Lythrum and Peplis from the Late Cretaceous and Cenozoic of North America and Eurasia: New evidence suggesting early diversification within the Lythraceae more

American Journal of Botany 98(11): 1801–1815. 2011. LYTHRUM AND PEPLIS FROM THE LATE CRETACEOUS AND CENOZOIC OF NORTH AMERICA AND EURASIA: NEW EVIDENCE SUGGESTING EARLY DIVERSIFICATION WITHIN THE LYTHRACEAE1 Friðgeir Grímsson2,3, Reinhard Zetter2, and Christa-Charlotte Hofmann2 2 University of Vienna, Department of Palaeontology, Althanstraße 14, A-1090 Vienna, Austria • Premise of the study: To fully understand the evolution of today’s angiosperms, the fossil record of plant families and genera must be used to determine their time of origin and phytogeographic history. As within many angiosperm families, the interrelationships of extant Lythraceae are hard to resolve without sufficient data from the geological past. Here we establish the earliest fossil occurrences of Lythraceae and start resolving the interrelationships and evolution of two of its genera, Lythrum and Peplis. • Methods: We studied several Cretaceous and Cenozoic palynofloras from the northern and southern hemispheres. Using the single-grain technique, we screened the treated samples for Lythrum- and Peplis-type pollen. The same individual pollen grains were observed under both the light- and scanning electron microscope, allowing a high taxonomic resolution to be achieved. • Key results: Fossil Lythraceae pollen grains are rare in palynological samples. Nevertheless, we were able to identify Lythrum and Peplis pollen from Late Cretaceous sediments and thereby extend the fossil record of the two genera by ca 70 million years. • Conclusions: The appearance of Lythrum and Peplis in North America and Peplis in Asia at approximately the same interval in the mid Late Cretaceous points to an already wide geographical distribution by then. These findings add vital information for the time of origin of the Lythraceae and suggest a higher diversity within the family. They also indicate that the distribution of particular genera during the Cretaceous was wider than previously thought. Key words: biogeography; Cenozoic; Cretaceous; evolution; fossils; Lythraceae; Lythrum; Peplis; pollen; speciation. The origin and evolution of angiosperms, today’s most advanced and widespread plant group, has been the subject of numerous research articles, each providing new information on how these plants came to dominate Earth’s terrestrial vegetation. To be able to fully understand the present number and distribution of plant taxa, their fossil record must be used to determine their time of origin and trace their diversification with time. The interrelationships of extant Lythraceae genera are hard to resolve without sufficient data from the geological past. Since the classic monograph by Koehne (1903) on extant Lythraceae, new morphological studies have shown that the original divisions established are, to a large part, artificial (Graham et al., 1993; Tobe et al., 1998). In an early study, Koehne (1903) included 22 genera in the Lythraceae. Subsequently, though, the number of genera within the Lythraceae has changed continually as genera have been moved in and out of the family or given their own family status (Koehne, 1881, 1903; Melchior, 1964; Hutchinson, 1973; Dahlgren, 1975; Cronquist, 1981; Thorne, 1981, 1992; Dahlgren and Thorne, 1 Manuscript received 2 May 2011; revision accepted 9 September 2011. The authors thank Dr. M. Tekleva, Dr. O. Volkova, and Dr. S. A. Graham for sending them herbarium material of various Lythrum and Peplis species and Dr. L. J. Hickey for providing Cretaceous samples from North America. They also thank Dr. D. K. Ferguson, Dr. H. Rice, and the two reviewers for reading and improving their manuscript. This study was funded by the FWF (Austrian Science Fund) with a grant to F.G. (Lise Meitner Position, project number M 1181-B17). 3 Author for correspondence (e-mail: fridgeir.grimsson@univie.ac.at) doi:10.3732/ajb.1100204 1984; Johnson and Briggs, 1984; Graham et al., 1993, 2005, 2011; Conti et al., 1997; Tobe et al., 1998; Shi et al., 2000; Huang and Shi, 2002; Morris, 2007). At present, the Lythraceae include ca 28–31 genera and ca 600 species (Graham et al., 2005, 2011). In recent years, molecular studies on various Lythraceae taxa have resulted in new ideas and hypotheses on species/genera interrelationships (Conti et al., 1997; Shi et al., 2000; Huang and Shi, 2002; Graham et al., 2005; Morris, 2007). Despite all these studies, the basal relationships within the family remain uncertain. However, Morris (2007) pointed out that the poor resolution at the base of the family might reflect an early, rapid radiation, which resulted in the production of several lineages within a relatively short period of time. According to molecular analyses, Lythrum L. and Peplis L. apparently form a well-supported clade (Graham et al., 2005; Morris, 2007). In Graham et al. (2005), the Lythrum/Peplis clade is suggested, in a maximum likelihood as well as Bayesian tree, to be the sister clade to Decodon, which is more basal; all other Lythraceae belong to a more advanced major clade (see fig. 4 A, B in Graham et al., 2005). In the most recent and comprehensive molecular study on Lythraceae in a Bayesian tree by Morris (2007), the Lythrum/Peplis + Decodon clade is part of one of two super clades constituting the family and sister to the remainder of the clade consisting of subclade Ammannia/Nesaea/Ginoria/Tetrat axis/Lawsonia and subclade Duabanga/Lagerstroemia/Sonner atia/Trapa (see fig. 1–4 in Morris, 2007). Morphologically, Peplis and Lythrum species are quite similar (Koehne, 1903), and therefore Webb (1967) suggested that Peplis should be merged with Lythrum. Recent molecular analyses also have nested Peplis within Lythrum, suggesting that Peplis taxa should be included as species of Lythrum (Morris, American Journal of Botany 98(11): 1801–1815, 2011; http://www.amjbot.org/ © 2011 Botanical Society of America 1801 1802 American Journal of Botany [Vol. 98 2007). On the other hand, Tobe and Raven (1983) have shown that floral nectaries are absent in the generally recognized P. portula and P. alternifolia, as well as in P. erecta (L. borysthenicum) but are present in all other Lythrum species. Furthermore, Graham et al. (1987) reported a clear difference between Peplis pollen (P. portula and P. alternifolia) and Lythrum pollen (including L. borysthenicum). Our own studies on Peplis and Lythrum pollen support the general results of Graham et al. (1987) that these pollen types are clearly distinct, though we consider the P. erecta/L. borysthenicum pollen to be morphologically more closely related to P. alternifolia and P. portula than all other Lythrum pollen figured so far (see scanning electron microscopic [SEM] work in Graham et al., 1987; Halbritter, 2000 onward; Halbritter and Weber, 2000 onward; Booi et al., 2003). In this study, therefore, we regard Peplis and Lythrum as separate genera. In this connection, it also should be pointed out that the description of P. erecta/L. borysthenicum provided by Booi et al. (2003) is largely consistent with our own observations of herbarium material, but their figured material (see Plates 1 and 2 in Booi et al., 2003) does not correspond well either with their description of pollen from this species or to the pollen we have studied from the same taxon. Lythrum has 26–36 species but Peplis only two to three (Koehne, 1903; Webb, 1967; Graham et al., 1987; Morris, 2007). All species of both these genera are herbaceous, ranging from very small (10–30 cm) single plants to relatively tall (up to ca 1.5 m) plants in large colonies (Koehne, 1903). Both genera produce very small seeds, and all species are considered to be insect pollinated. Many of the species commonly are found in wetland areas, in water or close to the shoreline of lakes, in marshlands, and in swamps (Koehne, 1903; Graham et al., 1987; Morris, 2007). The age of origin of the Lythraceae has been debated for some time. Recent molecular analyses have suggested that the family diverged from other myrtalean-based groups during the early Late Cretaceous (Sytsma et al., 2004); so far very few Lythraceae fossils have been recorded from this time period (cf. Graham and Graham, 1971; Rodríguez-de la Rosa et al., 1998; Graham et al., 2005; Estrada-Ruiz et al., 2009). In this study, we provide new data from pollen on the earliest known fossil occurrence of the family and thereby help to clarify some of the poor resolution noted by molecular analyses at the base of the Lythraceae. For this study, we used the single-grain technique to find and identify rare fossil pollen from palynological samples. We describe new species of Lythrum and Peplis from Cretaceous and Cenozoic sediments of North America, Europe, and Asia. These rare but important findings show that the Lythraceae family had already diverged in the Late Cretaceous and had a wide northern hemispheric distribution. pollen of Lythrum and/or Peplis. These are documented for the first time here. Information on the localities is included in the systematic part of the paper (see Results). Preparation of samples—To prepare the palynological samples, the sedimentary rock first was washed, dried, and hand ground in a mortar with a pestle. The resulting powder was boiled in concentrated HCl for 5 min. After decanting most of the HCl liquid, the remainder was boiled for ca 10 min in HF. The solution then was transferred to a 4-L beaker of water. After settling, the liquid was decanted, and the remainder was boiled again in HCl for 5 min. After cooling and settling, most of the HCl was decanted, the remaining solution centrifuged, and the deposit washed 3 or 4 times with water. The sample then was acetolyzed, washed again with water, and centrifuged up to 4 times. The final remaining organic material was mixed with glycerin and stored in small sample tubes. Single-grain technique—Fossil pollen grains were investigated both by light microscope (LM) and scanning electron microscope (SEM) by using the single-grain technique (cf. Zetter, 1989; Hesse et al., 2009). This technique has proved to be very useful for studying fossil palynofloras, compared with using LM only, because it allows a more accurate, systematic identification of the pollen to be made (cf. Ferguson et al., 2007; Grímsson et al., 2008, 2011; Denk et al., 2010). Drops from the sample tubes were transferred to glass slides, and single pollen grains were picked out by using a preparation needle with a human nasal hair mounted on it. The grains were placed on a separate slide within fresh drops of glycerin for photography in an LM. The pollen grains then were transferred to SEM stubs by using the preparation needle and were washed with drops of absolute ethanol to remove the remaining glycerin. The stubs then were sputter coated with gold and the pollen photographed in an SEM. Terminology for pollen description follows Punt et al. (2007) and Hesse et al. (2009). Extant material—Pollen of Lythrum and Peplis species were obtained from the herbariums of the Missouri Botanical Garden (MO), the University of Vienna (WU), and the Moscow State University (MW). Flowers from herbarium material were studied under a dissecting microscope. Mature anthers were removed from the flowers and transferred into large drops of acetolyzing liquid on a glass slide. The anthers then were broken gently by squeezing them with a dissecting needle to free the pollen. The slides were heated over a candle until the cell contents were removed and the pollen wall had acquired the desirable color for photography. Some pollen grains then were transferred into drops of glycerin on new slides and photographed in an LM. Other pollen grains from the same sample were transferred onto SEM stubs, sputter coated with gold, and photographed in an SEM. Conservation of material—All slide preparations and SEM stubs of fossil and extant pollen are stored in the collection of the Department of Palaeontology, University of Vienna, Austria (holotypes and paratypes: nos. IPUW 20100012 to 2010-0015). RESULTS Systematic paleobotany— According to Patel et al. (1984), Graham et al. (1985, 1987, 1990) and Booi et al. (2003), pollen grains of Lythraceae are mostly tricolporate, rarely four colporate. One of the characteristic features for this family is the frequent presence of pseudocolpi; these are present in 15 genera: six pseudocolpi occur in Ammannia, Crenea, Ginoria, Haitia, Hionanthera, Lafoensia, Lagerstroemia, Lawsonia, Nesaea, Pemphis, and Rotala, whereas three pseudocolpi occur in Koehneria, Lythrum, Peplis, and in some Sonneratia (Graham et al., 1985, 1987, 1990). Pseudocolpi rarely occur within angiosperm families and to date have been documented in only a few other families: Acanthaceae, Apocynaceae, Boraginaceae, Combretaceae, Crypteroniaceae, Melastomataceae, Oliniaceae, Papilionaceae, and Rhynchocalycaceae. The pollen morphology of genera/species with pseudocolpi and methods for distinguishing between these taxa have been described using LM, SEM, and TEM by Patel et al. (1984), Ghazali and Krzywinski (1989), MATERIALS AND METHODS Sedimentary samples—Numerous sedimentary samples from the northern and southern hemispheres were studied and screened for Lythraceae pollen grains. Over a 30-yr period, we have studied fossil palynofloras from the Late Cretaceous (from Austria, Germany, Russia, and the United States; see among others Hofmann and Zetter, 2007, 2010; Zetter and Hickey, 2010) and the Cenozoic (Argentina, Australia, Austria, Bulgaria, Canada, China, Columbia, Germany, Greece, Hungary, Iceland, Italy, New Zealand, Spain, Thailand, United States, and Venezuela; see among others Ferguson et al., 1998; Kovar-Eder et al., 1998; Hofmann, 2002; Zetter et al., 2002; Hofmann and Zetter, 2005; Zetter, 2006; Liu et al., 2007; Grímsson et al., 2008; Denk et al., 2010). We used the single-grain technique described herein and found a few localities with November 2011] Grímsson et al.—Fossil Lythraceae pollen from the northern hemisphere 1803 and Wang and Harley (2004). These studies show that even though species belonging to the previously mentioned families can have three pseudocolpi, their SEM sculpturing, being mostly rugulate but also psilate and reticulate, is different from the striate sculpturing of Lythrum and Peplis. Combretaceae is the only family that has species that produce striate pollen similar to those occurring in the Lythraceae; the difference between the pollen of these families/genera has been described in detail by Patel et al. (1984). Pollen grains of Lythrum are tricolporate with three pseudocolpi (heteroaperturate), pseudocolpi are narrower and shorter than the colpi, and exine striae under SEM are mostly parallel to the polar axis (excluding P. erecta/L. borysthenicum; Graham et al., 1987; Booi et al., 2003). Pollen grains of Peplis are tricolporate with three pseudocolpi (heteroaperturate), pseudocolpi are wider and shorter than colpi, whereas striae under SEM are mostly perpendicular to the polar axis (including P. erecta/L. borysthenicum; Graham et al., 1987; Booi et al., 2003). Some paleobotanists might argue that Cretaceous pollen grains, like those presented here, should not be assigned to modern genera, especially if a few minor differences exist between the fossil and the extant species (Table 1). However, the shape, outline, aperture type and number, presence of pseudocolpi, and especially the sculpturing (striation) of the Late Cretaceous pollen grains described later correspond well with the generally accepted characteristic features of extant species of both Lythrum and Peplis (cf. Graham et al., 1987; Booi et al., 2003). On the basis of the features described here, the pollen grains fit only within the family Lythraceae and can be assigned only to these two modern genera. The new species of fossil pollen described in this paper are arranged in alphabetical order of the genera, followed by their stratigraphic age, the older fossils being listed first. The descriptions contain no synonyms as there are no fossil records of these taxa prior to this study. All new species also are listed in Table 1 for comparison with their modern counterparts. Lythraceae J.St.-Hil— Lythrum L. Generic pollen diagnoses–We agree with the description provided by Graham et al. (1987) with the following additions. The most characteristic features of extant Lythrum species are the combination of three colpori and three pseudocolpi, along with parallel striae that run along the polar axis, a combination found in no other Lythraceae. In all Lythrum species, the colpori are longer and wider than the shorter pseudocolpi (see Table 1). The endopori in Lythrum are always present within the margin of the colpi and are equipped with an annulus. The surface sculpture of the colpus membrane and the pseudocolpi is microverrucate, microechinate to microrugulate. The sculptural elements mostly are widely spaced. The following descriptions of fossil Lythrum pollen grains are based on the type species, L. salicaria L. (Linnaeus, Sp. pl. 446, 1753), for which pollen was described by Booi et al. (2003). Other extant species examined include L. anceps, L. californicum, L. curtissii, L. hyssopifolia, L. lineare, and L. virgatum. Lythrum elkensis sp. nov. (Fig. 1A–F; Table 1) Diagnosis– Pollen tricolporate with three alternating pseudocolpi (heteroaperturate), colpi longer than pseudocolpi, endopori circular as wide as colpi, pseudocolpi wide in SEM, sculpturing striate, striae parallel and running along the polar axis, sometimes irregularly arranged around pori, colpus membrane microrugulate to microverrucate, pseudocolpi rugulate to microechinate, sculptural elements densely packed in both colpi and pseudocolpi. Holotype–IPUW 2010-0012-0001 (Fig. 1A–C). Paratype–IPUW 2010-0012-0002 (Fig. 1D–F). Type locality–Park County, Wyoming, northwestern USA (lat. 44°59′27″N, long. 108°52′05″W). Stratigraphy–Elk Basin, Montana Group, Eagle Formation (cf. Eldridge, 1888, 1889; Hicks, 1993; Van Boskirk, 1998). Origin of samples–The sedimentary samples were provided by Leo Hickey. For more information on geographic location and collecting, see Van Boskirk (1998). Age–Late Cretaceous (Lower Campanian), 82–81 Ma (cf. Hicks, 1993). Etymology–The species is named after the Elk Basin, where the sediments containing the Lythraceae pollen are found. Description–Pollen a monad, prolate to subprolate, elliptic in equatorial view, triangular in polar view (LM); 18–20 µm wide in LM, 14–16 µm in SEM, polar axis 21–23 µm in LM, 18–20 µm in SEM; tricolporate with three alternating pseudocolpi (heteroaperturate), colpi as long as pseudocolpi (LM, SEM), colpi 1.5–2.1 µm wide; endopori circular with an indistinct annulus, as wide as colpus (LM); pseudocolpi 5–6 µm wide in SEM; exine 2.0–3.5 µm thick, nexine thinner than sexine; tectate, sculpturing scabrate in LM, striate in SEM, striae longer in polar regions than in equatorial areas, striae parallel and running along the polar axis, sometimes irregularly arranged around pori, striae 0.2–0.4 µm wide, colpus membrane microrugulate to microverrucate, pseudocolpi rugulate to microechinate, sculptural elements densely packed in both colpi and pseudocolpi (SEM). Comparison–The Lythrum elkensis pollen grains are smaller than most pollen from extant Lythrum species (Table 1) but still comparable in size to taxa like L. alatum. The wall thickness in the fossil taxon is relatively thick and thicker than in most extant taxa, but some species like L. virgatum and L. salicaria also produce pollen with comparatively thick walls (Table 1). The striae in L. elkensis are parallel and mostly running along the polar axis, as seen in all extant species, except near the pori, where the striae are often irregularly arranged. The position and outline of the colpori are comparable to observations in extant species. In comparison to extant species, the colpori in L. elkensis are relatively narrow, and the pseudocolpi are relatively wide. The annulus in L. elkensis can be seen in LM but is less obvious in SEM, and it is clearly not as distinct as in extant species (compare Fig. 1A, B, D, E to Fig. 2A, B, D, E, G, H, J, K). The sculpturing of the colpus membrane and the pseudocolpi in L. elkensis is microrugulate to microechinate, similar to that seen in extant taxa, though in the fossil species the sculptural elements are more densely packed (compare Fig. 1C, F with Fig. 2F, I, L). Remarks–This is the earliest record of Lythrum worldwide. Occurrence–Angiosperm pollen grains are frequent in this sample, but the Lythrum pollen is rare. Lythrum wilhelmii sp. nov. (Fig. 1G–L; Table 1) Diagnosis–Pollen tricolporate with three alternating pseudocolpi (heteroaperturate), colpi longer than pseudocolpi, endopori large circular with a prominent annulus, pseudocolpi wide, sculpturing striate, striae parallel and running along the polar axis, colpus membrane microverrucate to microrugulate, pseudocolpi microrugulate to microechinate, sculptural elements densely to widely spaced in colpi but densely packed in pseudocolpi. 1804 American Journal of Botany [Vol. 98 Table 1. Comparison of fossil and extant Lythrum and Peplis pollen. L. elkensis L. wilhelmii Late Cretaceous Late Miocene Wyoming, USA Rechnitz, Austria Shape Outline eq.v. Outline p.v. Size in LM (eq.d./p.a.; µm) Size in SEM (eq.d./p.a.; µm) Aperture Colpi width (µm) Colpi vs. pseudocolpi Width of pseudocolpi in SEM (µm) Exine thickness (µm) Exine Prolate to subprolate Elliptic Triangular 18–20 / 21–23 14–16 / 18–20 Tricolporate 1.5–2.1 Equal in length 5.0–6.0 2.0–3.5 Nexine thinner than sexine Oblate to spheroidal Elliptic to circular Hexagonal to circular 22–25 / 21–23 21–25 / 21–22 Tricolporate 4.1–6.2 Colpi longer 2.1–5.1 1.3–1.7 Nexine as thick or slightly thicker than sexine Striate, striae parallel and running along the polar axis L. alatum Extant SE Canada, E USA Oblate to spheroidal Elliptic to circular Circular 17–19 / 16–17 12–16 / 13–15 Tricolporate 3.7–5.2 Colpi longer 1.2–2.3 1.9–2.1 Nexine thinner or as thick as sexine Striate, striae parallel and running along the polar axis L. anceps1 Extant Japan Spheroidal to subprolate Circular to elliptic Circular to hexagonal — 20–21 / — Tricolporate 5.3–6.0 Colpi longer 2.4–4.8 — Nexine thinner or as thick as sexine Striate, striae parallel and running along the polar axis L. californicum1 L. curtissii L. hyssopifolia2,3 Extant SW USA Extant SE USA Extant Eurasia Spheroidal to subprolate Circular to elliptic Circular to hexagonal — — / 18–20 Tricolporate 4.1–4.8 Colpi longer — — Nexine thinner or as thick as sexine Striate, striae parallel and running along the polar axis Spheroidal to oblate Circular Circular 25–27 / 23–26 24–28 / 24–26 Tricolporate 7.8–8.2 Colpi longer 1.8–2.3 1.0–1.9 Nexine thinner or as thick as sexine Striate, striae parallel and running along the polar axis Spheroidal to subprolate Circular to elliptic Circular to hexagonal 21–26 / 20–25 23–25 / 22–24 Tricolporate 5.8–6.9 Colpi longer 2.4–3.4 1.5–2.2 Nexine thinner or as thick as sexine Striate, striae parallel and running along the polar axis L. lineare1 Extant SE USA Spheroidal to subprolate Circular to elliptic Circular to hexagonal — 28–29 / 27–28 Tricolporate 6.1–6.9 Colpi longer — — Nexine thinner or as thick as sexine Striate, striae parallel and running along the polar axis Striate, striae parallel and running along the polar axis, sometimes irregularly arranged around pori Width of striae 0.2–0.4 in SEM Colpus Microrugulate membrane to microverrucate, elements densely packed Pseudocolpi Sculpturing in SEM 0.4–0.6 0.2–0.4 0.3–0.5 0.2–0.3 Microverrucate to microechinate, element widely spaced Microverrucate to microechinate, elements widely spaced 0.3–0.5 0.3–0.5 0.3–0.5 Microverrucate to microrugulate, elements widely spaced Microverrucate to microrugulate, elements widely spaced Microverrucate Microverrucate, Microrugulate to to microrugulate, elements widely microverrucate elements densely spaced to microbaculate, to widely spaced elements widely to densely spaced Microverrucate Microverrucate to microrugulate, to microrugulate elements widely to microechinate, spaced elements widely spaced Microverrucate to microrugulate, elements widely spaced Microverrucate to microrugulate to microechinate, elements widely spaced Rugulate to Microrugulate to Microverrucate, Microrugulate to microechinate, microechinate elements widely microverrucate elements densely spaced to microbaculate, packed elements widely to densely spaced Note: Numbers accompanying the names of some extant species refer to figured material that is the basis for the description and measurements given LM = light microscope; SEM = scanning electron microscope; eq.v = equatorial view; p.v. = polar view; eq.d. = equatorial diameter; p.a. = polar axis. Holotype–IPUW 2010-0015-0001 (Fig. 1G–I). Paratype–IPUW 2010-0015-0002 (Fig. 1J–L). Type locality–Rechnitz, Badersdorf, southeastern Austria (lat. 47°12′06″N, long. 16°22′23″E). Stratigraphy–Pannonian Lignite Series op2 (cf. Nebert, 1979). Origin of samples–All sedimentary samples were collected by Reinhard Zetter and Christa-Charlotte Hofmann. For more information on geographic location and collecting, see Zetter (1987) and Hofmann and Zetter (2005). Age–Late Miocene (Upper Pannonian), ca 9.5 Ma (cf. Zetter, 1987; Daxner-Höck, 1996; Rögl and Daxner-Höck, 1996; Hofmann and Zetter, 2005). Etymology–The species name honors the palynologist Prof. Wilhelm Klaus, who was one of the first palynologists to use the combination of LM and SEM to study paleo-palynofloras. Description–Pollen a monad, oblate to spheroidal, elliptic to circular in equatorial view, hexagonal to circular in polar view, 22–25 µm wide in LM, 21–25 µm in SEM, polar axis 21–23 µm in LM, 21–22 in SEM; tricolporate with three alter- November 2011] Grímsson et al.—Fossil Lythraceae pollen from the northern hemisphere 1805 L. virgatum2 Extant Eurasia Spheroidal to subprolate Circular to elliptic Circular to hexagonal 20–39 / 20–38 19–20 / 18–19 Tricolporate 5.5–6.5 Colpi longer 1.9–2.2 2.5–3.5 Nexine thinner than sexine Striate, striae parallel and running along the polar axis L. salicaria1,2,4 Extant Eurasia Spheroidal to subprolate Circular to elliptic Circular to hexagonal 21–35 / 20–35 18–28 / 19–31 Tricolporate 5.2–8.5 Colpi longer 1.0–3.0 2.5–3.5 Nexine thinner or as thick as sexine Striate, striae parallel and running along the polar axis P. eaglensis Late Cretaceous Wyoming, USA Subprolate to prolate Elliptic Triangular to circular 23–25 / 24–29 17–21 / 22–24 Tricolporate 1.0–4.1 Equal or colpi longer 3.0-4.0 2.6–2.8 (−3.6) Nexine thinner or as thick as sexine Striate, striae mostly parallel and running perpendicular to the polar axis P. yakutiana Late Cretaceous W Siberia, Russia Prolate Elliptic Triangular to circular 19–29 / 25–35 12–23 / 21–27 Tricolporate 2.1–4.7 Equal or colpi longer 6.0–8.0 2.3–2.6 (−3.4) Nexine thinner than sexine Striate, striae mostly parallel and running perpendicular to the polar axis P. alternifolia Extant Europe, C Asia Spheroidal to subprolate Circular to elliptic Triangular to circular 15–19 / 15–18 12–15 / 13–16 Tricolporate 2.0–3.6 Colpi longer 2.8–5.5 0.7–1.1 Nexine thinner or as thick as sexine Striate, striae are parallel and running perpendicular to the polar axis P. erecta Extant Eurasia Spheroidal to oblate Circular to elliptic Circular to hexagonal 16–22 / 15–20 14–17 / 14–16 Tetracolporate, rarely tricolporate 2.8–4.0 Colpi longer 3.0–3.5 0.8–1.4 Nexine as thick as sexine Striate, striae are irregularly arranged P. portula Extant Eurasia Oblate to spheroidal Elliptic to circular Triangular to circular 16–18 / 12–15 12–14 / 12–13 Tricolporate 2.2–4.3 Colpi longer 3.3–7.3 0.8–1.4 Nexine as thick as sexine Striate, striae are parallel and running perpendicular to the polar axis 0.2–0.4 Microverrucate to microrugulate, widely spaced 0.3–0.6 Microverrucate to microrugulate, elements widely spaced Microverrucate to microrugulate, elements widely spaced 0.1–0.2 Microrugulate, elements densely packed 0.3–0.6 Shortly striate 0.1–0.3 Microverrucate to microrugulate, elements widely spaced 0.1–0.3 Microrugulate to microverrucate and microbaculate, elements densely to widely spaced Microrugulate to microverrucate and microbaculate, elements densely to widely spaced 0.1–0.3 Microrugulate, microverrucate, microechinate, and microbaculate, elements widely spaced Microrugulate, microverrucate, microechinate, and microbaculate, elements widely spaced Microverrucate to microrugulate, widely spaced Microrugulate, elements densely packed Rugulate to shortly striate Microverrucate to microrugulate, elements widely spaced here. 1 = Graham et al., 1987; 2 = Booi et al., 2003; 3 = Halbritter, 2000 onward; 4 = Halbritter and Weber, 2000 onward. nating pseudocolpi (heteroaperturate), colpi longer than pseudocolpi (LM, SEM), colpi 2.2–3.2 µm wide; endopori large circular with an annulus surrounding the pore (LM, SEM); pseudocolpi 2.1–5.1 µm wide in SEM; exine 1.3–1.7 µm thick, nexine as thick or slightly thicker than sexine; tectate, sculpturing psilate to slightly scabrate in LM, striate in SEM, striae parallel and running along the polar axis, striae 0.4–0.6 µm wide, colpus membrane microverrucate to microrugulate, pseudocolpi microrugulate to microechinate, sculptural elements densely to widely spaced in colpi but densely packed in pseudocolpi (SEM). Comparison–The Lythrum wilhelmii pollen grains are comparable in size and wall thickness to most extant species. The striae are parallel to the polar axis as in extant species (Table 1). The colpori are longer than the pseudocolpi, as in living species. The widths of the colpi and pseudocolpi are more or less identical (Fig. 1H, I, K, L), whereas in extant species, the pseudocolpi are usually narrower. The pori are positioned within the margin 1806 American Journal of Botany [Vol. 98 Fig. 1. Fossil Lythrum pollen. (A–F) Lythrum elkensis sp. nov. from the Late Cretaceous, Eagle Formation, Wyoming, USA. (A, D) Light micrographs, polar and equatorial view. (B, C, E, F) Scanning electron micrographs of the same pollen grains showing the surface sculpturing. (G–L) Lythrum wilhelmii sp. nov. from the Late Miocene Lignite Series, Badersdorf, southeastern Austria. (G, J) Light micrographs, polar view. (H, I, K, L) Scanning electron micrographs of the same pollen grains showing the surface sculpturing. Scale bar = 10 µm in A, B, D, E, G, H, J, and K; 1 µm in C, F, I, and L. November 2011] Grímsson et al.—Fossil Lythraceae pollen from the northern hemisphere 1807 Fig. 2. Extant Lythrum pollen. (A–F) Lythrum alatum, J. L. Hillyer, 004692 (WU). (A, D) Light micrographs, polar view (A) and equatorial view (D). (B, C, E, F) Scanning electron micrographs, polar view (B), detail of polar area (C), equatorial view (E), and detail showing colpus membrane (F). (G–L) Lythrum curtissii, A. H. Curtiss, 0030617 (WU). (G, J) Light micrographs, polar view (G) and equatorial view (J). (H, I, K, L) Scanning electron micrographs, polar view (H), detail of polar area showing striae and part of colpus membrane (I), oblique view (K), and detail showing distal part of pseudocolpi and surrounding striae (L). Scale bar = 10 µm in A, B, D, E, G, H, J, and K; 1 µm in C, F, I, and L. 1808 American Journal of Botany [Vol. 98 Fig. 3. Fossil Peplis pollen. (A–L) Peplis eaglensis sp. nov. from the Late Cretaceous, Eagle Formation, Wyoming, USA. (A, D, G, J) Light micrographs, equatorial view (A, D, G) and polar view (J). (B, C, E, F, H, I, K, L) Scanning electron micrographs of the same pollen grain from both sides, equatorial view with pseudocolpi in middle (B), detail of polar region showing parallel striae (C), detail showing sexine arching over pori (E), detail of polar region showing the microrugulate sculpturing of the pseudocolpi and parallel striae running perpendicular to the polar axis (F), equatorial view with colpi in the middle (H), detail showing the microrugulate colpus membrane (I), detail showing the microrugulate pseudocolpi (K), and detail showing the striae in the polar region (L). Scale bar = 10 µm in A, B, D, G, H, and J; 1 µm in C, E, F, I, K, L. November 2011] Grímsson et al.—Fossil Lythraceae pollen from the northern hemisphere 1809 of the colpi, as in extant species, and bear a distinct annulus (Fig. 1K, L). The sculpturing of the colpus membrane and the pseudocolpi in L. wilhelmii is microverrucate to microrugulate. In the colpi, the sculptural elements are densely to widely spaced, but in extant species they mostly are widely spaced (compare Fig. 1H, K, L with Fig. 2E, F, H, I). In the pseudocolpi of L. wilhelmii, the microverrucate to microrugulate elements are densely packed, whereas in extant species, the elements are usually widely spaced (compare Fig. 1G, I, K with Fig. 2K, L). Lythrum wilhelmii differs from Lythrum elkensis in being larger and having striae that run parallel to the polar axis even in the equatorial region. The striate area is also slightly wider, the pseudocolpi are much narrower, and in the colpori, the sculptural elements of the colpus membrane are not as densely packed. Remarks–Similar pollen grains have been described from Late Miocene and Pliocene sediments of other European localities (see Discussion). Those few reports are based on LM studies only, making further comparison impossible. Occurrence–The pollen are quite rare in the sample; other insect-pollinated angiosperm pollen are also rare. The frequently occurring pollen types are all from wind-pollinated taxa. Peplis L. Generic pollen diagnosis–We agree with the description provided by Graham et al. (1987), with the following additions. The most characteristic features of extant Peplis species are the combination of three colpori and three pseudocolpi, with parallel striae, which, in contrast to those in Lythrum, run perpendicular to the polar axis. In Peplis, the colpori are longer and narrower than the relatively wide pseudocolpi (see Table 1). The endopori in Peplis, like those in Lythrum, are within the margin of the colpi, but the annulus is often less distinct. The sculpturing of the colpus membranes and the pseudocolpi in Peplis shows the same variability as in Lythrum. Sculptural elements in the colpi and pseudocolpi of Peplis are also widely spaced. The following descriptions of fossil Peplis pollen grains are based on the type species, P. portula L. (Linnaeus, Sp. Pl. 332, 1753), for which pollen was described by Graham et al. (1987). Other extant species examined include P. alternifolia and P. erecta. Peplis eaglensis sp. nov. (Figs. 3A–L, 4A–I; Table 1)Diagnosis–Pollen tricolporate with three alternating pseudocolpi (heteroaperturate), colpi slightly longer or as long as pseudocolpi, endopori elliptic to circular, sexine sometimes arching over pori, sculpturing striate, striae mostly parallel and running perpendicular to the polar axis, colpus membrane microrugulate, pseudocolpi microrugulate, sculptural elements densely packed in both colpi and pseudocolpi. Holotype–IPUW 2010-0013-0001 (Fig. 3A–L). Paratype–IPUW 2010-0013-0002 (Fig. 4A–C), 2010-00130003 (Fig. 4D–F), 2010-0013-0004 (Fig. 4G–I). Type locality–Park County, Wyoming, northwestern USA (lat. 44°59′27″N, long. 108°52′05″W). Stratigraphy–Elk Basin, Montana Group, Eagle Formation (cf. Eldridge, 1888, 1889; Hicks, 1993; Van Boskirk, 1998). Origin of samples–The sedimentary samples were provided by Leo Hickey. For more information on geographic location and collecting, see Van Boskirk (1998). Age–Late Cretaceous (Lower Campanian), 82–81 Ma (cf. Hicks, 1993). Etymology–The species is named after the Eagle Formation. The sediments containing the new Peplis pollen type are part of this formation. Description–Pollen a monad, subprolate to prolate, elliptic in equatorial view, triangular to circular in polar view (LM); 23– 25 µm wide in LM, 17–21 µm in SEM, polar axis 26–29 µm in LM, 22–24 in SEM; tricolporate with three alternating pseudocolpi (heteroaperturate), colpi slightly longer or the same as pseudocolpi (LM, SEM), colpi 1.2–4.5 µm wide; endopori elliptic to circular (LM), sexine sometimes arching over pori (SEM), pseudocolpi 3–4 µm wide in SEM; exine 2.6–2.8 (–3.6) µm thick, nexine thinner or as thick as sexine; tectate, sculpturing scabrate in LM, striate in SEM, striae mostly parallel and running perpendicular to the polar axis, striae 0.1–0.2 µm wide, colpus membrane microrugulate, pseudocolpi microrugulate, sculptural elements densely packed in both colpi and pseudocolpi (SEM). Comparison–The Peplis eaglensis pollen grains are larger than grains of extant Peplis species, and the pollen wall is also thicker (Table 1). The striae are parallel and run perpendicular to the polar axis as in the recent P. erecta and P. portula. The colpi in P. eaglensis are longer or equal in length to the pseudocolpi; in extant taxa the pseudocolpi are always shorter. Pseudocolpi are wider than the colpi in both P. eaglensis and the extant species. The pori are positioned within the margin of the colpi, as in extant species, but there is a conspicuous arching of the sexine over the pori (Fig. 3B, E, H), compared with a relatively lowrelief annulus seen in extant taxa (Figs. 5D-L, 6). The sculpturing of the colpus membrane and the pseudocolpi in P. eaglensis is microrugulate, but it is more variable in extant taxa (Table 1). In the fossil Peplis, the sculptural elements are also densely packed compared with the wider spacing in extant species (compare Fig. 3F, I, K and Fig. 4B, C with Fig. 5H, I, K, L). Remarks–This is the earliest record of Peplis worldwide. Occurrence–This is a rare element in the sample but is found more frequently than the new Lythrum species described from the same sediments. Peplis yakutiana sp. nov. (Figs. 4J–L, 5A–C; Table 1) Diagnosis–Pollen tricolporate with three alternating pseudocolpi (heteroaperturate), colpi slightly longer or as long as pseudocolpi, endopori circular, sexine sometimes arching over pori, pseudocolpi wide, sculpturing striate, striae mostly parallel and running perpendicular to the polar axis, colpus membrane shortly striate, pseudocolpi rugulate to shortly striate, sculptural elements densely packed in both colpi and pseudocolpi. Holotype–IPUW 2010-0014-0001 (Fig. 4J–L). Paratype–IPUW 2010-0014-0002 (Fig. 5A–C). Type locality–Tyung River, Yakutia (Sakha), western Central Siberia, Russia (lat. 64°15′16″N, long. 121°21′32″E). Stratigraphy–Vilyuy (Vilui) Basin, Timerdyakh Formation (cf. Samoilovitch and Mchedlishvili, 1961; Samoilovitch, 1965). Origin of samples–Sedimentary samples were provided by Robert A. Spicer. For more information on geographical location and collecting, see Spicer et al. (2008) and Hofmann and Zetter (2010). Age–Late Cretaceous (Upper Campanian/Lower Maastrichtian), 72–68 Ma (cf. Hofmann and Zetter, 2007, 2010). Etymology–The species is named after the Yakutia (Sakha) Republic, which belongs to the Russian Federation. The fossiliferous locality lies within this republic. Description–Pollen a monad, prolate, elliptic in equatorial view, triangular to circular in polar view (LM); 19–29 µm wide 1810 American Journal of Botany [Vol. 98 November 2011] Grímsson et al.—Fossil Lythraceae pollen from the northern hemisphere 1811 in LM, 12–23 µm wide in SEM, polar axis 25–35 µm in LM, 21–27 µm in SEM; tricolporate with three alternating pseudocolpi (heteroaperturate), colpi slightly longer or as long as pseudocolpi (LM, SEM), colpi 2.1–4.7 µm wide; endopori circular (LM), sexine sometimes arching over pori (SEM), pseudocolpi 6–8 µm wide in SEM; exine 2.3–2.6 (–3.4) µm thick, nexine thinner than sexine; tectate, sculpturing scabrate in LM, striate in SEM, striae mostly parallel and running perpendicular to the polar axis, striae 0.3–0.6 µm wide, colpus membrane shortly striate, pseudocolpi rugulate to shortly striate, sculptural elements densely packed in both colpi and pseudocolpi (SEM). Comparison–The Peplis yakutiana pollen grains can be divided into two groups according to size. The large pollen grains (Fig. 4J–L) are bigger and seem to have a thicker wall than the extant Peplis species (Fig. 5G, J, 6A, D, G, J) The smaller pollen grains (Fig. 5A–C) are within the size range of extant taxa but still have thicker walls (Table 1). The striae are parallel and run perpendicular to the polar axis, as in extant taxa (Figs. 5D-L, 6), and are slightly wider than the striae in extant species. The colpi in P. yakutiana are longer or equal in length to the pseudocolpi, compared with longer pseudocolpi in extant taxa (Fig. 5H, K, 6B, E, H, K). The pseudocolpi are wider than the colpi, as in extant species (Table 1). The pori are positioned within the margin of the colpi, as in extant species, but there is a conspicuous arching (similar to the one in P. eaglensis) of the sexine over the pori (Fig. 5B, C). In extant taxa, the pori have a low-relief annulus. The sculpturing of the colpus membrane and the pseudocolpi in P. yakutiana is shortly striate (Fig. 5B, C) to rugulate (Fig. 5K, L). Although this sculpturing of the colpus membrane and the pseudocolpi is not known in any extant Peplis species, all the remaining pollen characters fit this genus. The main differences between Peplis eaglensis and P. yakutiana are that in the latter, the striae along the colpori are broader and also farther apart, whereas the sculpturing in the colpus area and the pseudocolpi is striate in P. yakutiana but microrugulate in P. eaglensis. The P. yakutiana pollen grains are often slightly larger than pollen of P. eaglensis. Remarks–This is the earliest record of Peplis for Siberia (Russia) and also for the whole of Eurasia. Occurrence–Angiosperm pollen compose 40–70% of the taxa identified from these sediments. The Peplis type pollen is relatively rare compared with most of the other angiosperm pollen types. DISCUSSION Fossil record of Lythrum and Peplis and ideas on the paleobiogeography of the two genera— The fossil record of Lythrum and Peplis is very sparse. Being herbaceous, Lythrum and Peplis plants do not shed their leaves in winter. Their leaves remain attached to the stem and disintegrate (decay) along with the rest of the plant. Leaf production per plant is small compared with that of woody plants, and being generally more membranous, they are more easily destroyed when transported by water and ← winds than are leaves of hardy woody plants. As a result, their preservation potential is quite low (Ferguson, 1985). It therefore comes as no surprise that there are no fossil leaf records for Lythrum or Peplis. The seeds of Lythrum and Peplis are relatively small and have not been recognized in the fossil record either. As mentioned earlier, all Lythrum and Peplis species are insect pollinated. Insect-pollinated plants produce relatively few pollen grains compared with wind-pollinated plants and therefore rarely are found in any large quantities in fossil palynological assemblages (cf. Fægri and Iversen, 1989). Insectpollinated plants identified from fossil pollen often originated from plants growing in or close to some type of water body (lake, river, swamp, etc.) and often are believed to have reached the depocenter as whole flowers or flower parts that have fallen or been transported into lacustrine sedimentary environments (cf. Fægri and Iversen, 1989), a factor that also explains the rarity of some species in the fossil record. The sparse pollen record of Lythrum, Peplis, and other Lythraceae genera has been summarized by Graham and Graham (1971) and Muller (1981). Only a few of the extant genera have been reported from the fossil record so far; Crenea seems to occur from the Eocene onwards and Cuphea and Lagerstroemia from the Miocene onwards (cf. Graham and Graham, 1971; Muller, 1981; Liu et al., 2008). Lagerstroemia pollen also have been encountered in middle Eocene sediments of Hainan, China (C.-C. Hofmann, personal observation), and of the Princeton chert in British Columbia, Canada (R. Zetter, personal observation). Pollen of Decodon occur in sediments from the Eocene up to the present (Grímsson et al., work in progress). The oldest record of Lythrum or Peplis before this study is Lythrum pollen from the Late Miocene of Spain (Van Campo, 1976). Younger records of Lythrum pollen are from the Pliocene of Germany (Menke, 1976), whereas Lythrum and Peplis pollen grains are also known from the Pliocene of eastern Macedonia, Greece (Wijmstra, 1969). So far, there are no macro- or mesofossil records (vegetative parts, leaves, fruits, seeds) of Lythrum or Peplis from sediments older than the Holocene. All records before this study are younger than 10 Ma. The oldest remains presented here are between 82–81 Ma in age, extending the fossil record of the two genera by ca 70 million years, to the Late Cretaceous. This is also one of the earliest records for the family of Lythraceae (cf. Graham and Graham, 1971; Rodríguez-de la Rosa et al., 1998; Graham et al., 2005; Estrada-Ruiz et al., 2009). On the basis of the new discoveries, it is clear that pollen with morphological characteristics of Lythrum and Peplis already existed in the Late Cretaceous (Lower Campanian, 82–81 Ma; Hicks, 1993) of North America. Clear Peplis type (different from the North American one) pollen grains now also are known from the latest Cretaceous (Upper Campanian/Lower Maastrichtian, 72–68 Ma; Hofmann and Zetter, 2007, 2010) of Siberia, Russia. These findings suggest that Lythrum and Peplis must have had an extensive northern hemisphere distribution on the Laurasian landmass during the Late Cretaceous. Authors of previous studies on the family (cf. Graham et al., 2005; Estrada-Ruiz et al., Fig. 4. Fossil Peplis pollen. (A–J) Peplis eaglensis sp. nov. from the Late Cretaceous, Eagle Formation, Wyoming, USA. (A, D, G) Light micrographs, equatorial view (A, D) and oblique polar view (G). (B, C, E, F, H, I) Scanning electron micrographs of the same pollen grains, equatorial view (B), detail showing pseudocolpi in the equatorial region (C), equatorial view (E), detail of colpi showing colpus membrane (F), oblique polar view (H), and detail of polar area (I). (J–L) Peplis yakutiana sp. nov. from the latest Late Cretaceous, Timerdyakh Formation, W. Siberia, Russia. (J). Light micrographs, polar view (upper right) and equatorial view (lower left). (K, L) Scanning electron micrographs of the same pollen grain, equatorial view (K) and detail of pseudocolpi showing striae (L). 1812 American Journal of Botany [Vol. 98 Fig. 5. Fossil and extant Peplis pollen. (A–C) Peplis yakutiana sp. nov. from the Late Cretaceous, Timerdyakh Formation, W. Siberia, Russia. (A). Light micrographs, polar view (upper) and equatorial view (lower). (B, C) Scanning electron micrographs of the same pollen grain, equatorial view (B) and detail of colpori showing sexine arching over pori (C). (D–F) Peplis alternifolia, V. Tikhomirov and A. Timonin, 12618 (MW). (G–L) Peplis alternifolia, N. N. Tzelev and E. V. Petchenyuk, s.n. in 1983 (MO). (D, G, J) Light micrographs, polar view (D, upper part of G) and equatorial view (lower part of G, J). (E, F, H, I, K, L) Scanning electron micrographs, polar view (E), detail of polar region (F), oblique equatorial view with the pseudocolpi in middle (H), detail of pseudocolpi showing the sculptural elements (I), equatorial view with the colpori in the middle (K), and detail showing the pori and the colpus membrane (L). November 2011] Grímsson et al.—Fossil Lythraceae pollen from the northern hemisphere 1813 Fig. 6. Extant Peplis pollen. (A–F) Peplis erecta, N. N. Tzelev and E. V. Petchenyuk, 5689091 (MO). (A, D) Light micrographs, polar view (A) and equatorial view (D). (B, C, E, F) Scanning electron micrographs, polar view (B), detail of polar region showing irregular striae and colpus membrane (C), equatorial view (E), and detail of pseudocolpi showing the sculptural elements (F). (G–I) Peplis portula, O. Murmann, 0041882 (WU). (J–L) Peplis portula, N. Shvedchikova, s.n. 1997. (G, J) Light micrographs, polar view (G) and equatorial view (J). (H, I, K, L) Scanning electron micrographs, polar view (H), detail of polar region (I), equatorial view (K), and detail of the colpus showing the pori and the colpus membrane (L). 1814 American Journal of Botany [Vol. 98 2009) have speculated about the time and especially the place of origin of the Lythraceae. Although it might be impossible to determine the time and place, given the presence and/or absence of Lythrum and Peplis in the various fossil floras of the world, it seems most likely that these two genera began their diversification on Laurasia during the mid Late Cretaceous. Effect and correlation of the fossils to cladistic and phylogenetic analysis within Lythraceae and related groups— Divergence dates for families within the order Myrtales have been estimated from molecular analyses and suggest that the Lythraceae–Onagraceae lineage already was established during the latest Early Cretaceous (end of the Albian, ca 100 Ma; Sytsma et al., 2004). The two families are believed to have diverged soon thereafter, during the early Late Cretaceous (end of the Cenomanian, ca 94 Ma; Sytsma et al., 2004). Given the new Late Cretaceous (Lower Campanian to Lower Maastrichtian) fossil records of Lythrum and Peplis, the time of origin for the Lythraceae proposed by Sytsma et al. (2004) appears quite plausible. At least it is clear from the fossil record that Lythrum and Peplis were already established genera by the Late Cretaceous (Lower Campanian), suggesting that the family diverged some time earlier, most likely during the early mid Late Cretaceous (ca. 90 Ma). This still needs to be proved and will be possible only with additional information from the fossil record. Conclusions— The use of advanced methods, such as the single-grain technique, when studying microfloras makes it possible to find extremely rare elements within fossil floras. In this case, the Lythrum and Peplis pollen reported here are unique to the fossil record both in time and place, and they established the presence of the two genera far earlier than previously known. Further, the appearance of Lythrum in North America and Peplis in Asia, at approximately the same time, points to an already wide geographical distribution by the mid Late Cretaceous. These new findings provide fundamental information for future cladistic and phylogenetically based studies, as the fossils can be used to anchor the age of origin and time of divergence within the Lythraceae family. LITERATURE CITED Booi, M., W. Punt, and P. P. Hoen. 2003. Lythraceae. The northwest European pollen flora 68. Review of Palaeobotany and Palynology 123: 163–180. Conti, E., A. Litt, P. G. Wilson, S. A. Graham, B. G. Briggs, L. A. S. Johnson, and K. J. Sytsma. 1997. Interfamiliar relationships in Myrtales: Molecular phylogeny and patterns of morphological evolution. Systematic Botany 22: 629–647. Cronquist, A. 1981. An integrated system of classification of flowering plants. Columbia University Press, New York, New York, USA. Dahlgren, R. 1975. A system of classification of the angiosperms to be used to demonstrate the distribution of characters. Botaniska Notiser 128: 119–147. Dahlgren, R., and R. F. Thorne. 1984. The order Myrtales: Circumscription, variation, and relationships. Annals of the Missouri Botanical Garden 81: 419–450. Daxner-Höck, G. 1996. Faunenwandel im Obermiozän und Korrelation der MN-“Zonen” mit den Biozonen des Pannons der Zentralen Paratethys. Beiträge zur Paläontologie 21: 1–9. Denk, T., F. Grímsson, and R. Zetter. 2010. Episodic migration of oaks to Iceland: Evidence for a North Atlantic “land bridge” in the latest Miocene. American Journal of Botany 97: 276–287. El Ghazali, G. E. B., and K. Krzywinski. 1989. An attempt to clarify the term heterocolpate. Grana 28: 179–186. Eldridge, G. H. 1888. On some stratigraphical and structural features of the country about Denver, Colorado. Colorado Scientific Society Proceedings 3: 86–118. Eldridge, G. H. 1889. Some suggestions upon the methods of grouping the formations of the Middle Cretaceous and the employment of an additional term in its nomenclature. American Journal of Science 38: 313–321. Estrada-Ruiz, E., L. Calvillo-Canadell, and S. R. S. CevallosFerriz. 2009. Upper Cretaceous aquatic plants from Northern Mexico. Aquatic Botany 90: 282–288. Fægri, K., and J. Iversen. 1989. Textbook of pollen analysis, 4th ed. [revised by K. Fægri, P. E. Kaland, and K. Krzywinski]. Blackburn Press, Caldwell, New Jersey, USA. Ferguson, D. K. 1985. The origin of leaf-assemblages—New light on an old problem. Review of Palaeobotany and Palynology 46: 117–188. Ferguson, D. K., M. Pingen, R. Zetter, and C.-C. Hofmann. 1998. Advances in our knowledge of the Miocene plant assemblage from Kreuzau, Germany. Review of Palaeobotany and Palynology 101: 147–177. Ferguson, D. K., R. Zetter, and K. N. Paudayal. 2007. The need for SEM in palaeopalynology. Comptes Rendus Palévol 6: 423–430. Graham, A., and S. A. Graham. 1971. The geologic history of the Lythraceae. Brittonia 23: 335–346. Graham, A., S. A. Graham, J. W. Nowicke, V. Patel, and S. Lee. 1990. Palynology and systematics of the Lythraceae. III. Genera Physocalymma through Woodfordia, addenda, and conclusions. American Journal of Botany 77: 159–177. Graham, A., J. W. Nowicke, J. J. Skvarla, S. A. Graham, V. Patel, and S. Lee. 1985. Palynology and systematics of the Lythraceae. I. Introduction and genera Adenaria through Ginoria. American Journal of Botany 72: 1012–1031. Graham, A., J. W. Nowicke, J. J. Skvarla, S. A. Graham, V. Patel, and S. Lee. 1987. Palynology and systematics of the Lythraceae. II. Genera Haitia through Peplis. American Journal of Botany 74: 829–850. Graham, S. A., J. V. Crisci, and P. C. Hoch. 1993. Cladistic analysis of the Lythraceae sensu lato based on morphological characters. Botanical Journal of the Linnean Society 113: 1–33. Graham, S. A., M. Diazgradados, and J. C. Barber. 2011. Relationships among the confounding genera Ammannia, Hionanthera, Nesaea and Rotala (Lythraceae). Botanical Journal of the Linnean Society 166: 1–19. Graham, S. A., J. Hall, K. Sytsma, and S. Shi. 2005. Phylogenetic analysis of the Lythraceae based on four gene regions and morphology. International Journal of Plant Sciences 166: 995–1017. Grímsson, F., T. Denk, and R. Zetter. 2008. Pollen, fruits, and leaves of Tetracentron (Trochodendraceae) from the Cainozoic of Iceland and western North America and their palaeobiogeographic implications. Grana 47: 1–14. Grímsson, F., R. Zetter, and C. Baal. 2011. Combined LM and SEM study of the Middle Miocene (Sarmatian) palynoflora from the Lavanttal Basin, Austria: Part I. Bryophyta, Lycopodiophyta, Pteridophyta, Ginkgophyta, and Gnetophyta. Grana 50: 102–128. Halbritter, H. 2000 onward. Lythrum hyssopifolia. In R. Buchner and M. Weber, PalDat—Palynological database: Descriptions, illustrations, identification, and information retrieval. Website http://www. paldat.org/.Accessed 30-07-2010. Halbritter, H., and M. Weber. 2000 onward. Lythrum salicaria. In R. Buchner and M. Weber [eds.], PalDat—Palynological database: Descriptions, illustrations, identification, and information retrieval. Website http://www.paldat.org/.Accessed 30-07-2010. Hesse, M., H. Halbritter, R. Zetter, M. Weber, R. Buchner, A. FroschRadivo, and S. Ulrich. 2009. Pollen terminology–An illustrated handbook. Springer Verlag, Wien, Austria. Hicks, J. F. 1993. Chronostratigraphic analysis of foreland basin sediments of the latest Cretaceous, Wyoming, USA. Ph.D. thesis, Yale University, New Haven, Connecticut, USA. Hofmann, C.-C. 2002. Pollen distribution in sub-recent sedimentary environments of the Orinoco Delta (Venezuela)—An actuo-palaeobotanical study. Review of Palaeobotany and Palynology 119: 191–217. November 2011] Grímsson et al.—Fossil Lythraceae pollen from the northern hemisphere 1815 Hofmann, C.-C., and R. Zetter. 2005. Reconstruction of different wetland plant habitats of the Pannonian Basin system (Neogene, Eastern Austria). Palaios 20: 266–279. Hofmann, C.-C., and R. Zetter. 2007. Upper Cretaceous pollen from the Vilui Basin, Siberia: Circumpolar and endemic Aquilapollenites, Manicorpus, and Azonia species. Grana 46: 227–249. Hofmann, C.-C., and R. Zetter. 2010. Upper Cretaceous sulcate pollen from the Timerdyakh Formation, Vilui Basin (Siberia). Grana 49: 170–193. Huang, Y.-L., and S.-H. Shi. 2002. Phylogenetics of Lythraceae sensu lato: A preliminary analysis based on chloroplast rbcL gene, psaAycf3 spacer, and nuclear rDNA internal transcribed spacer (ITS) sequences. International Journal of Plant Sciences 163: 215–225. Hutchinson, J. 1973. The families of flowering plants, 3rd ed. Clarendon, Oxford, England. Johnson, L. A. S., and B. G. Briggs. 1984. Myrtales and Myrtaceae—A phylogenetic analysis. Annals of the Missouri Botanical Garden 71: 700–756. Koehne, E. 1881. Lythraceae monographice describuntur. Botanische Jahrbucher fur Systematik, Pflanzengeschichte und Pflanzengeographie 1: 142–157. Koehne, E. 1903. Lythraceae. In A. Engler [ed.], Das Pflanzenreich IV, 216, Heft 17. W. Engelmann, Germany. Kovar-Eder, J., B. Meller, and R. Zetter. 1998. Comparative investigations on the basal fossiliferous layers at the opencast mine Oberdorf (Koflach-Voitsberg lignite deposit, Styria, Austria; Early Miocene). Review of Palaeobotany and Palynology 101: 125–145. Linnaeus, C. 1753. Species plantarum: Exhibentes plantas rite cognitas, ad genera relatas, cum differentiis specificis, nominibus trivialibus, synonymis selectis, locis natalibus, secundum systema sexuale digestas. Tomus I. Laurentii Salvii, Stockholm, Sweden. Liu, C. Y. S., R. Zetter, D. K. Ferguson, and B. A. R. Mohr. 2007. Discriminating fossil evergreen and deciduous Quercus pollen: A case study from the Miocene of eastern China. Review of Palaeobotany and Palynology 145: 289–303. Liu, C. Y. S., R. Zetter, D. K. Ferguson, and C. Zou. 2008. Lagerstroemia (Lythraceae) pollen from the Miocene of eastern China. Grana 47: 262–271. Melchior, H. 1964. Myrtiflorae. In H. Melchior [ed.], A. Engler’s Syllabus der Pflanzenfamilien. 12. Auflage, Band 2. Gebrüder Bornträger, Berlin, Germany. Menke, B. 1976. Pliozäne und ältestquartäre Sporen- und Pollenflora von Schleswig-Holstein. Geologisches Jahrbuch A 32: 3–197. Morris, J. A. 2007. A molecular phylogeny of the Lythraceae and inference of the evolution of heterostyly. Ph.D. thesis. Kent State University, Kent, Ohio, USA. Muller, J. 1981. Fossil pollen records of extant angiosperms. Botanical Review 47: 1–142. Nebert, K. 1979. Die Lignitvorkommen Südostburgenlands. Jahrbuch der Geologischen Bundesanstalt 122: 143–170. Patel, V. C., J. J. Skarla, and P. H. Raven. 1984. Pollen characters in relation to the delimitation of Myrtales. Annals of the Missouri Botanical Garden 71: 858–969. Punt, W., P. P. Hoen, S. Blackmore, S. Nilsson, and A. Le Thomas. 2007. Glossary of pollen and spore terminology. Review of Palaeobotany and Palynology 143: 1–81. Rodriguez-de la Rosa, R., S. R. S. Cevallos-Ferriz, and A. SilviaPineda. 1998. Paleobiological implications of Campanian coprolites. Palaeogeography, Palaeoclimatology, Palaeoecology 142: 231–254. Rögl, F., and G. Daxner-Höck. 1996. Late Miocene Paratethys correlations. In R. L. Bernor, V. Fahlbusch, and H.-W. Mittmann [eds.], The evolution of Western Eurasian Neogene mammal faunas. Columbia University Press, New York, New York, USA. Samoilovitch, S. R. 1965. The description of new pollen species of the Upper Cretaceous angiospermic flora. Trudy VNIGRI 239: 121–141 (in Russian). Samoilovitch, S. R., and N. D. Mchedlishvili. 1961. Pollen and spores of western Siberia. Jurassic to Palaeocene. Trudy VNIGRI 177 (in Russian). Shi, S., Y. Huang, F. Tan, X. He, and D. E. Boufford. 2000. Phylogenetic analysis of the Sonneratiaceae to Lythraceae based on ITS sequences of nrDNA. Journal of Plant Research 113: 253–258. Spicer, R. A., A. Ahlberg, A. B. Herman, C.-C. Hofmann, M. Raikevich, P. J. Valdes, and P. J. Markwick. 2008. The Late Cretaceous continental interior of Siberia: A challenge for climate models. Earth and Planetary Science Letters 267: 228–235. Sytsma, K. J., A. Litt, M. L. Zihra, J. C. Pires, M. Nepokroeff, E. Conti, J. Walker, and P. G. Wilson. 2004. Clades, clocks, and continents: Historical and biogeographical analysis of Myrtaceae, Vochysiaceae, and relatives in the southern hemisphere. International Journal of Plant Sciences 165 (suppl): S85–S105. Thorne, R. F. 1981. Phytochemistry and angiosperm phylogeny, a summary statement. In D. A. Young and D. S. Seigler [eds.], Phytochemistry and angiosperm phylogeny, Praeger, New York, New York, USA. Thorne, R. F. 1992. An updated phylogenetic classification of the flowering plants. Aliso 13: 365–389. Tobe, H., S. A. Graham, and P. H. Raven. 1998. Floral morphology and evolution in Lythraceae sensu lato. In J. Owens and P. J. Rudall [eds.], Reproductive biology. Royal Botanic Gardens, Kew, Richmond, UK. Tobe, H., and P. H. Raven. 1983. An embryological analysis of Myrtales: Its definition and characteristics. Annals of the Missouri Botanical Garden 70: 71–94. Van Boskirk, M. C. 1998. The flora of the Eagle Formation and its significance for Late Cretaceous floristic evolution. Ph.D. thesis, Yale University, New Haven, Connecticut, USA. Van Campo, E. 1976. La flore sporopollenique du gisement Miocène terminal de Venta del Moro (Espagne). Ph.D thesis, University of Science and Techniques of Languedoc, Montpellier II, Montpellier, France. Wang, W.-M., and M. M. Harley. 2004. The Miocene genus Fupingopollenites: Comparison with ultrastructure and pseudocolpi in modern pollen. Review of Palaeobotany and Palynology 131: 117–145. Webb, D. A. 1967. Generic limits in European Lythraceae. Feddes Repertorium 74: 10–13. Wijmstra, T. A. 1969. Palynology of the first 30 meters of a 120 m deep section in northern Greece. Acta Botanica Neerlandica 18: 511–527. Zetter, R. 1987. Bemerkungen zur Mikroflora der Kohleschichten der Südostburgenländischen Schwelle. Biologisches Forschungsinstitut Burgenland Bericht 68: 159–166. Zetter, R. 1989. Methodik und Bedeutung einer routinemäßig kombinierten lichtmikroskopischen und rasterelektonenmikroskopischen Untersuchung fossiler Mikrofloren. Courier Forschungsinstitut Senckenberg 109: 41–50. Zetter, R. 2006. The Middle Eocene microflora of the Princeton chert of southern British Columbia (Canada). 7th European PalaeobotanyPalynology Conference, Prague. Program and abstracts, p. 163. Zetter, R., and L. J. Hickey. 2010. New evidence for Late Cretaceous angiosperm diversity. 8th European Palaeobotany-Palynology Conference, Budapest, Hungary. Program and abstracts, p. 261. Zetter, R., M. Weber, M. Hesse, and M. Pingen. 2002. Pollen, pollenkitt, and orbicules in Craigia bronnii flower buds (Tilioideae, Malvaceae) from the Miocene of Hambach, Germany. International Journal of Plant Sciences 163: 1067–1071.