes, ribbed or striate surfaces, and the presence of a distinct sulcus, characters that are similar to extant Welwitschia pollen. Welwitschia is considered to be entomophilous because of the stickiness of the pollen that would hinder wind dispersal (Endress, 1996
; Wetschnig and Depisch, 1999
; J. Armstrong, Illinois State University, personal communication). Generally, entomophilous species have very limited pollen dispersal. Therefore, we might expect to find the megafossils of Welwitschiaceae in the same sediments that yield pollen types similar to extant Welwitschia.
The first to recognize gnetalean megafossils in the Crato Formation, which tentatively were attributed to Ephedraceae and Welwitschiaceae, were Pons et al. (1992)
and Bernardes-de-Oliveira et al. (1999
, 2000)
. A diverse assemblage of plants with gnetalean affinities was noted in Mohr and Friis (2000)
, while Rydin et al. (2003)
proposed Cratonia cotyledon, for a seedling they relate to Welwitschia. In this paper we present both vegetative and reproductive fossils of Welwitschiaceae from the Crato Formation.
The abundant and diverse plant macrofossils from the Crato Formation are frequently well preserved because of their calcification. The first published research on the fossil plants of the Santana Group was by Duarte (1985)
. She identified Brachyphyllum from the Crato and Romualdo Formations and Podozamites, Nymphaeites, and Choffatia from the Crato Formation. Since then, there has been a growing body of literature of the megafossil plants from the Crato and/or Santana Formations (Crane and Maisey, 1991
; Pons et al., 1992
; Martill et al., 1993
; Barreto et al., 2000
; Mohr and Friis, 2000
; Mohr and Rydin, 2002
; Mohr and Eklund, 2003
; Rydin et al., 2003
; Mohr and Bernardes-de-Oliveira, 2004
). An overview of the Crato Formation flora presented by Martill et al. (1993
, plates 10–13) demonstrated the presence of gymnosperms, angiosperms, and possible gnetaleans. Bernardes-de-Oliveira et al. (1993
, 1999
, 2000
) recorded the presence of Araucariaceae and Welwitschiaceae. A winged fruit, which was caliciform, syncarpic, and tetralocular, was described by Barreto et al. (2000)
. A synthesis of fossil plants from the scientific collections of the University of São Paulo, the Federal University of Pernambuco, the Departamento Nacional Produção Mineral Museum (in Crato, state of Ceará [CE]), and the Paleontological Museum of Rural Cariri University (URCA, in Santana do Cariri, CE) was presented by Dilcher et al. (2000a
, b
). The collections include leaves attached to stems articulated like Schizoneura, bulblike structures of Isoetites with preserved sporophylls, fern fronds, conifer bracts, a bifoliate seedling and young isolated leaves of Welwitschiaceae, Ephedra-type stems with leaves and attached strobili, angiosperm leaves, folicules, winged fruits, and petals. Mohr and Eklund (2003)
noted, based upon a collection of Crato Formation plants purchased by the Museum der Naturekunde in Berlin, that the flora represents elements from terrestrial environments and also a few reputed aquatic taxa. Of the 80 species they recognize from this collection, 20 taxa are assigned to angiosperms, predominately dicotyledons, and one putative monocotyledon.
Locality and stratigraphy
The Araripe Basin is situated in the interior of northeastern Brazil and extends into the states of Piauí, Pernambuco, and Ceará occupying an area of 8000 km2. It is located between meridians 38°30' and 4050' W longitude and 7°5' and 7°50' S latitude. It consists of Paleozoic and Mesozoic sediments, forming one of the inland Mesozoic basins originating at the time of the breakup of Gondwana and the initial opening of the South Atlantic Sea (Berthou, 1990
; Brito-Neves, 1990
; Martill et al., 1993
; Ponte and Ponte Filho, 1996
) (Fig. 1).
|
). A new lithostratigraphic column for the Aptian to Albian sequence in this basin was formally proposed by Neumann (1999)
and Neumann and Cabrera (1999)
and is followed here (Fig. 3). In this proposal, the Santana Group includes all sedimentary post-rift sequences of the Araripe Basin consisting of the Rio Batateira, Crato, Ipubi, Romualdo, and Arajara Formations.
|
; Neumann et al., 2002
) (Fig. 2). The environment suggested by Bechly (1998)
and Mohr and Rydin (2002)
was marine or lagoonal. However, Maisey (1991)
makes a strong case for the basin to represent low-energy inland freshwater lakes. The sediments indicate that there were some episodic higher energy events in the lake system, and the few echinoids near the top of the sediments suggest a late marine incursion into this freshwater lake system.
The underlying Rio da Batateira Formation may be gradational or interdigitate with the Crato Formation and sometimes the Crato Formation lies disconformably on the Abaiara Formation (Neocomian) or on the Cristalin Basement rocks. Its maximum thickness is about 30 m (Martill et al., 1993
). The upper contact with the Ipubi Formation is normal to gradational. In the sediments of the Crato Formation there are excellently preserved ostracods, insects, arachnids, bivalves, gastropods, fishes, amphibians, crocodiles, pterosaurs, lizards, birds, coprolites, algae, pteridophytes, gymnosperms, angiosperms, and their palynomorphs (Maisey, 1991
). The Crato Formation limestones are exploited for the cement industry and ornamental rocks, and during this mining abundant fossil remains are exposed.
Based on palynomorphs and ostracods, Hashimoto et al. (1987)
, Arai et al. (1989
, 1999
, 2001
), and Coimbra et al. (2002)
considered that the Rio da Batateira and Crato Formations belong to the lower Alagoas stage. This included the Sergipea variverrucata palynozone that corresponds to the Upper Aptian (Fig. 3). The presence of pollen grains in these sediments, including Classopollis, monosulcate cycadophytes, and polyplicates related to Gnetales (including probable representatives of Ephedra and Welwitschia), has been used to suggest a dry regional climate, with high diurnal temperatures in the area where the Crato sediments were deposited (Lima, 1983
). A similar environment also has been postulated for other parts of Northern Gondwana (Vakhrameev, 1970
; Brenner, 1976
; Doyle, 1977
; Herngreen and Chlonova, 1981
; Herngreen et al., 1996
), although Doyle et al. (1982)
suggested that some regional climates were warm, but not arid.
MATERIALS AND METHODS
Most of the plant fossils described in this study belong to the "Murilo Rodolfo de Lima" Collection of the Departamento de Geologia Sedimentar e Ambiental, Instituto de Geociências, Universidade de São Paulo. They are catalogued under GP/3E numbers. Other specimens catalogued as SMB were made available by Dr. Volker Wilde from the collections of the Senckenberg Museum of Natural History, Frankfurt am Main, Germany. The plant fossils were collected from several outcrops of the Crato Formation, all situated in the northern flank of the "Chapada do Araripe." Collections were made by amateur collectors or purchased from collectors and donated to the Universidade de São Paulo. Collectors include Mrs. Urania Gusmão Corradini, Mrs. Maria Aparecida Vulcano, and Prof. Dr. Murilo Rodolfo de Lima. One specimen from the "Murilo Rodolfo de Lima" Collection was given to the Paris Museum of Natural History (MNHN) and was loaned for this investigation. The material from the Senckenberg Museum of Natural History was purchased from a collector who purchased and exported them from the same region and formation in Brazil. Modern material were examined from the Paleobotany Division, Florida Museum of Natural History, University of Florida, and supplied by Dr. Joseph E. Armstrong, Illinois State University.
Some plant megafossils from the Crato Formation are preserved as impressions in thin laminated limestone, but the majority contains delicately preserved morphological details. According to X-ray analysis done at the Instituto de Geociências of the Universidade de São Paulo, permineralization occurred through partial replacement of plant tissues by iron minerals (goethite) and calcite. The plant fossils in this report were studied under a stereomicroscope with a camera lucida and with a Zeiss Axiophot optical binocular microscope with epifluorescence illumination (Zeiss, Thornwood, New York, USA).
SYSTEMATICS
Division
Trachaeophyta.
Class
Gnetatae/Gnetopsida.
Order
Gnetales.
Family
Welwitschiaceae Markgraf (1926).
Genus
Welwitschiella Dilcher, Bernardes-de-Oliveira, Pons et Lott, gen. nov.
Type species
Welwitschiella austroamericana Dilcher, Bernardes-de-Oliveira, Pons et Lott sp. nov.
Generic diagnosis
An ovoid-shaped principal axis bearing two small opposite lanceolate cotyledons diverging from the apical area. Cotyledons with first-order venation parallel and equidistant throughout, each vein originating successively from a marginal vein. Parallel second-order veins interspersed between first-order veins.
Etymology
Genus "Welwitschia" plus "ella" (Lat.) diminutive suffix for "little" in order to say "small Welwitschia."
Species
Welwitschiella austroamericana Dilcher, Bernardes-de-Oliveira, Pons et Lott sp. nov. (Figs. 4, 7–10).
Specific diagnosis
Principal axis oval to round, 9–10 mm long and 7–9 mm wide, with broad end-bearing cotyledons. Two opposite cotyledons arise from the axis at a 90° angle, often arching out to an angle of 45–60°. The cotyledons are oblong-lanceolate, 5.8–6.3 cm long and 0.7–1 cm wide. The cotyledons have a dichotomous pair of first-order veins just above attachment to the axis. One vein appears unbranched and traverses half the cotyledon distance. The second vein produces parallel veins, each vein originating successively upward, 0.6–1 mm from each other, at an angle of divergence of 20–30°. Eight to 10 parallel first-order veins observed. Parallel second-order veins interspersed between first-order veins with possibly a few thin oblique lateral second-order veins present that arise from first-order veins at a 10–20° angle, joining opposite second-order veins forming apically oriented, poorly developed chevrons.
Etymology
Species "austroamericana," from South America.
Holotype
GP/3E-7529, "Murilo Rodolfo de Lima" Collection, Departamento de Geologia Sedimentar e Ambiental, Instituto de Geociências, Universidade de São Paulo, Brazil.
Additional specimens
GP/3E-7530a, b.
Number of specimens examined
Three.
Locality
Chapada do Araripe, northeastern Brazil.
Age and stratigraphy
Late Aptian, Crato Formation.
Description
Welwitschiella austroamericana consists of an axis bearing two lateral cotyledons. The main axis is round, to ovoid, to broadly triangular in shape and laterally compressed. It probably was cylindrical in form during life. The base appears rounded or it may taper, both ending in a narrow hypocotyl (broken off in Figs. 4, 7, and 10 [about 1 mm in diameter] or the hypocotyl may be absent in Fig. 9). The axis apex appears to be flat, and the cotyledons appears to arise from either side of the axis and extend upward and arch outward from this disc (Figs. 4, 7, 9, 10). The surface of some axes may be rugose while others are smooth. The cotyledons are oblong-lanceolate in shape, often with a missing apex, and the ends are often frayed. The basal portion of the cotyledons taper to their point of attachment with the axis and may appear slightly twisted. The cotyledon bases expand down below the point of attachment and appear to clasp or arise from the sides of the young woody axis, resulting in swollen areas on either side of the axis (Figs. 4, 7, 9). These bases seem to remain intact, even when the axis begins to disintegrate. Near the cotyledon attachment is a pair of first-order veins that appear to dicotomize. Because of the twisted nature of most of the cotyledons, it is difficult to trace the complete venation of a cotyledon. But as shown in Fig. 8, the venation of half of a cotyledon is seen. One vein branches five times as it continues along the cotyledon margin. This results in several parallel veins that extend to the tip of the cotyledon where some appear to fuse. A total of 8–10 parallel, first-order veins are present. Several parallel second-order veins are interspersed between the first-order veins. Toward the apex, most of the lateral veins end at the lateral margin. In only a few areas of one specimen is it possible to observe weakly developed oblique lateral second-order opposing veins joining to form poorly defined apically oriented chevrons (Fig. 8).
Discussion
The morphology of these cotyledons exclude a close relationship to monocotyledons by the organization of the venation, such as an absence of a midvein, and longitudinal veins that end blindly at the lateral margins. Also, monocotyledons have an alternate, rather than an opposite phyllotaxy. A close relationship to palm seedlings is further excluded by the absence of plications (Uhl and Dransfield, 1987
). The morphology of these cotyledons are similar to those of Welwitschia mirabilis seedlings, in characters such as their opposite and divergent orientation, the tapering at their attachment to the young axis, their expanded and clasping base that is broadly attached to the axis (Figs. 4, 7, 9), and their parallel first-order venation with weakly developed chevrons (Rodin, 1953
; Sykes, 1910
, 1911
; Figs. 5, 6). We were able to observe 18-mo-old greenhouse seedlings (J. Armstrong, Illinois State University, photographed, measured, and sent material) that show persistent cotyledons as the young leaves begin to grow (Figs. 5, 6). The size ranges of extant cotyledons are 2.8–3.5 cm long and 3–5 mm wide, each with 6–8 veins (Sykes, 1910
; Rodin, 1953
; J. Armstrong, Illinois State University, personal communication). The fossil cotyledons presented here are 6 cm long and 7–10 mm wide, each with 8–10 veins. The sizes of the fossil cotyledons are slightly larger than any recorded cotyledons from greenhouse-grown plants. The classic chevrons or inverted Ys seen in the venation of the cotyledons consists of hypodermal fibers (Rodin, 1953
), which do not seem to be well-preserved in the fossils, but can be observed with careful observation (Fig. 8).
Genus
Welwitschiophyllum Dilcher, Bernardes-de-Oliveira, Pons et Lott gen. nov.
Type species
Welwitschiophyllum brasiliense Dilcher, Bernardes-de-Oliveira, Pons et Lott sp. nov.
Generic diagnosis
Isolated leaves of indefinite length, short or elongated, triangular to linear, entire margin, isobilateral symmetry, coriaceous, with numerous subparallel vascular bundles, usually with high venation density. Base of maximum width and is curved/enrolled.
Etymology
Genus "Welwitschia" plus "phyllum" (Gr.), leaf.
Species
Welwitschiophyllum brasiliense Dilcher, Bernardes-de-Oliveira, Pons et Lott sp. nov. (Figs. 11, 12, 16).
|
Etymology
Species "brasiliense" (Lat.), from Brazil.
Holotype
GP/3E-6035, "Murilo Rodolfo de Lima" Collection, Departamento de Geologia Sedimentar e Ambiental, Instituto de Geociências, USP.
Paratypes
GP/3E-6033, -6034a and -6034b, SMB 16480.
Number of specimens examined
Seven.
Locality
Chapada do Araripe, northeastern Brazil.
Age and stratigraphy
Late Aptian, Crato Formation.
Description
The leaves (Figs. 11, 12, 16) are elongated to triangular shaped, ranging from 8.9 to 70 cm long and 2.8–5 cm wide with a basal width of 3.5–5.4 cm. The leaf symmetry is isobilateral with an entire margin, and the leaf lamina is thick and coriaceous in texture. The leaves are often creased longitudinally from the base to just less than half the leaf length. The apex is acute (Fig. 12) or may be frayed and worn, breaking the leaf into longitudinal segments (Fig. 16). The maximum width is at the base that may be semicircular or curved with tissue extending downward to form the curve (Fig. 16). Other leaf bases show a slight curvature and/or may appear to be thickened or rolled (Figs. 11, 12). These leaf base shapes suggest a direct attachment to the stem along the entire leaf base width. The venation is formed by subparallel vascular bundles, somewhat convergent near the apex with some disappearing into the margin (Figs. 11, 12). The preservation of typical leaves allows for only an approximate vein count, with a high vein density of about 9–15 bundles/cm, with 40– 50 veins from margin to margin. The preservation of one specimen (Fig. 11) allows for an exact count of 65 veins near the base and 44 veins in the middle region. Basal veins may be 0.8–1.0 mm thick, while the veins in the mid-section of the leaf are equidistant and 0.4–0.5 mm thick. Only seven specimens were available for study, and more variation in size and form probably exists, which should be seen as more material becomes available.
Discussion
Characters relating these leaves to Welwitshiella austroamericana and thus to the co-occurring fossils are parallel first-order veins that are equidistant, convergent near the apex, with some veins disappearing into the margin. Although a number of synapomorphies of Gnetales (Crane, 1996
) were not observed, the following characters are similar to Welwitschia: the isobilateral form of the leaves, possible thickening of the epidermis, triangular elongated leaf shape with a wide base, longitudinal splitting from a frayed leaf apex, a somewhat thickened or creased mid-leaf area (Figs. 11, 12), and numerous parallel veins. The frayed leaves are not particularly long and may have been broken in deposition rather than as in wind-blown frayed ends of the long and large leaves of living W. mirabilis. Some of the leaf blade bases suggest an attachment around a stem, long tapering leaf length, and the rolled base, suggesting the leaf may have been enrolled into or ensheathed in deep grooves (Kubitzki, 1990
). These isolated leaves somewhat resemble the long, linear, parallel-veined Desmiophyllum leaves that are associated with microsporophylls of Piroconites from the Early Liassic of Franken, Germany (van Konijnenburg-van Cittert, 1992
; Crane, 1996
) (Table 2). The Desmiophyllum/Piroconites complex may be related to Gnetales (van Konijnenburg-van Cittert, 1992
), with Desmiophyllum leaves similar to Welwitschia but without the cross veins (Doyle, 1996
). Desmiophyllum-like leaves have been noted in the Crato Formation (Crane, 1996
), with four illustrated specimens described as "large parallel-veined reed-like leaves of uncertain affinity" (Maisey, 1991
, p. 429).
Genus
Welwitschiostrobus Dilcher, Bernardes-de-Oliveira, Pons et Lott gen. nov.
Type species
Welwitschiostrobus murili Dilcher, Bernardes-de-Oliveira, Pons et Lott sp. nov.
Generic diagnosis
Cones are terminal or axillary on slender striated axes. Cones with a decussate pattern of striated scales. Each scale is folded along its median longitudinal line, giving the cone a square cross section. Scale apex acute.
Etymology
Genus "Welwitschia" and "strobus" (Lat.), cone.
Species
Welwitschiostrobus murili Dilcher, Bernardes-de-Oliveira, Pons et Lott sp. nov. (Figs. 13, 14, 17, 18).
Specific diagnosis
Striated axes 30–210 mm long by 2.5– 3.4 mm wide. Reproductive cones narrowly elliptic to elliptic. Cone axis diameters are 0.5–1.1 mm.
Etymology
Species "murili" (Lat.), in honor of Prof. Murilo Rodolfo de Lima, the palynologist who assembled much of the collection of fossils used in this paper.
Holotype
MNHN 40033, Muséum National d'Histoire Naturelle (MNHN), Paris.
Additional specimens
GP/3E-5802, -7531.
Number of specimens examined
Three.
Locality
Chapada do Araripe, northeastern Brazil.
Age and stratigraphy
Late Aptian, Crato Formation.
Description
The reproductive cones are terminal or axillary, with a decussate pattern of paired scales. All the scales have an acute apex. The specimen MNHN 40033 (Figs. 13, 14) has an axis 33 mm long and 3 mm wide, terminating in a central cone and two opposite branches. Each branch is 30– 35 mm long and 2.5 mm wide, with the apices bearing cones organized in a dichasium. Cone scales are striated, with the apices forming the corners of a square, and curved toward the apex. The apex of the left branch bears one well-preserved cone and two fragmented and poorly preserved cones (Fig. 14). The well-preserved cone is 20 mm long and 6.5 mm wide. The central cone is partially preserved and is 17 mm long and 4 mm wide, showing a central axis approximately 0.5 mm in diameter with axillary bracts. The right lateral cone is too poorly preserved to be measured. The right axis bears cones that appear smaller and perhaps less developed than those just described. The cones at the base of this dichasium are 10 mm long and 6 mm wide. The central cone is incomplete, with scales preserved at the base and distally axillary bracts. The reproductive cones of GP/3E-5802 (Fig. 17) and GP/3E-7531 (Fig. 18) are axillary and terminal, on striated monopodial articulated axis, measuring 55–210 mm long and 3–4 mm wide. These axes bear opposite, lateral axes that emerge at nodes. Both the main axes and lateral axes are striated, and the lateral axes appear to ensheathe the axis from which they arise. It is difficult to determine whether the lateral axes represent leaves, bracts, axes, or remnants of old cone axes. The cones measure 40 mm long and 10 mm wide, with an axis 1.1 mm in diameter (Fig. 18).
Discussion
The dichasial organization, in which reproductive spikes are borne both terminally and laterally at the apex of stems found in Welwitschiostrobus murili, resemble Drewria potomacensis Crane and Upchurch (1987)
. However, the pollen cones described here are compact cones (Figs. 13, 14, 17, 18) and are quite similar to recent Welwitschia pollen cones (Fig. 15). They are unlike the loose reproductive structures (spikes) of D. potomacensis. The specimens illustrated in Figs. 17 and 18 have opened their cone scales after pollen dispersal. Welwitschiostrobus murili represents pollen cones that are quite similar to recent Welwitschia pollen cones described by Pearson (1929)
and ones we examined from our collection in the Florida Museum of Natural History (Fig. 15) in size, form, decussate arrangement of scales, and position on long slender axes. Attempts to prepare pollen from these fossil cones and to observe pollen in or attached to them by epifluorescence microscopy was not successful. The scales of W. murili (Figs. 13, 14) are similar to the basal scales with acute apices of Welwitschia male cones. It is difficult to observe the subtending bracts typical of Welwitschia (Pearson, 1929
) in the cones of Welwitschiostrobus murili. The reproductive cones illustrated in Figs. 17 and 18 are somewhat similar to the cones in Martill et al. (1993
: Plate 12, Fig. 2). For this study, only one species of fossil cone is proposed. When more material is available, the cones illustrated in Figs. 17 and 18 may yield sufficient characters to keep or remove them from W. murili.
DISCUSSION
The flora of the Crato Formation is considered allochthonous (Martill et al., 1993
) that is reflected in the state of the preservation in our specimens. The broken or frayed tips of some leaves (Fig. 16) and a few cotyledons (Figs. 4, 9) suggest that some transport of the fossils reported here took place when much of the Crato Formation plant material was deposited. The basal portions of seedling specimens are abraded and broken at the hypocotyl with no roots preserved (Figs. 4, 7, 9, 10), suggesting that the young seedlings were broken from their growth position and transported to a site of deposition. The overall placement and character of the seedlings are reminiscent of the dispersal pattern of Welwitschia mirabilis seedlings in dry riverbeds of Southwest Africa (Pearson, 1906
). It appears that these Welwitschiella seedlings represent very young plants that were ripped up and deposited when still in a cotyledon stage before bearing their first pair of foliar leaves. We suggest here that Welwitschiella specimens are probably young seedling plants in the cotyledon stage because the fossils demonstrate numerous characters of Welwitschia seedlings in the cotyledon stage detailed next. Data on extant Welwitschia is based on literature (Sykes, 1911
; Rodin, 1953
) and photographs, specimens, and personal communications (supplied by J. Armstrong, Illinois State University). The seeds germinate readily and remain in a cotyledon stage for several months. After approximately 6–8 mo, paired leaves are produced at the apex of the epicotyl, or short broad axis, from which the cotyledons remain attached (Fig. 5). These paired leaves are held together and grow straight up between the cotyledons. Just such a 2-yr-old seedling is illustrated by Sykes (1911)
. Eventually the cotyledons drop away, and the two paired leaves will fold down pointing in opposite directions, the typical form of Welwitschia familiar to most. As the axis enlarges, the width of the leaf base also increases. So the leaves shown in Figs. 11, 12, and 16 probably came from young plants with larger axes than those illustrated here in the cotyledon stage (Figs. 4, 7, 9, 10). Plants of reproductive age were present as evidenced by the pollen cones also preserved in the Crato Formation, Welwitschiostrobus murili. While these cones do not have pollen preserved, there are numerous types of Welwitschia-like pollen grains described from the Crato Formation (Table 1).
Another fossil, Cratonia cotyledon, recently described from the Crato Formation, is considered to represent the cotyledon stage of a closely related genus to Welwitschia (Rydin et al., 2003
). Cratonia cotyledon appears to be represented by a single specimen and has distinct chevron or "Y" venation that allies it to Welwitschia. The specimen consists of two broad leaves that overlap one another, a feeder and a root, but the hypocytol is reported as missing. The leaves are 40 mm long and 18 mm wide at the base and have approximately 20 main veins. In C. cotyledon, the proposed cotyledons are entirely different from cotyledons of Welwitschia (Figs. 5, 6) and Welwitschiella austroamericana described here (Figs. 4, 7–10). The leaves of C. cotyledon differ from extant cotyledons and our specimens by their broad base, more or less uniform width, numerous veins, lack of a tapering base, leaves not clasping the epicotyl axis, and the pronounced display of the "Y" venation pattern. The cotyledons of extant Welwitschia often appear to spread out, are narrow (maximum of 5 mm width), have few main veins (6–10), taper near their base, clasp the young axis or epicotyl partially enclosing it, and the "Y" pattern of venation is less well formed than in the mature leaves. Also, the axis splits and increases in diameter as the leaves are produced (Fig. 6: extant; Fig. 7: fossil, before leaves are initiated). Cratonia cotyledon is similar to what we have seen for the first postcotyledonary leaves in seedlings of Welwitschia (Figs. 5, 6) where the leaves have a broad base, from which they continue to grow, and are oriented in an overlapping fashion. Because C. cotyledon does not exhibit these characters, we suggest that it may be a seedling in an early leaf stage from which the cotyledons and portions of the axis are missing. Conversely, it may be a seedling stage of a plant with cotyledons similar to but differing in the above characters from what is known for Welwitschia today, perhaps in an extinct line of plants.
Welwitschiella austroamericana, while very similar to Welwitschia seedlings in the cotyledon stage, does have some differences. The cotyledons of W. austroamericana are longer (fossil: 5–6 cm; extant: 3–4 cm) and wider (fossil: 5–6 mm; extant: 3–6 mm) compared to extant cotyledons. Also, the number of veins given in the literature (Sykes, 1911
; Rodin, 1953
) of approximately 6–8 in Welwitschia cotyledons is less than the 10–12 estimated in W. austroamericana. We observe only a few, poorly preserved chevrons or "Y" type patterns in the venation of the fossils. This may be because the chevrons or "Y" vein pattern in the cotyledons are the result of the presence of only a few hypodermal fibers (Rodin, 1953
), in contrast to a well-formed pattern in the vascular bundles of the mature leaves of Welwitschia (Rodin, 1953
). The hypodermal fibers of the cotyledons would not preserve as well as those of the mature leaf vascular bundles. Some of these differences between the extant and fossil material may be the result of where the seeds were grown. The fossils must have grown outside in an open environment, while seedlings reported in the literature were grown in greenhouses. Also, the Brazilian, Cretaceous-age plants may be a distinct population, with distinct characters, separate in time and space from the extant African populations.
The similarities and differences that are found in the various fossils described here, to extant Welwitschia mirabilis, raise questions of their relationships that are reflected by our choice of taxonomic assignment. This question is compounded by the nature of our paleobotanical data. That is, we are working with isolated organs such as seedlings at the cotyledon stage, leaves, and pollen cones. We may think that they grew together as one plant, but we cannot demonstrate this because they are not in organic connection. Because of this, it is necessary to assign to them an individual generic name for each isolated, unconnected organ. Also, in spite of many similarities, it is better not to assign these organs, at this time, to the modern W. mirabilis but only to the extant family.
The fossils presented in this paper, the fossil reported by Rydin et al. (2003)
, and earlier pollen records of Lima (1978a
, b
, c
, 1979
, 1980
, 1981
, 1982
) firmly establish the presence of Welwitschia-related plants in the Crato Formation of Brazil. Overall cotyledon form, size, basal insertion, and venation of Welwitschiella austroamericana are similar to cotyledons of Welwitschia mirabilis (Sykes, 1910
, 1911
). Similar leaf characters in Welwitschiophyllum brasiliense are found in Welwitschia (Kubitzki, 1990
). Cone characters such as dichasial organization, size, form, position on slender axes, and decussate arrangement of scales of Welwitschiostrobus murili are similar to the pollen cone characters of Welwitschia mirabilis (Pearson, 1929
; Leuenberger, 2001
). Cones similar to W. murili were noted by Martill et al. (1993)
from the Crato Formation. Because of all the characters of the fossils presented here, they are placed in the family Welwitschiaceae. The similarities and differences of the specific characters detailed in this report illustrate the extent to which these characters were already found in the ancestral stock of living Welwitschia.
The presence of Welwitschiaceae in the Lower Cretaceous of Brazil and in the modern flora of Africa provides support for floral exchange between Africa and South America during the Mesozoic (Raven and Axelrod, 1974
; Coetzee, 1993
). Other fossil data suggest it continued even through the Paleogene (Dilcher, 2000
; Jaramillo and Dilcher, 2001
). The presence of the fossils presented here in the Lower Cretaceous of Brazil is important in the evolution and dispersal of plants in the Mesozoic. This demonstrates clearly an exchange of biota during the Mesozoic between Africa and South America. A similar link has been established with the presence of dinosaurs that also link South America, Africa, and Asia during the Lower Cretaceous (Sereno et al., 2004
). This early exchange of gymnosperms must also have been taking place in a variety of angiosperms. It has been suggested that a Cretaceous biotic exchange may have been important in the dispersal of common tropical angiosperm elements between southeastern Asia, Africa, South America, and southeastern North America (Dilcher, 2000
).
Other megafossils with Welwitschia-like characters are recorded in both Mesozoic-age land masses of Gondwana and Laurasia (Fig. 19). They include a Cenomanian leaf from the Czech Republic, Conospermites hakeaefolius (Velenovsk
and Viniklá
, 1926
; Crane, 1988
), and Decheyllia gormani from the Upper Triassic of Arizona (Ash, 1972
; Crane, 1988
). Although pollen associated with D. gormani and Masculostrobus clathratus resembles pollen of both Ephedra and Welwitschia, an affinity with Gnetales is uncertain due to characters of the pollen, winged seeds, and "cones" of leaflike microsporophylls (Crane, 1988
; Zavada, 1990
; van Konijnenburg-van Cittert, 1992
; Osborn et al., 1993
). Duan (1998)
described Chaoyangia as an angiosperm fruiting axis from northeastern China, while Sun et al. (1998)
suggested that these "fruits" are the same as those that Krassilov (1986)
recognized as Gurvanella and considered them a protoangiosperm from Russia. Sun et al. (1998)
pointed out that Gurvanella, with its winged seeds, is similar to those of Welwitschia and suggested that Welwitschia-like plants appear to have extended to China and Russia during the Mesozoic. Some have suggested that the presence of winged seeds indicates a closer link between Ephedra and Welwitschia (Zhou et al., 2003
), while phylogenetic analyses based on morphology and molecular data suggest a closer link between Gnetum and Welwitschia (Crane, 1985
; Doyle and Donoghue, 1986
; Doyle, 1996
; Bowe et al., 2000
; Chaw et al., 2000
; Gugerli et al., 2001
; Rydin et al., 2002
). Table 2 lists the various reports of possible Welwitschia-like megafossils from the Mesozoic. However, the affinities of the Triassic fossils are in dispute.
|
; Table 1). Certainly, the widespread record of pollen ascribed to Welwitschia during the Mesozoic in both land-masses of Gondwana and Laurasia should make it clear that the Mesozoic distribution of Welwitschia-like plants must have been widespread. Welwitschia-like pollen was found with Drewria potomacensis (Crane and Upchurch, 1987
). Welwitschia-like pollen, Welwitschiapites has been reported from the Lower Cretaceous of the Caucasus, Crimea, Austria, Hungary, Canada, and Brazil (Müller, 1966
; Herngreen and Chlonova, 1981
), including the Crato Formation of Brazil (Lima, 1978c
; Osborn et al., 1993
). The affinity of the type species of fossil pollen (Welwitschiapites magniolobatus Bolkh. ex Potonié) to Welwitschia remains uncertain (Bolchovitina, 1953
; Kremp et al., 1959
; Krutzsch, 1961
; Jansonius and Hills, 1976
). Ephedripites, Equisetosporites, and Gnetaceaepollenites, which are closely related to Ephedra and Welwitschia (Lima, 1978a
; Trevisan, 1980
; Crane and Ligdard, 1990
; Osborn et al., 1993
), have been reported from the Lower Cretaceous Crato (Santana) Formation (Table 1). Based upon such widespread reports of the pollen, we suggest this is further evidence that Welwitschia and Welwitschia-like plants were much more widespread throughout the world than previously thought.
The presence of Welwitschiaceae in the Lower Cretaceous of Brazil may also provide insights into climatic requirements for modern Welwitschia (arid to semiarid) based upon the modern distribution of the genus, the nature of the associated fossils and the sediments, and Welwitschia-like fossils (suggested arid to semiarid to mesic). Welwitschia is capable of growing in more mesic climates and does so readily in greenhouse settings (von Willert et al., 1992
; Jacobson and Lester, 2003
). This suggests, along with possible aquatic plants in the Crato Formation (Duarte, 1985
; Mohr and Friis, 2000
), that regional climates were warm, but not arid (Doyle et al., 1982
). Welwitschia-like plants eventually reached the savanna-woodland habitat of Africa (Axelrod and Raven, 1978
) due to a widespread connection in the Cretaceous of South America and Africa (Raven and Axelrod, 1974
; Sereno et al., 2004
; Smith et al., 2004
). Desertification and isolation during the Tertiary and Quaternary limited the distribution of other less drought-adapted plants (Axelrod and Raven, 1978
), resulting in the isolation and endemic distribution of extant Welwitschia in southwestern Africa today (Jacobson and Lester, 2003
).
|
|
1 The help of Dr. Joseph E. Armstrong for access to living seedlings of Welwitschia and Dr. Volker Wilde (Senckenberg Museum of Natural History) and the Paris Museum of Natural History (MNHN) for access to fossil material used in this study was essential and very much appreciated. The authors thank two anonymous reviewers, James Doyle and Alan Graham, for their comments. Support was provided in part by the University of Florida, Dilcher Research Funds, the Becker-Dilcher Paleobotany Research Fund, and the Fundacão de Amparo à Pesquisa do Estado de São Paulo, FAPESP 03/09407-4. This paper is the University of Florida Contribution to Paleobiology publication no. 534. 
5 Author for correspondence (e-mail: dilcher{at}flmnh.ufl.edu
)
LITERATURE CITED
Arai M. J. C. Coimbra A. C. Silva-Telles Jr 1999 Biostratigraphical synthesis of the Araripe Basin, northeastern Brazil. Anais da Academia Brasileira de Ciências 71: 816 (Abstract)
Arai M. J. C. Coimbra A. C. Silva-Telles Jr 2001 Síntese bioestratigráfica da Bacia do Araripe (nordeste do Brasil). In L. M. Barros, P. C. Nuvens, and J. B. M. Filgueira [eds.], Atas do II simpósios sobre a Bacia do Araripe e bacias interiores do nordeste, 109–117; 122–124. Comunicações, Universidade Regional do Cariri (URCA), Crato, Ceará Province, Brazil (Coleção Chapada do Araripe no 1)
Arai M. A. T. Hashimoto N. Uesugui 1989 Significado cronoestratigráfico da associação microflorística do Cretáceo inferior do Brasil. Boletim de Geociências da Petrobrás 3: 87-103
Ash S. R. 1972 Late Triassic plants from the Chinle Formation in northeastern Arizona. Palaeontology 15: 598-618
Axelrod D. I. P. H. Raven 1978 Late Cretaceous and Tertiary vegetation history of Africa. In M. J. A. Werger [ed.], Biogeography and ecology of Southern Africa, 77–130. W. Junk, The Hague, Netherlands
Barreto A. M. F. M. E. C. Bernardes-De-Oliveira D. L. Dilcher A. F. Mandarim-De-Lacerda M. S. S. Viana 2000 Early Cretaceous monocarpelar fruit of the Crato Member, Santana Formation, Araripe Basin, northeastern Brazil. Revista Universidade Guarulhos, Geociências V 121-124
Bechly G. 1998 New fossil dragonflies from the Lower Cretaceous Crato Formation of north-east Brazil (Insecta: Odonata). Stuttgarter Beiträge zur Naturkunde Serie B 264: 1-66
Belsky C. Y. S. Jardiné B. Prestat P. Durif E. Grosdidier J. P. Cassan M. Gillmann 1975 Etude stratigraphique et sédimentologique d'une coupe du Crétacé-Tertiaire au large du Suriname. Proceedings of the Ninth Inter-Guyana Geological Conference. Boletin de Geologia, Venezuela 179–187
Bernardes-De-Oliveira M. E. C. D. L. Dilcher A. F. Mandarim-De-Lacerda D. Pons 1999 Registro Aptiano-Albiano de Welwitschiaceae na Formação Santana, Bacia do Araripe, nordeste do Brasil. XVI Congresso Brasileiro de Paleontologia. Sociedade Brasileira de Patologia/Universidade Regional do Cariri (SBP/URCA), Crato, Ceará Province, Brazil. Boletim de resumos, 26
Bernardes-De-Oliveira M. E. C. D. L. Dilcher A. F. Mandarim-De-Lacerda D. Pons F. Ricardi-Branco 2000 Gnetalean macrofossils of the Crato Member, Santana Formation, Late Aptian–Early Albian, Chapada do Araripe, Brazil. Thirty-first session of the International Geological Congress, 6–17 August, 2000, Booth E 43-CD. Geological Survey of Brazil, Rio de Janeiro, Brazil
Bernardes-De-Oliveira M. E. C. M. R. Lima D. Pons 1993 Folhas de Araucariaceae da Formação Santana, Cretáceo do nordeste Brasileiro. Anais da Academia Brasileira de Ciências, Resumos de Comunicações 65: 329-330
Berry E. W. 1911 Systematic paleontology of the Lower Cretaceous deposits of Maryland, Angiospermae. In W. B. Clark [ed.], Lower Cretaceous, 214–508. Maryland Geological Survey, Baltimore, Maryland, USA
Berthou P. Y. 1990 Le bassin d'Araripe et les petits bassins intracontinentaux voisins (N.E. du Brésil): formation et évolution dans le cadre de l'ouverture de l'Atlantique équatorial: comparaison avec les bassins ouest-africains situés dans le même contexte. In De A. Campos, M. S. S. Viana, P. M. Brito, and G. Beurlen [eds.], Atas do I Simpósio sobre a Bacia do Araripe e bacias interiores do nordeste, Crato, 1990, 113– 134. Departamento Nacional da Produção Mineral (DNPM), Crato (CE), Brazil
Bolchovitina N. A. 1953 Spore and pollen characteristics of Cretaceous deposits in the central areas of the U.S.S.R. Trudy Instituta Geologii Nauk S.S.S.R., Geologii Series 145: 1-184
Bowe L. M. G. Coat C. W. dePamphilis 2000 Phylogeny of seed plants based on all three genomic compartments: extant gymnosperms are monophyletic and Gnetales' closest relatives are conifers. Proceedings of the National Academy of Sciences, USA 97: 4092-4097
Brenner G. J. 1968 Middle Cretaceous spores and pollen from northeastern Peru. Pollen et Spores 10: 341-383
Brenner G. J. 1976 Middle Cretaceous floral provinces and early migration of angiosperms. In C. B. Beck [ed.], Origin and early evolution of angiosperms, 23–47. Columbia University Press, New York, New York, USA
Brito-Neves B. B. 1990 A Bacia do Araripe no contexto geotectônico regional. In D. De A. Campos, M. S. S. Viana, P. M. Brito, and G. Beurlen [eds.], Atas do I Simpósio sobre a Bacia do Araripe e bacias interiores do nordeste, Crato, June 1990, 21–33. Departamento Nacional da Produção Mineral (DNPM), Crato (CE), Brazil
Chaw S.-M. C. L. Parkinson Y. C. Cheng T. M. Vincent J. D. Palmer 2000 Seed plant phylogeny inferred from all three plant genomes: monophyly of extant gymnosperms and origin of Gnetales from conifers. Proceedings of the National Academy of Sciences, USA 97: 4086-4091
Coetzee J. A. 1993 African flora since the Terminal Jurassic. In P. Goldblatt [ed.], Biological relationships between Africa and South America, 37– 61. Yale University Press, New Haven, Connecticut, USA
Coimbra J. C. M. Arai A. L. Carreño 2002 Biostratigraphy of Lower Cretaceous microfossils from the Araripe Basin, northeastern Brazil. Géobios 35: 687-698[CrossRef][ISI]
Crane P. R. 1985 Phylogenetic analysis of seed plants and the origin of angiosperms. Annals of the Missouri Botanical Garden 72: 716-793[CrossRef][ISI]
Crane P. R. 1988 Major clades and relationships in the "higher" gymnosperms. In C. B. Beck [ed.], Origin and evolution of gymnosperms, 218– 272. Columbia University Press, New York, New York, USA
Crane P. R. 1996 The fossil history of the Gnetales. International Journal of Plant Sciences 157: S50-S57[CrossRef]
Crane P. R. C. D. Hult 1988 Welwitschia the wonderful: life as a survivor in the desert of southwestern Africa. Field Museum of Natural History Bulletin 59: 22-29
Crane P. R. S. Lidgard 1990 Angiosperm radiation and patterns of Cretaceous palynological diversity. In P. D. Taylor and G. P. Larwood [ed.], Major evolutionary radiations, 377–407. Clarendon Press, Oxford, UK
Crane P. R. J. G. Maisey 1991 Fossil plants. In J. G. Maisey [ed.], Santana fossils: an illustrated atlas. T. F. H., Neptune City, New Jersey, USA
Crane P. R. G. R. Upchurch 1987 Drewria potomacensis gen. et sp. nov., an Early Cretaceous member of Gnetales from the Potomac Group of Virginia. American Journal of Botany 74: 1722-1736
Dilcher D. L. 2000 Geological history of the vegetation in southeast United States. Sida, Botanical Miscellany 18: 1-21
Dilcher D. L. A. F. Mandarim-De-Lacerda A. M. F. Barreto M. E. C. Bernardes-De-Oliveira 2000a Selected fossils from the Santana Formation, Chapada do Araripe, Brazil. Geociências, (São Paulo) 5: 249 (Abstract)
Dilcher D. L. A. F. Mandarim-De-Lacerda A. M. F. Barreto M. E. C. Bernardes-De-Oliveira 2000b Selected fossils from the Santana Formation, Chapada do Araripe, Brazil. American Journal of Botany 87: 67
Dino R. 1992 Palinologia, bioestratigrafia e paleoecologia da Formação Alagamar—Cretáceo da Bacia Potiguar, nordeste do Brasil. Ph.D. dissertation, Universidade de São Paulo, Brazil
Dino R. 1994 Algunas espécies novas de grãos de pollen do Cretáceo inferior do nordeste do Brasil. Boletim de Geociências da Petrobrás 8: 79-98
Dino R. D. T. Pocknall M. E. Dettmann 1999 Morphology and ultrastructure of elater-bearing pollen from the Albian to Cenomanian of Brazil and Ecuador: implications for botanical affinity. Review of Palaeobotany and Palynology 105: 201-235[CrossRef]
Doyle J. A. 1977 Patterns of evolution in early angiosperms. In A. Hallam [ed.], Patterns of evolution, 501–546. Elsevier, Amsterdam, Netherlands
Doyle J. A. 1996 Seed plant phylogeny and the relationships of Gnetales. International Journal of Plant Sciences 157: S3-S39[CrossRef]
Doyle J. A. M. J. Donoghue 1986 Seed plant phylogeny and the origin of angiosperms: an experimental cladistic approach. Botanical Review 52: 321-431
Doyle J. A. S. Jardiné A. Doerenkamp 1982 Afropollis, a new genus of early angiosperm pollen, with notes on the Cretaceous palynostratigraphy of paleoenvironments of northern Gondwana. Bulletin des Centres de Recherches Exploration-Production Elf-Aquitaine 6: 39-117
Duan S. Y. 1998 The oldest angiosperm—a tricarpous female reproductive fossil from Western Liaoning Province, NE China. Science in China (Series D) 41: 14-20
Duarte L. 1985 Vegetais fósseis da Chapada do Araripe, Br. Seção Paleontologia e Estratigrafia 2: 557-563
Endress P. K. 1996 Structure and function of female and bisexual organ complexes in Gnetales. International Journal of Plant Sciences 157: S113-S125[CrossRef]
Gothan W. 1914 Die unterliassische (rhätische) Flora der Umgegend von Nürnberg. Abhandlungen Naturhistorische Gesellschaft Nürnberg 19: 91-186
Gugerli F. C. Sperisen U. Buchler I. Brunner S. Brodbeck J. D. Palmer Y.-L. Qiu 2001 The evolutionary split of Pinaceae from other conifers: evidence from an intron loss and a multigene phylogeny. Molecular Phylogenetics and Evolution 21: 167-175[CrossRef][ISI][Medline]
Hashimoto A. T. C. J. Appi A. N. Soldan J. R. Cerqueira 1987 O Neo-Alagoas nas bacias do Ceará, Araripe, Potiguar (Brasil): caracterização estratigráfica e paleoambiental. Revista Brasileira Geociências 17: 118-122
Herngreen G. F. W. 1973 Palynology of Albian-Cenomanian strata of borehole 1-QS-1-MA, state of Maranhão, Brazil. Pollen et Spores 15: 515-545
Herngreen G. F. W. 1974 Middle Cretaceous palynomorphs from northeastern Brazil. Results of a palynological study of some boreholes and comparison with Africa and the Middle East. Sciences Géologiques, Bulletin 27: 101-116
Herngreen G. F. W. 1975 Palynology of Middle and Upper Cretaceous strata in Brazil. Mededelingen Rijks Geologische Dienst, N.S 26: 39-91
Herngreen G. F. W. 1981 Microfloral relationships between Africa and South America in Middle and Upper Cretaceous time. IV International Palynological Conference, vol. 3, 406–417. Birbal Sahni Institute of Palaeobotany, Lucknow, India
Herngreen G. F. W. A. F. Chlonova 1981 Cretaceous microfloral provinces. Pollen et Spores 23: 441-555
Herngreen G. F. W. H. Duenas Jimenez 1990 Dating of the Cretaceous Une Formation, Colombia and the relationship with the Albian-Cenomanian African-South American microfloral province. Review of Palaeobotany and Palynology 66: 345-359[CrossRef]
Herngreen G. F. W. M. Kedves L. V. Rovnina S. B. Smirnova 1996 Cretaceous palynofloral provinces: a review. In J. Jansonius and D. C. McGregor [eds.], Palynology: principles and applications, vol. 3, 1157–1188. American Association of Stratigraphic Palynologists Foundation, Dallas, Texas, USA
Hooker J. D. 1863 On Welwitschia a new genus of Gnetaceae. Transactions of the Linnean Society of London 24: 1-48
Jacobson K. M. E. Lester 2003 A first assessment of genetic variation in Welwitschia mirabilis Hook. Journal of Heredity 94: 212-217
Jansonius J. 1962 Palynology of Permian and Triassic sediments, Peace River area, western Canada. Palaeontographica Abteilung B 110: 35-98
Jansonius J. L. V. Hills 1961, 1976, 1979 Genera file of fossil spores and pollen. Special publication, Department of Geology, University of Calgary, Calgary, Alberta, Canada
Jaramillo C. A. D. L. Dilcher 2001 Middle Paleogene palynology of Central Colombia, South America: a study of pollen and spores from tropical latitudes. Palaeontographica Abteilung B 258: 87-213
Kers L. E. 1967 The distribution of Welwitschia mirabilis Hook. Svensk Botanisk Tidskrift 61: 97-125
Krassilov V. A. 1982 Early Cretaceous flora of Mongolia. Palaeontographica Abteilung B 181: 1-43
Krassilov V. A. 1984 New paleobotanical data on origin and early evolution of angiospermy. Annals of the Missouri Botanical Garden 71: 577-592[CrossRef][ISI]
Krassilov V. A. 1986 New floral structure from the Lower Cretaceous of Lake Baikal area. Review of Palaeobotany and Palynology 47: 9-16[CrossRef][ISI]
Krassilov V. A. 1997 Angiosperm origins: morphological and ecological aspects. Pensoft, Sofia, Bulgaria
Krassilov V. A. E. V. Bugdaeva 1988 Gnetalean plants from the Jurassic of Ust-Balej, East Siberia. Review of Paleobotany and Palynology 53: 359-374[CrossRef]
Kremp G. O. W. H. T. Ames A. J. Kovar 1959 Catalog of fossil spores and pollen. Pennsylvania State University, University Park, Pennsylvania, USA
Krutzsch W. 1961 Über Funde von "ephedroiden" Pollen im deutschen Tertiär. Geologie 10: 15-53
Kubitzki K. 1990 Gnetatae—with the single order Gnetales. In K. U. Kramer and P. S. Green [eds.], The families and genera of vascular plants, 378–391. Springer, Berlin, Germany
Kuvaeva S. B. 1972 Spores of a new genus Corniculatisporites from Cretaceous deposits of the Crimea and the Caucasus. Journal of Palynology 7: 20-25
Leuenberger B. E. 2001 Welwitschia mirabilis (Welwitschiaceae), male cone characters and a new subspecies. Willdenowia 31: 357-381
Lima M. R. 1978a Palinologia da Formação Santana (Cretáceo do nordeste do Brasil). Ph.D. dissertation, Universidade de São Paulo, São Paulo, Brazil
Lima M. R. 1978b Microfósseis da Formação Exu, Cretáceo do nordeste do brasil. Anais do 30° Congresso Brasileiro de Geologia Recife 2: 965-969
