Detailed palaeontologic and taphonomic techniques to reconstruct an earliest Paleocene fossil flora: An example from southwestern North Dakota, USA

Review of Palaeobotany and Palynology 151 (2008) 136–146 Contents lists available at ScienceDirect Review of Palaeobotany and Palynology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / r ev p a l b o Detailed palaeontologic and taphonomic techniques to reconstruct an earliest Paleocene fossil flora: An example from southwestern North Dakota, USA Antoine BERCOVICI a,⁎, Jacqueline WOOD b, Dean PEARSON c a b c UMR 6118 du CNRS, Géosciences Rennes, Bat. 15-Université de Rennes 1, Campus de Beaulieu, 35042 Rennes Cedex, France Delgado Community College, Department of Science and Math, 615 City Park Avenue, New Orleans, LA 70119, USA Pioneer Trails Regional Museum, Paleontology Department, 12 First Avenue NE, Box 78, Bowman, ND 58623, USA A R T I C L E I N F O Article history: Received 6 October 2007 Received in revised form 3 February 2008 Accepted 15 March 2008 Available online 7 April 2008 Keywords: Cretaceous/Tertiary boundary Fort Union formation fossil flora leaf mat paleoenvironment taphonomy A B S T R A C T A stratigraphic section in the basal Fort Union Formation (Paleocene) in southwestern North Dakota was used to study in detail the post-crisis recovery as well as to reconstruct the local environment and its evolution using sedimentology, palynology, fossil floras and vertebrate data. This report will focus on the flora from this site, corresponding to the first appearance data for Paleocene floral recovery, just above the Cretaceous– Tertiary (K/T) Boundary. The studied flora consist of an assemblage of tightly stacked leaves preserved as carbon imprints (also called leaf mats), a preservation condition that makes the extraction of each individual leaf difficult to achieve directly on site. As a result, a new technique was tested, allowing the study of every leaf preserved, their sedimentological context as well as their position relative to each other. A large block of matrix including the leaf mat was jacketed in plaster and was taken to the museum laboratory for analysis under controlled conditions. Preparation consisted of removing sediments at a millimeter scale and recording of placement and orientation of all fossil materials for three-dimensional reconstructions. Using this technique, a description and census of more than 300 leaf specimens was possible within an area of only 0.5 m2. The general sedimentological context indicates that the leaves were deposited in a near-stream environment associated with short-term flood events. Detailed information on depositional environment was gathered both by cutting a stratigraphic column from the Hell Creek/Fort Union formational contact up through the basal 4 m of the Fort Union Formation and by studying sediments and leaf preservation mode in detail within the leaf mat. Significant changes in taxonomic abundances correlated with different lithologies was observed, and a leaf species new to the study area was reported. The new methodology proves to be an efficient way to recover additional taphonomic and paleoenvironmental information from leaf mats necessary to understand the depositional dynamics of a fossiliferous leaf site, as well as to improve the record of taxonomic census. In a biostratigraphical prospective, the specimens recovered represent a low-diversity Fort Union flora composed exclusively of dicots that do not exist in the Hell Creek Formation. Preliminary palynological analysis reveal a Cretaceous age for the entire stratigraphic section, implying that the studied leaf mat is part of the FU0 megafloral zone (as defined by the occurrence of a Fort Union flora with Cretaceous palynomorphs). However, this Cretaceous age attribution for the entire section is questioned due to the occurrence of Paranymphaea crassifolia (part of the Paleocene FUI megafloral zone) within another leaf mat located 266 cm above the base of the coal representing the formational contact, and the occurrence of Paleocene PU1 mammals, reported from sediment within the leaf block interval. © 2008 Elsevier B.V. All rights reserved. 1. Introduction In southwestern North Dakota, leaf fossils found in the Hell Creek and Fort Union Formations are usually preserved in poorly lithified sediments or occur within thin mud laminae that are extremely hard ⁎ Corresponding author. E-mail address: antoine.bercovici@univ-rennes1.fr (A. BERCOVICI). 0034-6667/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.revpalbo.2008.03.004 to collect. Large or entire specimens usually require extensive excavation where blocks of matrix are removed using pick axes and/ or rock hammers. Once extracted, these blocks of matrix are split along bedding planes and trimmed down in the field to extract entire or identifiable fragments of specimens, which are then censused, carefully wrapped, and returned to the lab for further identification and storage (K. R. Johnson, pers. com.). This process is commonly used for collecting identifiable partial and entire specimens. However, the integrity of leaf assemblages cannot be fully described, as specimens A. BERCOVICI et al. / Review of Palaeobotany and Palynology 151 (2008) 136–146 130 V02017 120 Edmonton 100 110 513 Vancouver Marmarth Montana Québec Montréal Minnesota Ottawa Bismarck Wisconsin Rhame Toronto South Dakota Idaho L 512 North Dakota Oregon itt le Be av er Regina Washington BURL ING TON NOR THERN SANTA FE Wyoming 40 Nevada Spring Lake CE DA RH ILL 50 Winnipeg on Cr eek SLOPE COUNTY BOWMAN COUNTY Cr e ek 12 90 C A N A D A 50 Ba c 137 40 Iowa Nebraska Utah Co yo te C re S 70 ek Williston ILLS Sprin gC ree k ri Riv MEDICINE ek le M is s ou bo Cre Skull C reek er Ho rse RATTLESNAKE C re ek BUTTES 510 30 Dickinson Fargo Bismarck SLOPE Litt Gum Grand Forks POLE H 511 SUNSET BUTTE Big Minot 100 Kilometer BOWMAN 90 CRETACEOUS MUD BUTTES 5 Kilometer 58 59 500 Kilometer 80 Hell Creek Formation TERTIARY Sentinel Butte Formation QUATERNARY White River Formation Coleharbor Group Fox Hills Formation Bullion Creek Formation Golden Valley Formation Quaternary/Holocene Pierre Formation Fort Union Formation 60 Fig. 1. Map of the United States highlighting the state of North Dakota, general geology of the southwestern corner of the state, and position of the site. from multiple depositional environments may be incorporated into the same census without being noticed. A new technique for recovering the details of fossil leaf assemblages was tested on the earliest Paleocene fossil flora found at the studied site. Applying collection procedures usually employed for fossil vertebrates, where the object of study is jacketed in plaster and returned to the lab for analysis, allowed for the description of the leaf assemblage with the greatest precision possible. Using this process, larger sized blocks of matrix can be collected from which individual depositional environments and associations can also be described. This new methodology was tested to characterize the earliest Paleocene flora found directly above the Cretaceous/Tertiary (K/T) boundary at the studied site. 1.1. Geological settings Southwestern North Dakota exposes extensive outcrops of the continental K/T boundary (Hartman, 2002). These outcrops are aligned along the Cedar Creek anticline, in the valley of the Little Missouri River (Fig. 1). The K/T boundary lies in close proximity to the contact between the Hell Creek and the Fort Union Formations. In the study area (covering Bowman and Slope Counties), the Hell Creek Formation is approximately 100 m, thick, consisting of cross-bedded sandstones from stream channels, rooted siltstones representing point bar deposition and near-stream environments, and floodplains consisting of darker gray to green mudstones (Johnson, 1989; Fastovsky, 1987). Thin, discontinuous lignitic coals and carbonaceous shales are present, but rare. This formation usually displays darker, more uniform colors in grayish tones than the overlying Fort Union Formation. The overlying Fort Union Formation is also approximately 100 m thick, and consists of numerous lignitic beds, carbonaceous shales, large scale coarse sandstone bodies, and ponded-water deposits represented by variegated beds containing ironstone, siltstone, sandstone, and mudstone sequences seen since lightercolored deposits when compared to the underlying Hell Creek Formation (Johnson, 1989; Fastovsky, 1987). The contact between the Hell Creek and Fort Union Formations is routinely used within this region as an important field datum as it can be easily recognized and is traceable over long distances (Johnson, 1992; Pearson et al., 2002). However, it does not always display the same recognizable lithologic features from one area to another: any one or more of the lithologic indicators of the formational contact such as color change, a break in slope, the presence of a contact coal, more defined bedding planes, or the presence or absence of modern vegetation can be used to distinguish the field datum (Murphy et al., 2002). A combination of these criteria can be seen in Fig. 2. Modern erosion generally covers the coal bed usually found at the formational contact, which when present, has a thickness ranging from 1 cm to more than 50 cm. Recognition of the formation contact datum is essential, since it lies in close proximity to the palynologically defined K/T boundary (Johnson and Hickey, 1990; Johnson, 1992; Pearson, 1997). In the study area, the boundary is reported as occurring somewhere within the basal 4 m of the Fort Union Formation, or it may be coincident with the formational contact (Nichols, 2002; Nichols and Johnson, 2002). This observation implies that the formation contact is diachronous. Our study concentrates on a single site that is located within the lithological units of a stratigraphic column that encompasses the formation contact. This stratigraphic column produced the first megafloral record of the Fort Union Formation found during subsequent excavation and collection of Pioneer Trails Regional Museum (PTRM) vertebrate site V02017. A detailed stratigraphic section for the site is presented in Fig. 3, associated with palaeontological data for first and last appearances of biostratigraphically indicative taxa. 1.2. Biostratigraphical context and previous paleobotanical studies Abundant and well-characterized floras were recovered from the Hell Creek and Fort Union Formations in southwestern North Dakota and surrounding areas within the Williston Basin (K. R. Johnson, pers. com.). Synthesis work on these floras conducted by Johnson and Hickey (1992), identified megafloral biozonation within the Hell Creek 138 A. BERCOVICI et al. / Review of Palaeobotany and Palynology 151 (2008) 136–146 Fig. 2. View of the Hell Creek Fort Union Formation contact in the study area. Black arrows point to the formational contact. Typical features such as color change (especially from gray to golden yellow), a break in slope, more defined bedding planes within the Fort Union Formation and more color banding, and the presence of modern vegetation just above the contact. Formation (zones HCI–HCIII) and the overlying Fort Union Formation (zone FUI). Due to the distinctly different nature of the Hell Creek (HC) versus Fort Union (FU) floral components and their stratigraphic placement, HC floras were assigned a Cretaceous age and FU floras a Paleocene age. However, redefinition of the biozonation as well as the addition of a new FU0 zone was later proposed (Johnson, 2002; Nichols and Johnson, 2002) when FU floras were found below the palynologically defined K/T boundary, identifying for the first time a Fort Union flora of Cretaceous age. Floras found at our site will be attributed to one of the FU zones according to this definition. 2. Materials and methods Eight horizons were found to produce leaf megafossils on the site, two of them being preserved as assemblages of tightly stacked leaves preserved as carbon imprints. This type of preservation does not allow any practical way to determine leaf density in the field without potentially compromising the integrity of specimens and associated data. Specimens are so fragile that any rain fall, wind, or sunlight may damage the fossil imprints and their context in a short period of time. Extracting specimens that are associated in leaf mats is also extremely difficult to undertake in the field, since the fine-grained and poorly consolidated rock does not allow for specimens to be preserved if the small layers are taken apart. Only extracting large blocks can provide enough stability to keep specimens from cracking, however no detailed observations can be made when extracting large blocks, as they may contain multiple lithologic units and possibly different depositional environments. In order to observe the fine-scale resolution, a new excavation and preparation method was implemented in the collection process. 2.1. Site V02017 This site was located after the discovery of a large turtle (Axestemys sp.) exposed on the surface, at a close proximity above the formational contact. It was decided to undertake an in-depth study of this site, in order to reconstruct the depositional environment as well as the occurrence and evolution of the fossil fauna across this critical interval. A broad, 1 m wide and 1 m deep stratigraphic column (PTRM designation #156) was excavated, exposing the lowermost 3.7 m of the Fort Union Formation. The sedimentological context described within this column is summarized by the log presented in Fig. 3. Lithological units were numbered in an upward sequence from the formation contact coal, with separations based on noticeable sedimentological changes (this numbering system is used here only as a convenient way for referencing lithological units and is not used as a correlation system between PTRM localities). An extensive campaign of wet-sieving of the sediments was undertaken and has yielded a collection of several thousand vertebrate specimens from the basal Fort Union Formation to date (Pearson et al., 2004). During the controlled excavation process, the eight leaf bearing horizons where identified, but no convenient way of analyzing leaf associations was possible on site when high densities were found. Among these fossiliferous horizons is the lowermost FU fossil flora found at the site, which occurs at 133 cm above the formation contact, within lithologic units #9 and #10 (Fig. 4). 2.2. Excavation The site was excavated using a 1 m2 grid during the microvertebrate wet-sieving process. During this matrix collection process, flat surfaces were excavated 2 to 5 cm at a time to preserve the integrity of the lithologic unit. A 1 m2 area containing the entire lithological sequence of units #7 to #10 was selected for excavation and removal for study. Excavation and removal of the adjacent areas had indicated that horizon #9 and possibly #10 contained large number of fossil leaves (Fig. 5). Techniques developed primarily for large vertebrate fossils was applied to remove and transport the roughly 1 m2 block of matrix from the site. To do this, a trench was cut around the perimeter of the block with the use of hand tools. Excess sediment from the top of the block (the majority of unit #10) was removed to lighten the weight of the block. No leaves were found in this process, implying that the entire fossiliferous leaf layer was preserved within the block. Unit #7 was shown not to be fossiliferous during the search for vertebrate fossils, so trenching was stopped when the upper 20 cm of unit #7 was exposed. The block was then excavated so that a pedestal of sediment remained beneath it (Fig. 6) and encased in a jacket made of burlap and plaster that covered all exposed surfaces to minimize and contain any breakage. Metal rods were driven into the pedestal to enable a controlled cleavage along this horizon, and a hydraulic jack was used to A. BERCOVICI et al. / Review of Palaeobotany and Palynology 151 (2008) 136–146 Column #156 - V02017 Sandstone, coarse Sandstone, medium Sandstone, fine Siltstone Mudstone Cross stratification Roots Rip up clasts Paleontology Flora Fauna Sedimentological events Terrestrial abd aquatic vertebrates 26 Coal Lithological Units meters Formations Grain size 25 4 24 4 FUI flora 23 22 3 PU1 Land Mammal age FAD 19 18 17 3.1. Benefits of the new excavation technique for leaf mats Dissecting the block in flat layers following the stratigraphic surface allowed the observation of plant associations and taphofacies at the smallest scale possible. As a result, three main diversity zones were identified within the leaf mat, each with different paleobotanical content, preservation conditions and lithology. These identified zones “FU0” flora Fort Union 20 16 15 14 13 at a time. In some instances, leaf compressions were separated by layers of sediment not exceeding a few hundred microns. This excavation was undertaken on the entire surface of the block to maintain an even, level surface. Every new leaf exposed was photographed, mapped, and its depth determined relative to a reference datum represented by a level string placed across the block, and sedimentological context was recorded. Specimens were then removed carefully, one piece at a time, and re-assembled onto foam board and glued in place. Each specimen was given a separate specimen number for census. Unfortunately not all specimens were kept for collection as some were too fragile and disintegrated upon attempted removal. Photographic mapping allowed for precise 2D placement: a model of the block was drawn using Adobe Illustrator by placing each photograph in reference to pre-positioned landmarks according to the excavated block outlines and major cracks in the matrix. Once precisely replaced, the leaf was outlined and filled with translucent gray to build a density map. Fig. 8 represents the results of this reconstitution in stratigraphic sequence, from drawings 1 to 8. Drawing 9 represents the total density map of the leaf mat. To allow for details to be seen, drawings 1 to 8 were separated according to lithology and density to minimize overlap between leaves. In addition, a series of identified specimens are depicted in Plates I–III for verification. 3. Results 21 2 139 3 11 10 9 8 2 Studied interval 12 FAD FAD 7 1 6 1 Hell Creek 0 1 NOT SAMPLED 3 2 Aquatic flora 5 4 Fig. 3. Sedimentological log of section #156 on site V02017 with biostratigraphical data. lift the block and separate it from the ground (Fig. 7). Using a system of levers, the block was flipped over onto a piece of plywood and carried out of the field on a trailer pulled by an all-terrain vehicle, for transportation to the lab. 2.3. Analysis In the laboratory, layers of sediment was carefully removed using knives and dental tools. Only small layers of sediment were removed Fig. 4. Lithological section showing the context of the block that was extracted. 140 A. BERCOVICI et al. / Review of Palaeobotany and Palynology 151 (2008) 136–146 and third drawing). Taphofacies consists of entire leaves sometime preserved as curled imprints within the silty matrix. (3) The top zone beginning at the base of lithologic unit #10 is represented by mudstone with a very dense mat of large Platanus raynoldsii leaves. The leaf density gradually decreases as the lithology changes from mudstone to silt. (Fig. 8, 4th to 8th drawing). Most of the leaves exhibit signs of maceration, implying an under water residence prior to being covered by sediments, whereas very few reveal cracks or fragmentation. This can be explained by a short residence time of the leaves on the open ground. Likely, most of the leaves were rapidly transported by wind or surface runoff to be collected in a larger and more permanent aquatic transportation system (Bercovici and Broutin, 2008). Most were then deposited in areas of lowest water flow or standing water (Rich, 1989), represented by this zone of the leaf mat. Fig. 5. The leaf-producing horizon as seen when first excavated on the site. Scale is 15 cm. are listed below in stratigraphic sequence, starting at the base of the leaf mat: (1) From the base of the leaf mat (corresponding to the base of lithological unit #9), the first 5-cm-thick mudstone layer is composed of fragmented reeds, Platanus raynoldsii and Cornophyllum newberryi leaves and Cercidiphyllaceae fructifications (Fig. 8, first drawing). Taphofacies observed is corresponding to macerated and fragmented plant debris, mostly not identifiable, associated with often fragmented and smaller leaves. Water current action is obvious, and indication of paleocurrent is visible at the very base of lithological unit #9 where the reeds occur. Orientation of plant fragments, as well as the collection of mud clasts behind larger debris subjected to water action, indicate a current flow to the south (Fig. 9). (2) A 4-cm interval is present in the center of the leaf mat that consists of a siltier component at the top of lithologic unit #9. This zone is dominated by cf. “Rhamnus” goldiana, a leaf species originally described from the Denver Basin (Barclay et al., 2003; Johnson et al., 2003), but reported for the first time in this study area (Fig. 8, second Fig. 6. The block prepared to be jacketed in plaster after having been excavated with a pedestal of sediment remaining beneath. Information gathered from the study of plant taphofacies (Behrensmeyer, 1991; Burnham et al., 1992; Denko, 1995) are important for providing extra information about the depositional environment and the paleoenvironmental significance of an observed leaf assemblage that a sedimentological study alone could not entirely resolve (Behrensmeyer and Kidwell, 1985; Greenwood, 1991; Behrensmeyer et al., 2000). Here, the analysis of the succession indicates that major variations in percentages of taxa occur within the leaf mat. This observation has not been previously reported, as such changes are not readily visible in outcrop. These percentage of variations are reported in Table 1 and are easily seen on the separate drawings of Fig. 8, where the taxon composition is seemingly correlated with lithological changes within the leaf mat. The high density of leaves represented in the leaf mat suggests a relatively broad sampling area (in the order of 1 km2 for floodplain environments, Behrensmeyer et al., 2000) of the local environment by slow moving river systems and accumulation in places of lower energy. Leaf marceration is commonly seen, implying a residence under water prior to being covered by sediments. 3.2. Composition and nature of the flora The fossil flora identified and retrieved from the leaf block is a lowdiversity FU flora. However, the dominant angiosperms and taxodiaceous conifers that are previously reported from the basal Fort Union Formation (Johnson 1989; Johnson and Hickey 1990; Johnson 1992; and Johnson 2002) are not all present: The leaf mat from lithologic units #9 and #10 contains Platanus raynoldsii, Cornophyllum Fig. 7. The block was lifted using a hydraulic jack after being jacketed in plaster. A. BERCOVICI et al. / Review of Palaeobotany and Palynology 151 (2008) 136–146 newberryi, Cercidiphyllum seeds, Populus nebrascensis, cf. “Rhamnus” goldiana and miscellaneous unidentified reed fragments, but does not contain any “Cocculus” flabella, Dicotilophyllum anomalum, Quereuxia angulata, Paranymphaea crassifolia, “Lemna” scutana, Glyptostrobus europaeus, or Metasequoia occidentalis. These taxa, along with P. reynoldsii and P. nebrascensis are the previously reported 141 dominants of the FUI floral zone. Later, Johnson (2002) added the families Cercidiphyllaceae and Cornaceae to this list of dominants, which does appear at our site. These results seem to support the interpretation that Paleocene floral localities have a more variable and facies-related pattern of relative abundance than those of the Hell Creek Formation. 1 4 2 5 3 6 0 5 Fig. 8. Drawings 1 to 8: Actual position of excavated fossils in the leafblock separated by stratigraphic intervals. Drawing 9: Total density map. 25cm 142 A. BERCOVICI et al. / Review of Palaeobotany and Palynology 151 (2008) 136–146 3.3. Biostratigraphical implications With the exception of the three Nelumbo leaves found in unit #4 (Fig. 3) that are not biostratigraphically indicative, all leaf-producing horizons are identified as low-diversity Fort Union flora composed exclusively of dicots that do not exist in the Hell Creek Formation. A 7 preliminary palynological study was undertaken for the placement of the K/T boundary and showed that taxa known to disappear at the boundary in association with the extinction event (also referred to as K-taxa) occurs up to the top of the column in significant numbers (T. J. Kroeger, pers. com.), implying a Cretaceous age for the entire section. This also imply that floras found in unit #9 to unit #18 need to be considered as FU0 floras, or Fort Union floras of Cretaceous age. However, a major problem is raised at this site, as the first appearance of Paranymphaea crassifolia, marker of the FUI zone and indicative of a Paleocene age, appears at +266 cm above the formational contact, associated with Cretaceous palynomorphs (Fig. 3). Also, first appearance data for Paleocene PU1 mammals was recorded in the sediment of the leaf block interval (Hunter et al., 2003) at +133 cm (unit #9 to unit #12—Fig. 3). These two Paleocene markers are in conflict with the data given by preliminary examination of the palynological samples. A complete set of palynological samples was recollected and is being studied (Bercovici et al., 2007), results from this study will be part of an upcoming publication. 4. Discussion and conclusions 8 9 This new collection technique has allowed paleoenvironmental details to be studied at a level not previously investigated in the study area. Evidence of short-term changes in macrofossil flora diversity were demonstrated between the three diversity zones within the leaf mat, which were previously over-looked when sampling on a larger scale level. The origin of this variation can however be of multiple origins: Seasonality can be invoked, as the time averaging of paleobotalical assemblages on floodplain dominated environment can be as low as one year (Behrensmeyer et al., 2000). Transportation and depositional bias can also be proposed, as variation in taxon abundances also occur in conjunction with changes in lithology. Taxonomical censuses can be improved using this technique as well: In this particular example, issues that prevented the finding of “Rhamnus” goldiana with regular on-site excavation involved the well developed cleavage plane materialized by the Platanus bearing mudstone layer. Separation was thus preferentially made at this cleavage zone rather than in the more consolidated silty layer below, containing the large abundance of “R.” goldiana. On a biostratigraphical prospective, the interval studied at this site pointed out a major problem with the palynological placement of the K/ T boundary. The Cretaceous age indicated by the palynological analysis conflicts with the identification of the first appearance of the FUI floral zone as well as the first appearance of PU1 mammals. The occurrence of Cretaceous palynomorphs in association with younger elements may be attributed to reworking, but no process can explain the incorporation of younger fossils into older sediments. As a result of this major observation, placement of the K/T boundary on the site cannot be higher than the point of appearance of the first Paleocene marker, the PU1 mammals, at +133 cm above the base of the coal representing the formational contact. This also implies that the flora studied within the leaf mat, occurring in the same interval as the PU1 mammals, is Paleocene in age, and not part of the FU0 floral zone. Whether this problem in the palynological identification of the K/T boundary is the result of an anomaly occurring at this site or something significant at a regional scale should be investigated. This site and study was the first one to allow direct comparison of all biostratigraphical proxies (palynology, fossil floras and vertebrates) in a single sedimentological section at the K/T boundary. Acknowledgments 0 Fig. 8 (continued ). 5 25cm We gratefully acknowledge the valuable assistance of the following organizations and individuals: the East Marmarth Pasture Members and the Robert Brooks family for allowing us access to the study area; A. BERCOVICI et al. / Review of Palaeobotany and Palynology 151 (2008) 136–146 143 Plate I. 1 to 4 and 6, Cornophyllum newberryi—5 and 7, Platanus raynoldsii—8, Cornophyllum newberryi. This last specimen was not recovered from the block but was found in the leafproducing horizon. Large indentation of the margin resulting of insect feeding can be observed. Terry, Nancy and Blaine Schaefer, Kathy and Don Wilkening, Merle Clark, and the Pioneer Trails Regional Museum Interns for helpful discussions, and for assisting with sampling the stratigraphic columns and providing the heavy equipment necessary to extract and transport the block. John Hunter, Ohio State University, and Kirk Johnson, Denver Museum of Nature & Science for helpful comments and identifications 144 A. BERCOVICI et al. / Review of Palaeobotany and Palynology 151 (2008) 136–146 Plate II. 1 to 3 cf. “Rhamnus” goldiana—4 and 5, Platanus raynoldsii. A. BERCOVICI et al. / Review of Palaeobotany and Palynology 151 (2008) 136–146 Plate III. 1 and 3, Platanus raynoldsii—2, cf. “Rhamnus” goldiana with fragmented margin. 145 A. BERCOVICI et al. / Review of Palaeobotany and Palynology 151 (2008) 136–146 Burrows Current flow 146 Table 1 Quantitative data for leaf species and other plant materials found in each stratigraphical subgroups of Fig. 8 Unit #9 2-3° dip Rip up clasts Unit #8 Fig. 9. Horizontal surface showing the transition from unit #8 to unit #9. Rip-up clasts and elements indicate paleocurrent direction to the south, supporting evidence found from other sedimentological features in the stratigraphic column. of mammalian and fossil flora specimens. Last but not least, we thank the reviewers for thoughtful comments and a meticulous review that helped improve the manuscript. References Barclay, R.S., Johnson, K.R., Betterton, W.J., Dilcher, D.L., 2003. Stratigraphy and megaflora of a K–T boundary section in the eastern Denver Basin. Rocky Mountain Geology 38-1, 45–71. Bercovici, A., Broutin, J., (2008). The Autunian flora of the Martenet site (Blanzy-Montceau les Mines, Burgundy, France): Description, taphonomic study and palaeoenvironmental implications. Compte rendus de l'Académie des Sciences—Palévol 7, 1–16. Bercovici, A., Pearson, D.A., Nichols, D.J., Wood, J., 2007. Enhancing the recognition of the palynologically defined K/T boundary in southwestern North Dakota, USA: Using multiproxy datasets for paleoenvironmental reconstruction to minimize effects of taphonomic biases. Paléobotanique et évolution du monde végétal: quelques problèmes d'actualité. Collège de France, Paris, p. 46. Behrensmeyer, A.K., Kidwell, S.M., 1985. Taphonomy's contribution to paleobiology. Paleobiology, 11 (1), 105–119. Behrensmeyer, A.K., 1991. Terrestrial vertebrate accumulations. In: Allison, P.A., Briggs, D.E.G. (Eds.), Taphonomy: Releasing the Data Locked in the Fossil Record. . Topics in geobiology, vol. 9. Plenum, New York, pp. 291–335. Behrensmeyer, A.K., Kidwell, S.M., Gastaldo, R.A., 2000. Taphonomy and paleobiology. Paleobiology 26 (4), 103–147. Burnham, R.J., Wing, S.L., Parker, G.G., 1992. The reflection of deciduous forest communities in leaf litter: implications for autochtonous litter assemblages from the fossil record. Paleobiology 18 (1), 30–49. Denko, T.M., 1995, Taphonomy of fossil plants in the upper Triassic Chinle Formation. PhD thesis, University of Arizona, Tucson. 259 p. Fastovsky, D.E., 1987. Paleoenvironments of vertebrate-bearing strata during the Cretaceous–Paleogene transition, vol. 2. Palaios, eastern Montana and western North Dakota, pp. 282–295. Greenwood, D.R., 1991. The taphonomy of plant macrofossils. In: Donovan, S.K. (Ed.), Fossilization: The Processes of Taphonomy. Columbia University Press, New York, pp. 141–169. Hartman, J.H., 2002. Hell Creek Formation and the early picking of the Cretaceous– Tertiary boundary in the Williston Basin. In: Hartman, J.H., Johnson, K.R., Nichols, D.J. (Eds.), The Hell Creek Formation and the Cretaceous–Tertiary Boundary in the Northern Great Plains: An Integrated Continental Record of the End of the Cretaceous. Special Paper, vol. 361. Geological Society of America, pp. 1–7. Hunter, J.P., Pearson, D.A., Hartman, J.H., 2003. Additions to the Puercan mammals of North Dakota and a framework for the study of post-K/T vertebrate recovery. Society of Vertebrate Palaeontology 23, 64A Abstracts. Platanus raynoldsii % Cornophyllum newberryi % Populus nebraskensis % cf. “Rhamnus” goldiana % Cercidiphyllum seeds % Reeds % Sample size 1 2 3 4 5 6 7 8 16.9 26.0 5.2 1.3 2.6 48.0 77 11.1 79.6 0.0 9.3 0.0 0.0 54 30.0 48.0 12.0 8.0 0.0 2.0 50 75.0 25.0 0.0 0.0 0.0 0.0 20 100.0 0.0 0.0 0.0 0.0 0.0 19 100.0 0.0 0.0 0.0 0.0 0.0 14 77.7 22.3 0.0 0.0 0.0 0.0 9 100.0 0.0 0.0 0.0 0.0 0.0 2 Johnson, K.R., 1989, A high-resolution megafloral biostratigraphy spanning the Cretaceous–Tertiary boundary in the northern Great Plains, Ph.D. thesis, New Haven, Connecticut, 556 p. Johnson, K.R., 1992. Leaf-fossil evidence for extensive floral extinction at the Cretaceous–Tertiary boundary, North Dakota, USA. Cretaceous Research 13, 91–117. Johnson, K.R., 2002. Megaflora of the Hell Creek and lower Fort Union Formations in the western Dakotas: vegetational response to climate change, the Cretaceous–Tertiary boundary event, and rapid marine transgression. In: Hartman, J.H., Johnson, K.R., Nichols, D.J. (Eds.), The Hell Creek Formation and the Cretaceous–Tertiary Boundary in the Northern Great Plains: An Integrated Continental Record of the End of the Cretaceous. Special Paper, vol. 361. Geological Society of America, pp. 329–391. Johnson, K.R., Hickey, L.J.,1990. Megafloral change across the Cretaceous/Tertiary boundary in the northern Great Plains and Rocky Mountains, USA. In: Sharpton, V.L., Ward, P.D. (Eds.), Global Catastrophes in Earth History: An Interdisciplinary Conference on Impacts, Volcanism and Mass Mortality. GSA Sp. Pap., vol. 247, pp. 433–444. Johnson, K.R., Reynolds, M.L., Werth, K.W., Thomasson, J.R., 2003. Overview of the late Cretaceous, early Paleocene, and early Eocene megafloras of the Denver Basin. Rocky Mountain Geology 38-1, 101–120. Murphy, E.C., Hoganson, J.W., Johnson, K.R., 2002. Lithostratigraphy of the Hell Creek Formation in North Dakota. In: Hartman, J.H., Johnson, K.R., Nichols, D.J. (Eds.), The Hell Creek Formation and the Cretaceous–Tertiary Boundary in the Northern Great Plains: An Integrated Continental Record of the End of the Cretaceous. Special Paper, vol. 361. Geological Society of America, pp. 9–34. Nichols, D.J., 2002. Palynology and palynostratigraphy of the Hell Creek Formation in North Dakota: a microfossil record of plants at the end of Cretaceous time. In: Hartman, J.H., Johnson, K.R., Nichols, D.J. (Eds.), The Hell Creek Formation and the Cretaceous–Tertiary Boundary in the Northern Great Plains: An Integrated Continental Record of the End of the Cretaceous. Special Paper, vol. 361. Geological Society of America, pp. 393–456. Nichols, D.J., Johnson, K.R., 2002. Palynostratigraphy and microstratigraphy of Cretaceous–Tertiary boundary sections in southwestern North Dakota. In: Hartman, J.H., Johnson, K.R., Nichols, D.J. (Eds.), The Hell Creek Formation and the Cretaceous– Tertiary Boundary in the Northern Great Plains: An Integrated Continental Record of the End of the Cretaceous. Special Paper, vol. 361. Geological Society of America, pp. 95–143. Pearson, D.A., 1997. Searching for the last dinosaur: a systematic survey in the latest Cretaceous rocks of southwestern North Dakota. North Dakota Academic Science Proceedings 5, 213. Pearson, D.A., Schaefer, T., Johnson, K.R., Nichols, D.J., Hunter, J.P., 2002. Vertebrate biostratigraphy of the Hell Creek Formation in southwestern North Dakota and northwestern South Dakota. In: Hartman, J.H., Johnson, K.R., Nichols, D.J. (Eds.), The Hell Creek Formation and the Cretaceous–Tertiary Boundary in the Northern Great Plains: An Integrated Continental Record of the End of the Cretaceous. Special Paper, vol. 361. Geological Society of America, pp. 145–167. Pearson, D.A., Knauss, G., Wilkening, D., Wilkening, K., Bercovici, A., Schumaker, K., 2004. Comparing faunas and floras from the latest Cretaceous and early Paleocene in southwestern North Dakota: a microscopic analysis of environmental and taxonomic composition. Geological Society of America, Annual Meeting Abstracts 36 (5), 178. Rich, F.J., 1989. A review of the taphonomy of plant remains in lacustrine sediments. Review of Palaeobotany and Palynology 58 (1), 33–46.