| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Paleobotany |
2Department of Biology, University of Nebraska-Omaha, Omaha, Nebraska 68182 USA; 3Florida Museum of Natural History, University of Florida, Gainesville, Florida 32611 USA; 4Department of Botany, University of Florida, Gainesville, Florida 32611 USA
Received for publication November 14, 2002. Accepted for publication April 11, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSSYSTEMATICSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Eocene • flower • Green River Formation • inflorescence • infructescence • Pseudosalix handleyi • Salicaceae • Utah
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSYSTEMATICSDISCUSSIONLITERATURE CITED |
|---|
The family Salicaceae, formerly circumscribed narrowly to include only Populus and Salix, was recently emended to include most of the noncyanogenic genera formerly placed in Flacourtiaceae (Chase et al., 2002
). The revised classification is necessary because rbcL and morphological characters indicate that traditional Flacourtiaceae are rendered paraphyletic if Salix and Populus are treated as a distinct family (Judd, 1997
; Chase et al., 2002
; Judd et al., 2002
). Many earlier systematists also noted the similarity between Populus and Salix and various genera formerly attributed to Flacourtiaceae, especially the genera Itoa, Poliothyrsis, Carrierea, Idesia, Bennettiodendron, and other members of the tribe Flacourtieae (Holm, 1969
; Dahl, 1972
; Hegnauer, 1973
; Keating, 1973
; Meeuse, 1975
; Miller, 1975
; see also discussion in Judd [1997]
). In this article, we introduce the new genus Pseudosalix and interpret its phylogenetic position to be within the paraphyletic tribe Flacourtieae, as the immediate sister group to the clade containing extant Salix and Populus (tribe Saliceae of Chase et al. [2002]
).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSYSTEMATICSDISCUSSIONLITERATURE CITED |
|---|
The fossils were collected from lacustrine shales of the Parachute Creek Member of the Green River Formation in the vicinity of Bonanza, Utah, USA. The very fossiliferous horizon, from which these specimens were collected, has been traced laterally over several kilometers and is situated about 129 m above the Mahogany marker bed (Bruce Handley, personal communication) and includes UCMP site PB02016 and UF sites 15753, 15884. The fossiliferous layers correspond approximately to the level of the Wavy tuff, recently dated by 40Ar/39Ar laser fusion of biotite and hornblende crystals at 48.13 ± 0.71 and 48.22 ± 0.71 million years (Machlus et al., 2002
), and are considered early Middle Eocene.
The flora of the Green River Formation in eastern Utah and adjacent western Colorado was monographed by Knowlton (1923)
, Brown (1929
, 1934
), and MacGinitie (1969)
, but new discoveries and continuing comparative work have led to improved understanding of the floristic composition. Also common at this horizon of the Parachute Creek Member are leaves of Macginitiea (Platanaceae; Manchester, 1986
), leaves and fruits of Cedrelospermum (Ulmaceae; Manchester, 1989
), leaves and fruits of Populus (Salicaceae; Manchester et al., 1986
), and foliage of Parvileguminophyllum (Leguminosae; Call and Dilcher, 1994
).
Fine details of leaf margins, styles, and sepals were exposed by removal of thin layers of sediment with a needle under the dissecting microscope. Exposure of the stipule scars required the brushing away of carbonaceous matter from the impression surface. Cellular and structural details were further examined by preparing silicone casts of the flowers, which were subsequently observed with a Philips Model 515 (Philips Electron Optics, Hillsboro, Oregon, USA) scanning electron microscope (SEM).
The initial keying to family was performed using computerized software including MEKA version 3.0 for Windows using the "angiofam" database (Duncan and Meacham, 1996
) and INTKEY version 5.0 (Watson and Dallwitz, 2000
). The terminology used in description of the fossils follows that of the Leaf Architecture Working Group (1999)
for leaf architecture and Judd et al. (2002)
for morphology of inflorescences, flowers, and fruits. Herbarium specimens from Harvard University (Gray Herbarium and Arnold Arboretum) and University of Florida (FLAS) were consulted in the comparative analyses.
|
SYSTEMATICS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSYSTEMATICSDISCUSSIONLITERATURE CITED |
|---|
|
Type species
Pseudosalix handleyi gen. et sp. nov.
Specific diagnosis
As for genus.
Holotype
UMNH PB-1 (Figs. 1–8, 16, 18–20).
|
Other specimens
UCMP 390084, 390085, UF 15753-22965 (leaves); DMNH 24354 (locality DMNH 323) (detached pistillate flower); UF15753-22931, UF 15753-34806, UF 15753-34807, UF 15884-8925, USNM 38625, USNM 38628, UCMP 390081 (locality UCMP PB02016), DMNH 23391 (locality DMNH 2347), DMNH 23392 (locality DMNH 316), DMNH 24355, 24454 (locality DMNH 323) (staminate flowers).
Etymology
Pseudo, meaning "false," and Salix, emphasizing that it is not the genus Salix, although the leaves are similar. The specific epithet recognizes the contributions of Bruce Handley, whose magnificent collection of Green River fossil plants, now housed at UCMP, facilitated the completion of this study.
Description
Inflorescence terminal, 7.9–8.2 cm long and 2.0–2.7 cm wide, compound, panicle-like, with an apparently determinate main axis and cymose side branches (Figs. 1–2). Buds and flowers pedicellate, pedicels 2.5–4 mm long with bract scars at their junction with peduncle (Fig. 3). Flower buds ovoid, smooth, about 3 mm long and 2 mm wide with valvate sepals. Flowers unisexual, apetalous. Pistillate flowers actinomorphic with hypogynous calyx of four prominent, basally connate sepals; staminodes lacking (Figs. 3–4 and 24). The single pistil of each flower consisting of a smooth, ovoid ovary about 3 mm long and 2 mm wide, rounded basally and apically, with at least three (perhaps four in some specimens) recurved style arms, 0.9 mm long, radiating from the apex and terminated by globose stigmas (Figs. 5–8). Three to four longitudinal grooves on the ovary wall (Figs. 7–8). Attached opened flowers 7 mm in diameter, detached flowers up to 11 mm in diameter (Fig. 28). Fruits ovoid capsules 5.5–7.5 mm long and 3.0–4.5 mm wide with 3–4 valves, splitting longitudinally (Figs. 12–15).
|
|
Leaves simple, attached alternately along the axis with an internode width of 1.0 cm (Figs. 1 and 9). Laminae generally symmetrical and slender, ovate to oblong, up to 8.3 cm long and 1.0 cm wide; length to width ratio is 6 : 1 to 15 : 1. Petioles up to 1.4 cm long and about 1 mm wide with marginal attachment and not significantly swollen at the base. Stipule scars present on the twig immediately lateral to the attachment of the petiole (Fig. 19). Lamina base and apex both acute; base decurrent to cuneate; apex shape acuminate to straight. Margin serrate with circular glands on the teeth (Figs. 16–18). Venation is pinnate with a moderately thick midvein. Secondary venation semicraspedodromous with irregular spacing, uniform angles of 50–60 degrees from the midvein, and with frequent weak intersecondary veins (Figs. 16, 21). Position of intersecondaries somewhat variable, but often central between adjacent secondary veins. Secondary veins usually branch before reaching the margin, sending one branch to the tooth and the other to loop with the supra-adjacent secondary (Figs. 21–23). Each tooth usually enervated by three veins: the principal vein of the tooth, often extending directly from the secondary vein and entering the tooth medially and a pair of adjacent, slightly thinner accessory veins on each side that originate from veins above and below (Fig. 20). Teeth somewhat irregularly spaced, typically six per centimeter. Space between secondary veins approximately equivalent to the distance between successive teeth along the margin. Teeth simple; each one less than 1 mm from base to apex; however, there appear to be two orders of teeth near the middle of the lamina, arranged by alternating size along the margin (Fig. 16). Tooth shape concave on the apical side of the tooth and convex to straight on the basal side, with a rounded sinus. The tooth apex ends in a circular gland (Figs. 18, 20–23). Tertiary venation random reticulate to possibly alternate percurrent, with the vein angle variably obtuse to the primary vein. Highest order of veins appears to be dichotomizing quaternary veins, but not readily apparent. Areoles not preserved.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSYSTEMATICSDISCUSSIONLITERATURE CITED |
|---|
Leaves
The spherulate gland at the tooth apex is most similar to what has been termed a salicoid tooth (Hickey and Wolfe, 1975
; Judd et al., 2002
). There is no indication that it was deciduous, as the setae of theoid teeth. The tooth apex with its gland is enervated by a medial principal vein (secondary or intersecondary) and usually a pair of thinner accessory veins, one along the apical margin of the tooth, the other along the basal flank (Fig. 20). Fine venation in the leaf appears to be poorly developed, but this may be due to poor preservation of the higher orders of veins. Based on tertiary venation and the relatively poor petiole-blade separation, the leaf is considered low ranking, i.e., is not highly organized (Leaf Architecture Working Group, 1999
). Although stipules are not directly observed on the fossil twigs, a pair of scars immediately below the petiole attachment (Fig. 19) are positioned precisely where the stipules would be expected, indicating that stipules were likely present, but deciduous.
Among leaves previously described from the Green River flora, the leaf of Pseudosalix handleyi is most similar to Salix cockerelli Brown and S. longiacuminata Brown (Brown, 1934
; USNM 38594–38596, 168963, 38587, 38588). However, in those syntypes of S. cockerelli with the base preserved, the base is obtuse and more rounded than in Pseudosalix handleyi. In addition, there are notable differences in the venation of S. cockerelli. The secondary veins and intersecondaries are both more prominent and organized in S. cockerelli than in P. handleyi leaves. Further, the spacing and looping of the secondaries are better defined in S. cockerelli and the tertiary venation is mixed alternate-opposite percurrent. In S. longiacuminata, the secondary veins are more prominently looped than in P. handleyi, and the angle of the secondary veins increases significantly towards the base. However, it is possible the mature vegetative leaves of Pseudosalix were larger and perhaps varied in their leaf architecture. Other leaves comparable to Pseudosalix, but not previously described from the Green River, are listed above under "Other specimens."
Flowers and fruits
The inflorescence of Pseudosalix differs from that of Salix in its broader ovaries, well-developed calyx, and wider spacing of the flowers. It differs from both Salix and Populus in having a paniculoid cyme with lateral paraclades of pedicellate flowers, as in Idesia, Bennettiodendron, and Poliothyrsis, rather than the racemose inflorescence of Salix and Populus. In addition, the branch bearing the terminal inflorescence is fully developed, bearing large leaves, and not reduced as in Salix and Populus.
By studying the surface impression of the same flowers from both counterparts, it was possible to count the number of valves in ovaries of pistillate flowers and in the capsules of mature fruits. This ranged from three to four (Figs. 7–8 and 12–15). The number of styles per ovary was not immediately apparent in the natural fracture of the shale, although two styles could easily be seen in some of the flowers (Figs. 5 and 8). However, it was possible to expose additional styles that had been protruding into the shale (Figs. 6–7), indicating that there were at least three per ovary. The styles appear to be aligned with the center of each valve of the developing fruit, so it is likely that the number of styles varied from three to four as did the number of fruit valves.
The holotype specimen shows the pistillate inflorescence with both closed buds and opened flowers nearing anthesis (Figs. 2–8). The buds, which are protected by the fully closed valvate sepals, are easily distinguished from mature fruits because there is no scar of shed calyx on the pedicel and by their relatively small size. Isolated pistillate flowers show the same arrangement of sepals and a three or four valved gynoecium, but differ from the attached flowers by somewhat larger size, the presence of more obvious longitudinal venation within the sepals, and by the orientation of the style arms. Whereas the style arms in the attached flowers are distally recurved, those in the detached specimen are erect. It is likely that these detached flowers represent a more mature stage of development. This kind of isolated flower was described by Brown (1929)
as Antholithes polemonioides.
Associated, but unattached, isolated staminate flowers occur in the same horizon as the twigs described above. Previously illustrated as Potentilla (?) byrami Cockerell (1925
, Pl. 2, Fig. 9) and "species of Antholithes" Brown (1934
, Pl. 15, Figs. 3 and 6), they are similar to the pistillate flowers in being pedicellate and typically four-sepalled, with longitudinally aligned isodiametric epidermal cells. They show 3–5 longitudinal veins in each sepal, as is also observed in the detached pistillate flowers mentioned above. They bear up to about 30 stamens, each with a short (1.5–2 mm) filament, and globose anther. Attempts to process pollen from stamens of these flowers were unsuccessful. Globose anthers and short filaments also characterize Populus and Salix, but their flowers do not have such a well-developed calyx (Populus) or entirely lack a calyx (Salix). However, laboratory-generated anther-derived plants of Populus maximowiczii bore unusual single terminal flowers that possessed a "cup-shaped, calyx-like structure with 3–4 lobes from which the stamens emerged" (Stoehr et al., 1988
). These male flowers had 14–18 stamens per flower with globose anthers on short filaments. These abnormal Populus flowers, although smaller (about 3 mm diameter) than those associated with Pseudosalix, closely resemble the fossil flowers in venation and shape of the calyx lobes as well as androecial morphology.
Familial assignment
Salient features of the fossil, which we used in our search of extant families via the MEKA 3.0 angiofam key (Duncan and Meacham, 1996
), included: leaves alternate, simple, pinnately veined, serrate margined, stipules present, flowers unisexual and actinomorphic with four sepals, 3–4 carpels, capsular fruits, inflorescence compound, superior ovary (receptacle small), nectariferous disk absent, and multiple styles. The MEKA search indicated that the following eight families share the above set of characters: Dilleniaceae, Cunoniaceae, Flacourtiaceae, Malvaceae (cited as both Sterculiaceae and Malvaceae), Euphorbiaceae, Sapindaceae, Rubiaceae, and Caryophyllaceae. We note that Salicaceae was not included in the results because the database uses the older concept of this family, and only Salix and Populus characters were included. This family was excluded by the database based on the scoring of four sepals. However, Flacourtiaceae, a large number of which are now included in Salicaceae (Chase et al., 2002
; Judd et al., 2002
), are among the results. A review of each of the other conforming families indicates discrepancies in other characters that are inconsistent with the fossil, such as the lack of salicoid teeth in all but the Flacourtiaceae.
We also compared the fossil with the angiosperm family database of INTKEY 5.0 (Watson and Dallwitz, 2000
), using a different but somewhat overlapping set of characters from those used in MEKA: fertile gynoecium present, syncarpous, superior, fruit a capsule, calyx four-parted, flowers unisexual, small, actinomorphic, aggregated into inflorescences, leaves stipulate, well-developed, simple, alternate, spiral, petiolate, serrate, pinnately veined, medium-sized, lanceolate, nonsheathing, and nonsucculent. Fifteen families conformed to this set of search criteria: Actinidiaceae, Buddlejaceae, Capparaceae, Caryophyllaceae, Celastraceae, Cunoniaceae, Elaeocarpaceae, Flacourtiaceae, Gyrostemaceae, Hamamelidaceae, Molluginaceae, Salicaceae, Sapindaceae, Sterculiaceae, and Zygophyllaceae. Again, most of these families can be dismissed as being related to our fossil by various characters not included in the available database. Significantly, both Salicaceae sensu stricto (s.s.) and Flacourtiaceae are among the candidates from this search. Those families selected in both MEKA and INTKEY searches include the Cunoniaceae, Flacourtiaceae (i.e., Salicaceae s.l. and Achariaceae; Chase et al., 2002
; Judd et al., 2002
), Sterculiaceae, Sapindaceae, and Caryophyllaceae. The presence of circular glands on teeth of the leaf margin, i.e., salicoid teeth, distinguishes Salicaceae s.l. (and our fossil) from the other candidate families. These distinctive teeth are likely a synapomorphy of all members of Salicaceae s.l. except for the basal genus Casearia (Chase et al., 2002
; Judd et al., 2002
; see also Fig. 34), and their presence in Pseudosalix provides strong support for its placement in Salicaceae s.l. The presence of alternate, stipulate leaves is also consistent with a placement in this family.
|
). The capsular fruits with valvate dehiscence and bracts on the pedicels are similar to those observed in Populus. Most Flacourtieae have fleshy fruits, but capsular fruits (bearing wind-dispersed seeds) are a probable synapomorphy for the clade containing Poliothyrsis, Carrierea, Itoa, Pseudosalix, Salix, and Populus (Table 1; Fig. 34). The capsules of Poliothyrsis, Carrierea, and Itoa are large, and tend to open basally as well as apically, while those of Populus and Salix are smaller and open only apically, and these smaller capsules may be synapomorphic for the Populus + Salix clade. Since the capsules of Pseudosalix are similar to those of Populus, the fossil is likely a sister to the Populus + Salix clade (Fig. 34). The Saliceae, the tribe containing Populus and Salix, are nested within the cladistic structure of the Flacourtieae (see Fig. 34 and Chase et al. [2002]
), and the monophyly of Saliceae is supported by the additional apomorphies of racemose inflorescences (catkins) terminating reduced branches, flowers with a very reduced calyx, and plumose seeds (replacing the winged seeds characteristic of Itoa, Poliothyrsis, and Carrierea).
|
) together with Populus leaves and fruits, and it is possible that similar seeds were also produced by Pseudosalix. It is worth noting that surveys of large collections of Green River plant-bearing shales from several museum collections have failed to reveal any winged seeds matching those in other extant genera of capsule-fruited Salicaceae. Given the similarity of the fossil leaves to those of some extant species of Salix, it is worth pointing out the morphological differences that preclude placement of the fossil in that genus. Whereas Salix has narrow fruits of only two valves, those of the fossil are broader and have three or four valves (variable among fruits on the same infructescence). Whereas Salix infructescences are compact and racemose (without cymose branching), Pseudosalix infructescences are looser and branched. We note that some species of Salix sect. Longifoliae Pax have branched catkins, with secondary racemose catkins near the base of the primary catkin; this compound catkin is very different from, and not homologous with, the cymose branching inflorescence characteristic of Pseudosalix and members of the Flacourtieae. The calyx is well developed in the pistillate and staminate flowers of Pseudosalix, but is lacking in flowers of Salix. Salix typically has prominent subtending bracts to the inflorescence and associated nectar glands, which may be modified stipules. We have not observed these features in the fossil. Finally, it is important to note that a racemose inflorescence and reduced calyx are synapomorphic for the Salix + Populus clade and the lack of these features in Pseudosalix indicates that it cannot be placed within this monophyletic group. Thus, even if it were discovered that Pseudosalix had plumose seeds, it would still be unparsimonious to place the fossil within either Populus or Salix.
The elliptical elongate leaves with pinnate venation distinguish the fossil from many species of Populus that have broadly ovate laminae with actinodromous venation and long petioles; yet some living species of Populus, such as P. angustifolia James, have more slender leaves with pinnate venation and short petioles similar to those of Pseudosalix. There is also some indication from ontogeny that slender leaves may be plesiomorphic in Populus (Eckenwalder, 1980
, p. 367). The morphology of the capsules in Pseudosalix compares favorably with those of Populus. However, as discussed above, the fossil is distinguished from all modern species of Populus by the compound inflorescence terminating a well-developed leafy twig and a well-developed calyx in both pistillate and staminate flowers. The corresponding synapomorphies for the taxa of Saliceae are absent in Pseudosalix, indicating clearly that the fossil must be placed outside of the Populus + Salix clade. We consider that the fossil is best placed immediately adjacent to the Saliceae clade, forming a larger clade, defined by terminally opening capsules, comose seeds, and absence of staminodes in pistillate flowers (Fig. 34).
In summary, Pseudosalix has stipulate, alternate leaves with salicoid teeth, flowers with globose anthers, and gynoecia with parietal placentation (inferred by the close similarity of the capsule morphology of Pseudosalix to that of Salix and Populus), indicating with some confidence a taxonomic placement within the Salicaceae s.l. (Fig. 34). It would be useful to determine whether the fossil had centrifugal initiation of its stamens, as this is likely an additional synapomorphy of Salicaceae s.l. (Bernhard and Endress, 1999
). The likely dioecious condition, unisexual, and apetalous flowers of Pseudosalix, point to a placement within the tribe Flacourtieae. The capsular fruits of Pseudosalix indicate that this fossil belongs in a monophyletic group consisting of Poliothyrsis, Carrierea, Itoa, Populus, and Salix, all of which have capsules. Idesia and Bennettiodendron are morphologically similar, although with fleshy fruits, and are likely also related (Fig. 34). Pseudosalix may be phylogenetically adjacent to Itoa, Poliothyrsis, and Carrierea on one hand and Populus and Salix on the other. The fossil is most likely sister to Populus and Salix (i.e., the Saliceae, a tribe nested within the Flacourtieae; see Fig. 34) based on similarities in the structure of its capsules and lack of staminodes in the carpellate flowers, but cannot be placed within either Salix or Populus because it possesses a well-developed calyx and branched inflorescences. Both of these are plesiomorphic features, also seen, for example, in Poliothyrsis, Idesia, and Bennettiodendron. Taken together, the synapomorphic features evident in the fossil clearly support a placement, well nested within Salicaceae s.l.
Other fossil Salicaceae s.l
The fossil record of Salicaceae s.l. extends back to the late Paleocene based on reports of Populus foliage from sites in western North America (Wing, 1981
; Collinson, 1992
), but the first unequivocal records supported by both fruits and foliage are from the early Middle Eocene Green River Formation of Utah and Colorado. Populus wilmattae, typically with wide-ovate leaves, is one of the dominant elements of the Green River leaf flora and is also known from a twig with attached infructescence and leaves (Manchester et al., 1986
). A second morphotype, with linear leaves having strongly ascending basal secondary veins, Populus cinnamomoides, is also common in the Green River leaf flora and is thought to represent the foliage of young trees of Populus wilmattae (Eckenwalder, 1980
). Based on comparison with the heteromorphism observed in a similar modern relative Populus mexicana, Eckenwalder concluded that Populus wilmattae and P. cinnamomoides must be merged as one species and that the older name, P. cinnamomoides, would take priority.
Despite the reports of leaves identified to Salix from the Green River Formation (Brown, 1934
; MacGinitie, 1969
), it remains unknown whether the extant genus was really present by the Eocene. Neither the diagnostic racemose infructescences nor the bivalvate capsules characteristic of Salix have been recovered. It is likely that some of the Salix-like leaves represent Pseudosalix or other genera. For example, another specimen recovered from the Green River Formation of Colorado (Carpolithes serratifolius Brown, 1929
, p. 292, Pl. 76, Fig. 5) shows a fossil twig with elongate, Salix-like leaves and a terminal infructescence that is morphologically more similar to Populus (Collinson, 1992
, p. 164, Pl. 1 Fig. b).
MacGinitie (1937)
recognized Idesia based on leaves with well-preserved venation from the northern California Weaverville flora of probable early Miocene age (Barnett, 1989
). According to MacGinitie, every character of the fossil leaves of Idesia cordata MacGinitie is matched in the leaves of extant I. polycarpa Maxim. If correctly determined, the occurrence of Idesia in western North America indicates that this genus, now endemic to China, crossed between Asia and North America sometime during the Tertiary.
Iljinskaya and Chelebaeva (2002)
described a new genus, Utkholokia, based on leaf imprints from the Paleocene of western Kamchatka, which they think combine leaf characters of Populus with those of Idesia and Poliothyrsis. The leaves of Utkholokia ulevenensis (Cheleb.) Iljinskaja et Chelb. are ovate to wide-ovate and asymmetric. Several veins arise palmately from the base of the lamina, as in Idesia. The tertiary veins are mainly straight and perpendicular to the primary vein. Marginal teeth are small, often having attenuate tooth apices with the apical side of each tooth parallel to the basal side of the neighboring adjacent tooth, a feature these authors have observed in extant Idesia but not in Populus or Poliothyrsis. Iljinskaya and Chelebaeva (2002)
note that the fossils compared very favorably with leaves of extant Populus ciliata Wall ex Royle and the monotypic genera Idesia and Poliothyrsis. However, the leaves differ from those of all three modern genera in their prominent asymmetry. The reproductive structures of Utkholokia remain unknown.
Mai (1980)
recognized five species of Poliothyrsis based on lignitized seeds from the Tertiary of Europe and western Siberia. Poliothyrsis eurorimosa Mai, from the late Oligocene to middle Miocene of northwestern Saxony, Germany, was described both from fruits and seeds. The seeds are oval to oblong-oval, 0.7–1.5 mm long, 0.5–1.2 mm wide, and have a testa composed of a layer of thickened, radially arranged sclereids. The capsules are ovoid to oblong heart-shaped, 6–9 mm, and composed of three thin, woody-leathery valves with dehiscence slits that reach the lower one-third of the capsule and parietal placentae along the midline of each valve. The seeds are preserved without the encircling wing diagnostic of Poliothyrsis and are about half the size of seeds of the extant species. The younger species, P. lutetianoides (Szafer) Mai from the Pliocene of Kroscienko, Poland, and P. hercynica Mai from the Pliocene of Reuver, Netherlands, Berga, Germany, and Villafranca, d'Asti, Italy, are larger (1.9–2.5 mm long) and more similar in size to those of the extant species, P. sinensis.
The genus Saxifragispermum Reid and Chandler, based on permineralized, anatomically preserved fruits from the Eocene of England (Reid and Chandler, 1933
; Chandler, 1961
) and Oregon (Manchester, 1994
), has 3–4 carpellate capsules with parietal placentation that led to their assignment to Flacourtiaceae (Chandler, 1961
; Manchester, 1994
). As the Flacourtiaceae has been abandoned (Chase et al., 2002
), these characters now indicate that Saxifragispermum probably belongs in the Salicaceae s.l.
It is interesting that Pseudosalix appears at about the same time as the first unequivocal records of Populus. The extinct genus was sympatric with Populus wilmattae and P. cinnamomoides. Leaves have been identified to extant Salix from the Eocene of the Green River Formation (Brown, 1934
; MacGinitie, 1969
). We regard the assignment as possible, but uncertain, in view of the lack of Salix infructescences. Informed evaluation of whether these leaves represent an extinct taxon such as Pseudosalix or the extant genus requires more complete material. The main floral feature distinguishing Pseudosalix from Populus and Salix is the occurrence of well-developed sepals. Morphologically similar flowers of Idesia attract insects as pollinators. Pseudosalix may have been well suited for insect pollination and less adapted for wind pollination than extant Populus and Salix. Eckenwalder (1996)
noted that (except in some subtropical species) poplar species flower before leaf emergence in spring, as is common in wind-pollinated trees. Judging from the occurrence of mature flowers on a twig with expanded leaves, we infer that the timing of pollination in Pseudosalix was after the flush of new foliage.
Table 1 shows the geographic distribution and selected morphological characters of Pseudosalix and selected extant salicaceous genera with which it has been compared. Geographically, it is interesting that the successive outgroup genera to the Saliceae, including Itoa, Poliothyrsis, Carrieria, Idesia, and Bennettiodendron all are distributed in Asia today but not in North America. This distribution pattern, together with the cladistically nested position of the tribe Saliceae relative to these genera in the tribe Flacourtieae would lead to the suggestion that Saliceae originated in eastern Asia prior to dispersing across North America and Europe. However, the early fossil occurrence of both Populus and Pseudosalix in the Eocene of North America introduces the possibility of a North American diversification of the Saliceae. If this were the case, we would expect to find other fossil representatives of the Flacourtieae in North America. If the group actually evolved in Asia and subsequently dispersed to North America, then we may expect to find fossils of Pseudosalix and/or other extinct Flacourtieae, such as Utkholokia Iljinskaya & Chelebaeva, in the early Tertiary of Asia.
|
FOOTNOTES |
|---|
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSYSTEMATICSDISCUSSIONLITERATURE CITED |
|---|
Bernhard A. P. K. Endress 1999 Androecial development and systematics in Flacourtiaceae s.l. Plant Systematics and Evolution 215: 141-155[CrossRef][ISI]
Brown R. W. 1929 Additions to the flora of the Green River Formation. United States Geological Survey Professional Paper 154: -J 279-293
Brown R. W. 1934 The recognizable species of the Green River flora. United States Geological Survey Professional Paper 185: -C 45-77
Call V. B. D. L. Dilcher 1994 Parvileguminophyllum coloradensis, a new combination for Mimosites coloradensis Knowlton, Green River Formation of Utah and Colorado. Review of Palaeobotany and Palynology 80: 305-310[CrossRef]
Chandler M. E. J. 1961 The Lower Tertiary floras of southern England. 1. Paleocene Floras. London Clay Flora (Supplement). Text and Atlas. British Museum (Natural History), London, UK
Chase M. W. S. Zmarzty M. D. Lledó K. J. Wurdack S. M. Swensen M. F. Fay 2002 When in doubt, put it in Flacourtiaceae: a molecular phylogenetic analysis based on plastid rbcL DNA sequences. Kew Bulletin 57: 141-181[CrossRef]
Cockerell T. D. A. 1925 Plant and insect fossils from the Green River Eocene of Colorado. Proceedings of the United States National Museum 66, art. 19, 1–12
Collinson M. E. 1992 The early fossil history of Salicaceae: a brief review. Proceedings of the Royal Society of Edinburgh 98B: 155-167
Dahl A. O. 1972 Idesia polycarpa, an extraordinary oriental jewel. Morris Arboretum Bulletin 23: 55-57
Duncan T. C. A. Meacham 1996 MEKA: version 3.0 for Windows. University of California, Berkeley, California, USA
Eckenwalder J. E. 1980 Foliar heteromorphism in Populus (Salicaceae), a source of confusion in the taxonomy of Tertiary leaf remains. Systematic Botany 5: 366-383
Eckenwalder J. E. 1996 Systematics and evolution of Populus. In R. F. Stettler, H. D. Bradshaw, Jr., P. E. Heilman, and T. M. Hinkley [eds.], Biology of Populus and its implications for management and conservation, 7–32. National Research Council of Canada, Ottawa, Ontario, Canada
Hegnauer R. 1973 Salicaceae. In Chemotaxonomie der Pflanzen. Band 6, 241–258. Birkhäuser, Basel, Switzerland
Hickey L. J. J. A. Wolfe 1975 The bases of angiosperm phylogeny: vegetative morphology. Annals of the Missouri Botanical Garden 62: 538-589[CrossRef][ISI]
Holm L. 1969 An uredinological approach to some problems in angiosperm taxonomy. Nytt Magasin for Botanikk 16: 147-150
Iljinskaya I. A. A. I. Chelebaeva 2002 A new fossil genus Utkholokia combining leaf characters of Populus (Salicaceae) with those of Idesia and Poliothyrsis (Flacourtiaceae). Botanicheskii Zhurnal 87: 101-110
Judd W. S. 1997 The Flacourtiaceae in the southeastern United States. Harvard Papers in Botany 10: 65-79
Judd W. S. C. S. Campbell E. A. Kellogg P. F. Stevens 2002 Plant systematics: a phylogenetic approach, 2nd ed. Sinauer, Sunderland, Massachusetts, USA
Keating R. C. 1973 Pollen morphology and relationships of Flacourtiaceae. Annals of the Missouri Botanical Garden 60: 273-305[CrossRef][ISI]
Knowlton F. H. 1923 Revision of the flora of the Green River Formation. United States Geological Survey Professional Paper 131: 133-182
Leaf Architecture Working Group. 1999 Manual of leaf architecture: morphological description and categorization of dicotyledonous and net-veined monocotyledonous angiosperms. Smithsonian Institution, Washington, D.C., USA
MacGinitie H. D. 1937 The flora of the Weaverville beds of Trinity County, California with descriptions of plant-bearing beds. Carnegie Institute of Washington Publication 465: 84-151
MacGinitie H. D. 1969 The Eocene Green River flora of northwestern Colorado and northeastern Utah. University of California Publications in Geological Sciences 83: 1-203
Machlus M. S. R. Hemming P. E. Olson N. Christie-Blick 2002 40Ar/39Ar ages of selected tuffs of the Green River Formation: implications for interbasin and intrabasin correlations. Program Abstract, Paper No. 244–19, Geological Society of America Meeting Denver, 30 October 2002. Geological Society of America, Denver, Colorado, USA
Mai D. H. 1980 Zur Bedeutung von Relikten in der Florengeschichte. 100 Jahre Arboretum (1879–1979): 281-307
Manchester S. R. 1986 Vegetative and reproductive morphology of an extinct plane tree (Platanaceae) from the Eocene of western North America. Botanical Gazette 147: 200-226[CrossRef]
Manchester S. R. 1989 Attached reproductive and vegetative remains of the extinct American-European genus Cedrelospermum (Ulmaceae) from the Early Tertiary of Utah and Colorado. American Journal of Botany 76: 256-276[CrossRef][ISI]
Manchester S. R. 1994 Fruits and seeds of the Middle Eocene Nut Beds flora, Clarno Formation, north central Oregon. Palaeontographica Americana 58: 1-205
Manchester S. R. D. L. Dilcher W. D. Tidwell 1986 Interconnected reproductive and vegetative remains of Populus (Salicaceae) from the Middle Eocene Green River Formation, northeastern Utah. American Journal of Botany 73: 156-160[CrossRef][ISI]
Meeuse A. D. J. 1975 Taxonomic relationships of Salicaceae and Flacourtiaceae: their bearing on interpretative floral morphology and dilleniid phylogeny. Acta Botanica Neerlandica 24: 437-457[ISI]
Miller R. B. 1975 Systematic anatomy of the xylem and comments on the relationships of Flacourtiaceae. Journal of the Arnold Arboretum 56: 20-102[ISI]
Reid E. M. M. E. J. Chandler 1933 The London clay flora. British Museum (Natural History), London, UK
Stoehr M. U. L. Zsuffa J. E. Eckenwalder 1988 Anomalous flowers on anther-derived plants of Populus maximowiczii. American Journal of Botany 75: 594-597[CrossRef][ISI]
Watson L. M. J. Dallwitz 2000 The families of flowering plants: descriptions, illustrations, identification, and information retrieval. INTKEY, version 5.0, 14th December 2000, http://biodiversity.uno.edu/delta/
Wing S. L. 1981 A study of the paleoecology and paleobotany in the Willwood Formation (Early Eocene, Wyoming). Ph.D. dissertation, Yale University, New Haven, Connecticut, USA
This article has been cited by other articles:
|
|
 |
|
  W. L. Crepet, K. C. Nixon, and M. A. Gandolfo Fossil evidence and phylogeny: the age of major angiosperm clades based on mesofossil and macrofossil evidence from Cretaceous deposits Am. J. Botany, October 1, 2004; 91(10): 1666 - 1682. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. S. Judd and R. G. Olmstead A survey of tricolpate (eudicot) phylogenetic relationships Am. J. Botany, September 1, 2004; 91(10): 1627 - 1644. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
