Find research datasets worth reusing

opencc-by-4.0Feb 2019View details →

Figure 2. Comparison bar of statistic results SC 2015 – 2017

The theme of the Summer Reading and Creativity Campaign 2018, in which 148 members of the GLN participated, focused on “Favorite Data”. From June 20th to September 7th, children were called on to observe, record, comprehend, and report data through 46 different workshops tailored accordingly to innovative methodological approaches to pedagogy.

opencc-by-4.0Oct 2020View details →

SRGN: Pythia + Delphes pp --> tt-bar

A collection of datasets used in <a href="https://arxiv.org/abs/2010.03569">Parameter Estimation using Neural Networks in the Presence of Detector Effects</a> for the top quark mass extraction example. Sample code for reproducing the results is available on <a href="https://github.com/hep-lbdl/SRGN">GitHub</a>. Top quark pair production is generated using Pythia 8.230 and detector effects are modeled using Delphes 3.4.1 using the default CMS run card. One of the \(W\) bosons is forced to decay to \(\mu + \nu_\mu\) while the other \(W\) boson decays hadronically. Each event is recorded as a variable-length set of objects, consisting of jets, muons, and neutrinos. At simulation-level, the neutrino is replaced with the missing transverse momentum. Generator-level and simulation-level jets are clustered with the anti-\(k_t\) algorithm using \(R=0.4\) and are labeled as \(b\)-tagged if the highest energy parton inside the jet cone (\(\Delta R<0.5\)) is a \(b\) quark. Jets are required to have  \(p_T>20\) GeV and they can only be \(b\)-tagged if \(|\eta|<2.5\). Furthermore, jets overlapping with the muon are removed. Events are only saved if they have at least two \(b\)-tagged jets and at least two additional non \(b\)-tagged jets. Four observables are formed for performing the top quark mass extraction. First, the \(b\)-jet closest to the muon is labeled \(b_1\). Of the remaining \(b\)-tagged jets, the highest \(p_T\) one is labeled \(b_2\). The highest two \(p_T\)non \(b\)-tagged jets are labeled \(j_1\) and \(j_2\) . The four observables are given by: \(m_{b_1\mu\nu}\), \(m_{b_2\mu\nu}\), \(m_{b_1j_1j_2}\), and \(m_{b_2j_1j_2}\), where the four- momentum of the detector-level neutrino is determined by solving the quadratic equation for the \(W\) boson mass. Per event, the data is listed: \((m_{b_1\mu\nu}, m_{b_2\mu\nu}, m_{b_2j_1j_2}, m_{b_1j_1j_2}, M_t)\), where \(M_t\)denotes the top quark mass. Training Datasets: DCTR_Mt_train.npz contains two arrays, X and Y. X is an array of events and Y is 0 (1) if the jet was generated with default (non-default) top quark mass. For a non-default event (Y=1), the \(M_t\) in each list represents the value of the top quark mass that was used. Note that for a default event (Y=0) the \(M_t\) in each list is not the default top quark mass, but  \(M_t\) uniformly sampled in the same range as the Y=1 jets.  The top quark mass was uniformly sampled in \(M_t\in[170.0, 180.0]\). Test Datasets: The test datasets are arrays made up of the same observables discussed above but with the top quark mass removed (i.e. \((m_{b_1\mu\nu}, m_{b_2\mu\nu}, m_{b_2j_1j_2}, m_{b_1j_1j_2})\). SRGN_Mt_default.npz was generated with the default top quark mass, \(M_t=172.5\) GeV, and SRGN_Mt_unknown.npz was generated with \(M_t=175.0\) GeV.

Dataset and supplementary files - Behavioral response of chub (Squalius cephalus), barbel (Barbus barbus) and brown trout (Salmo trutta) to pulsed direct current electric fields and resulting optimal waveform for use at electrified bar racks

Behavior Library.zip: For each species and behavior observed during the experiments an exemplary video is provided. Behavior_all.pdf: Additional plots showing the thresholds for the first time each individual behavior was observed for all fish species and tested waveformsSpecies.pdf: Additional plot allowing direct comparison of observed thresholds for the tested species when subjected to different waveforms. data.csv: All data necessary to reevaluate the conducted experiments. The dataset consists of<ul><li>Experiment ID</li><li>waveform - indicating the set of electrical parameters used</li><li>fish species and fish id </li><li>behavior - observed behavior</li><li>time from and time to - time in s after the start of the experiment that a behavior was started and ended respectively</li><li>type - point or interval referring to whether a behavior is considered instantaneous or continuous</li><li>voltage - applied voltage at the start of the given behavior</li><li>experiment_timestamp - date and time of the start of the experiment</li><li>breathing rate start - breathing rate at the start of the experiment</li><li>water  conductivity - water conductivity at a reference temperature of 25°C [muS/cm]</li><li>water temperature [°C]</li><li>breathing rate end - breathing rate at the end of the experiment</li><li>meta behavior - assigned category of meta behavior based on the observe behavior category</li><li>standard length, total length and height - standard length, total length and height of the tested fish in [mm]</li><li>volume - calculated fish volume based on the measured length and height and an assumed elliptical form of the fish</li><li>Fangdatum - Date of catch</li><li>t.Pulse - pulse length of the tested waveform [ms]</li><li>Frequency - Frequency of the tested waveform</li><li>N.Pulses.Group - Number of pulses per group of pulses for the waveform pattern</li><li>t.Gap - time between two pulses within a group of pulses [ms]</li><li>DutyCycle - Percentage of time current is flowing for a given waveform. Calculated based on the waveform parameters</li><li>usage - first, second or third time a fish was used in the experiments. </li><li>field strength - field strength at the time of this behavior calculated based on the applied voltage</li><li>c_w  ambient water conductivity [muS/cm]</li><li>p_d - power density calculated based on the field strength and the ambient water conductivity</li><li>p_t - power transferred to the fish calculated based on the field strength, the ambient water conductivity and an assumed conductivity of the fish of 115 muS/cm</li></ul>

opencc-by-4.0Sep 2024View details →

Time-lapse electrical resistivity tomography and seismic reflection imaging of a shallow ground-water aquifer (0-50 m): Mississippi River levee seepage across the Duncan Point bar, Baton Rouge, Louisiana, U.S.A.

The electrical resisitivity raw data files are slightly processed to remove bad data points but can be inverted using tomographic inversion code.  The seismic data were assembled in Seismic Unix format, a shortened version of the SEG-Y format (Society of Exploration Geophysicists Exchange Format-Y https: //seg. org/Publications/SEG-Technical-Standards), that has the 3200-byte EBCDIC and 400-byte tape header removed. The data uploaded online (<a href="https://zenodo.org/records/14776025">https://zenodo.org/records/14776025</a>) is a CMP brute-stacked seismic section.   During data collection, shotpoint location changed proceeding along a 136-degree azimuth (south-easterly direction), and spaced every 1 m. A total of 48, horizontal-component 28-Hz nominal geophones were placed every one meter and shotpoints were located half-way between geophones. Geophones remained fixed at their locations throughout the survey and so the CMP spacing is nominally 0.5-m but fold varies linearly from a value of 1 from either side of the survey to a central maximum of 24.  The seismic source consisted of a partially buried 20-lb steel I-beam struck repeatedly on either side three times by an 8-lb sledge hammer.  Data of the same striking polarity were added in-phase in the field.  Data with opposing polarity at each shotpoint location were subtracted later to enhance SH-wave data and suppress converted SH-to-P waves. Seismic processing is minimal and consists of standard surface-wave muting, elimination of bad seismic traces, normal moveout, bandpass filtering (between 12 Hz and 50 Hz) and preliminary stacking with trace mixing every 3 CMPs.  The data were stacked with a single velocity throughout that ranged from 80 m/s (Vs) at 0.2 s, to 100 m/s at 0.35 s and reached 180 m/s at 0.5 s of two-way traveltime.

opencc-by-4.0Aug 2022View details →

Time series of Agulhas leakage from satellite altimetry. Update of Le Bars et al. 2014.

Agulhas leakage time series updated from Le Bars et al. 2014 and Kelly et al. 2016. Final results are stored in: Trp_track_twb122_barot30W.nc,Trp_track_twb20_barot30W.nc and Trp_track_twb198_barot30W.nc An example of analysis and comparison with older time series is in the Python 3.9 Jupyter notebook: AnalyseOutputs.ipynb (AnalyseOutputs.html to simply visualise) The code and intermediate results are also provided. See README.txt for instructions on how to use it. It was written in 2013 with the NCAR Command Language and Python 2.? but it can still run now with later version of NCL and with Python 2.7. References: Le Bars, D., Durgadoo, J. V., Dijkstra, H. A., Biastoch, A., and De Ruijter, W. P. M.: An observed 20-year time series of Agulhas leakage, Ocean Sci., 10, 601–609, https://doi.org/10.5194/os-10-601-2014, 2014. Kelly, K. A., K. Drushka, L. Thompson, D. Le Bars, and E. L. McDonagh (2016), Impact of slowdown of Atlantic overturning circulation on heat and freshwater transports, Geophys. Res. Lett., 43, doi:10.1002/2016GL069789.

opencc-by-4.0Aug 2023View details →

Bar graphs of DEMIX database tile classifications

The DEMIX database (Guth, 2023a) contains statistics from 6 test 1 arc second DEMs (ALOS, ASTER, CopDEM, FABDEM, NASADEM, and SRTM) compared to high resolution reference DEMs.  The database contains 236 DEMIX tiles (Guth and others, 2023) and forms the basis for the ranking of global DEMs in Bielski and others (2023). A K-means clustering of the database using MICRODEM (Guth, 2023b, 2023c), and an additional set of 4 land cover and landform classifications (Table 1) for the 236 DEMIX tiles computed the percentage of each DEMIX tile in each classification category.  Guth (2023d) has the raw data for the percentages of each category for each of the 236 tiles along with the K-means cluster assignments. This data set contains 3 figures for each of the 5 classification databases in Guth (2023d): <ul> <li>Bar graph of the category percentages for each of the test areas.  A composite version of these graphs is in Bielski and others (2023).</li> <li>Bar graph of the category percentages for each of the 236 test tiles.  These graphs are too large to include on a single page with readable legends.</li> <li>Legend for the classification</li> </ul>   References: Bielski, C.; López-Vázquez, C.; Grohmann, C.H.; Guth. P.L.; and the TMSG DEMIX Working Group, 2023. DEMIX Method Ranks COPDEM, and FABDEM as Top 1” Global DEMs: <a href="https://arxiv.org/abs/2302.08425v3">https://arxiv.org/abs/2302.08425v3</a> Guth, P. L., 2023a. DEMIX GIS Database Version 2 (2.0) [Data set]. Zenodo. <a href="https://doi.org/10.5281/zenodo.8062008">https://doi.org/10.5281/zenodo.8062008</a> Guth, P.L., 2023b. GIT-MICRODEM [Delphi source code, archived installation versions]. URL: https://github.com/prof-pguth/git_microdem  Guth, P.L., 2023c, MICRODEM: Open-source GIS with a focus on Geomorphometry [download latest Win64 executable and CHM help file] URL: <a href="https://microdem.org/">https://microdem.org/</a> Guth, P.L., 2023d, K-means clustering of the DEMIX data set (1.0) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.8283791 Guth, Peter L., Peter Strobl, Kevin Gross, & Serge Riazanoff. (2023). DEMIX 10k Tile Data Set (1.0) [Data set]. Zenodo. <a href="https://doi.org/10.5281/zenodo.7504791">https://doi.org/10.5281/zenodo.7504791</a>

opencc-zeroJun 2019View details →

IG. 6. — A, Trunk vertebra of Alsophis sp. 2 from Pointe du Helleux archaeological site (Square 2 – crab layer) located on Grande-Terre Island; B, trunk vertebra of Erythrolamprus juliae cf. copeae (Parker, 1936) from Sainte-Rose La Ramée archaeological site (US 2058) located on Basse-Terre Island. Abbreviations: cd., condyle; ct., cotyle; di., diapophysis; h. k., hemal keel; m. c., medial constriction; n. a., neural arch; n. s., neural spine; p. c., precondylar constriction; p. d., paracotylar depression; p. n., postero-medial notch of the zygantrum; pa., parapophysis; pz. f., prezygapophyseal facet; pz. p., prezygapophyseal process; s. d., subcentral depression; s. r., subcentral ridge; s. t., sub-cotylar tubercle; zs., zygosphene. Scale bars: 4 mm in Fossil dipsadid snakes from the Guadeloupe Islands (French West-Indies) and their interactions with past human populations

IG. 6. — A, Trunk vertebra of Alsophis sp. 2 from Pointe du Helleux archaeological site (Square 2 – crab layer) located on Grande-Terre Island; B, trunk vertebra of Erythrolamprus juliae cf. copeae (Parker, 1936) from Sainte-Rose La Ramée archaeological site (US 2058) located on Basse-Terre Island. Abbreviations: cd., condyle; ct., cotyle; di., diapophysis; h. k., hemal keel; m. c., medial constriction; n. a., neural arch; n. s., neural spine; p. c., precondylar constriction; p. d., paracotylar depression; p. n., postero-medial notch of the zygantrum; pa., parapophysis; pz. f., prezygapophyseal facet; pz. p., prezygapophyseal process; s. d., subcentral depression; s. r., subcentral ridge; s. t., sub-cotylar tubercle; zs., zygosphene. Scale bars: 4 mm

opencc-by-4.0Aug 2009View details →

Figure 1. Metatanais cylindricus Shiino 1952 female, lectotype A body, dorsal B body, lateral. Scale bar equals 1 in A new species of Metatanais Shiino, 1952 (Crustacea, Tanaidacea, Paratanaoidea) from Australian coral reefs, with a redefinition of the genus

Figure 1. Metatanais cylindricus Shiino 1952 female, lectotype A body, dorsal B body, lateral. Scale bar equals 1 mm.

opencc-by-4.0Dec 2017View details →

Morphodynamic stability of river and tidal bifurcations around bars tested in the Fast Flow Facility

Multithread rivers such as the Jamuna and Mekong have networks of channels and bars that change with every flood. Tidal systems such as the Scheldt, Humber and Columbia estuaries and short tidal basins in the Wadden Sea and in Florida, have perpetually changing and interacting channels and shoals formed by ebb and flood currents. Current models fail to forecast these natural dynamics, yet main channels are economically important shipping fairways, whilst shoal areas that emerge and submerge daily are ecologically valuable habitats. Human interference, changing river discharge and sealevel rise threaten all functions. Furthermore, there are strong indications that fairway deepening leads to reduced urban safety due to enhanced flow resistance by groynes in rivers and enhanced tidal range in estuaries (e.g. Bolla Pittaluga et al. 2015 in AWR, Seminara et al., in EH 2011). This enhances dike failure risk during low water level and flooding during high water level. We urgently need dynamic forecasting models to optimise management strategies for these multiple functions (Wang et al. 2012 in Ocean Coastal Manage., Coco et al. 2013 in Mar. Geol.). Here we target firstly river bifurcations and secondly the mutually evasive ebb- or flood-dominated channels that form around bars and are found in all sandy tidal systems in the world (van Veen 1950/2002 in J. R. Dutch Geograph. Soc.). The cause for the mutual evasion is still incompletely understood despite the fact that they also appear in our numerical model results and experiments (Canestrelli et al., in JGR 2010; Kleinhans et al. 2015 in JGR). The nodes where ebb and flood channels connect can be seen as asymmetric bifurcations where one channel is preferred during ebb and the other during flood. Such bifurcations are critical elements that partition flow and sediment through the channel network, govern bar merging and splitting and are locations where bed steps form in shipping lanes, as in river bifurcations. Stability and equilibrium configurations are mostly unknown for tidal bifurcations except for one recent theory (Wang et al in prep.). In particular, we have a fair understanding of the tidal dynamics, but this is incomplete for the morphodynamics, especially related to understanding the sediment division at the bifurcation. We take advantage of the better but yet incomplete understanding of river bifurcations. The stability of river bifurcations has been studied for two decades in fieldwork, experimentation, linear stability theory and numerical modelling (e.g. Wang et al. 1995, JHR, see review in Kleinhans et al. 2013, ESPL) and our recent theory (Bolla Pittaluga et al. 2015 in GRL) synthesises many of the earlier results as follows: In bedload-dominated rivers, symmetrical bifurcations are unstable and develop towards a highly asymmetrical division of discharge and sediment. The same is the case for suspended sediment-dominated rivers, but the theory predicts stable bifurcations for intermediate sediment mobility. However, there is very little data for conditions intermediate between low and high mobility rivers. Moreover, we have no idea whether bifurcations in reversing tidal flow are unstable for similar configurations and conditions as in rivers. Here we mean configurations that are entirely free of topographic forcings on the flow: straight channels split into two channels over some length and depth. Our objective was therefore to experimentally investigate bifurcation stability in a range of sediment mobilities in unidirectional flow and reversing tidal flow ceteris paribus.

opencc-by-4.0Oct 2020View details →

SWASH Model Files from Modeled Three-Dimensional Currents and Eddies on an Alongshore-Variable Barred Beach

This archive contains SWASH model input, MATLAB processing scripts, and the model output used to produce the figures in “Modeled Three-Dimensional Currents and Eddies on an A longshore-Variable Barred Beach." Support was provided by the Washington Royalty Research Fund, the National Science Foundation, the Office of Naval Research, a National Defense Science and Engineering Graduate Fellowship, a Vannevar Bush Faculty Fellowship, the United States Army Corps of Engineers, the United States Coastal Research Program, Sea Grant, and the WHOI Investment in Science Fund.   The model inputs, scripts, and data are contained in three zip files:  <ul> <li>model_input.zip: input files for all simulations presented in this paper</li> <li>model_output_processing.zip: MATLAB model output processing scripts.</li> <li>model_data.zip: model output used to produce the figures</li> </ul>   SWASH is an open source code and can be download at <a href="http://swash.sourceforge.net/home.htm">http://swash.sourceforge.net/home.htm</a>. Please contact C.M. Baker at cmbaker9@uw.edu with questions.

opencc-zeroNov 2020View details →

aquila (Whalen & Carter in Carter, Whalen & Guex, 1998), Orbuk-32; R, Pseudoeucyrtis busuangaensis (Yeh & Cheng, 1998), Orbuk-32; S, Farcus graylockensis Pessagno, Whalen & Yeh, 1986, Orbuk-32; T-U, Farcus sp. A, Orbuk-26; V, Farcus sp. B, Orbuk-27; W-X, Anaticapitula anatiformis (De Wever, 1982); W, Orbuk-32; X, Orbuk-38; Y, Saitoum sp. aff. S. triumphense Whalen & Carter in Carter, Whalen & Guex, 1998, Orbuk-32. Scale bar: A-C, K-L, T-U, W-Y, 100 µm; D-J, M-O, 120 µm; P-S, V, 150 µm. in Late Triassic to Early Jurassic radiolarian, conodont and ammonite assemblages from the Tavuscayiri block, Mersin Mélange, southern Turkey: Time constraints for the T/J boundary and sedimentary evolution of the southern margin of the northern Neotethys

aquila (Whalen & Carter in Carter, Whalen & Guex, 1998), Orbuk-32; R, Pseudoeucyrtis busuangaensis (Yeh & Cheng, 1998), Orbuk-32; S, Farcus graylockensis Pessagno, Whalen & Yeh, 1986, Orbuk-32; T-U, Farcus sp. A, Orbuk-26; V, Farcus sp. B, Orbuk-27; W-X, Anaticapitula anatiformis (De Wever, 1982); W, Orbuk-32; X, Orbuk-38; Y, Saitoum sp. aff. S. triumphense Whalen & Carter in Carter, Whalen & Guex, 1998, Orbuk-32. Scale bar: A-C, K-L, T-U, W-Y, 100 µm; D-J, M-O, 120 µm; P-S, V, 150 µm.

opencc-by-4.0Oct 2019View details →

Fig. 10. Middle Pleistocene Didacna species from the Hajigabul section. Scale bar 5 in Magneto-biostratigraphic age constraints on the palaeoenvironmental evolution of the South Caspian basin during the Early-Middle Pleistocene (Kura basin, Azerbaijan)

Fig. 10. Middle Pleistocene Didacna species from the Hajigabul section. Scale bar 5 mm. A-B. Didacna bergi (1954 m, early Khazarian); C-G. D. parvula (1795 m, Late Bakunian); H-I. D. cf carditoides (1795 m, Late Bakunian); J-K. Didacna sp (1795 m, Late Bakunian).

opencc-by-4.0May 2012View details →

Fig.ç4.C aligus longiramus sp. nov., holotype, female (KMNH IvR 500,511). A, leg 2, dorsal view; B, leg 3, dorsal view; C, leg 4, dorsal view; D, leg 5, ventral view. Scale bars: 0.1 mm. in Occurrence of Caligid Copepods (Crustacea) in Plankton Samples Collected from Japan and Ŋailand, with the Description of a New Species

Fig.ç4.C aligus longiramus sp. nov., holotype, female (KMNH IvR 500,511). A, leg 2, dorsal view; B, leg 3, dorsal view; C, leg 4, dorsal view; D, leg 5, ventral view. Scale bars: 0.1 mm.

opencc-by-4.0May 2012View details →

Fig.ç3.C aligus longiramus sp. nov., holotype, female (KMNH IvR 500,511). A, maxilla, dorsal view; B, maxilliped, dorsal view; C, sternal furca, dorsal view; D, leg 1, ventral view; E, exopod of leg 1 enlarged, ventral view. Scale bars: 0.1 mm. in Occurrence of Caligid Copepods (Crustacea) in Plankton Samples Collected from Japan and Ŋailand, with the Description of a New Species

Fig.ç3.C aligus longiramus sp. nov., holotype, female (KMNH IvR 500,511). A, maxilla, dorsal view; B, maxilliped, dorsal view; C, sternal furca, dorsal view; D, leg 1, ventral view; E, exopod of leg 1 enlarged, ventral view. Scale bars: 0.1 mm.

opencc-by-4.0May 2012View details →

Fig.ç2.A, Caligus latigenitalis Shiino, 1954, male (KMNH IvR 500, 510), habitus, dorsal view; B–F. Caligus longiramus sp. nov., holotype, female (KMNH IvR 500, 511): B, habitus, dorsal view; C, caudal rami, dorsal view; D, antennule, ventral view; E, antenna, postantennal process, and maxillule, ventral view; F, mandible. Scale bars: 1 mm (A, B); 0.1 mm (C–F). in Occurrence of Caligid Copepods (Crustacea) in Plankton Samples Collected from Japan and Ŋailand, with the Description of a New Species

Fig.ç2.A, Caligus latigenitalis Shiino, 1954, male (KMNH IvR 500, 510), habitus, dorsal view; B–F. Caligus longiramus sp. nov., holotype, female (KMNH IvR 500, 511): B, habitus, dorsal view; C, caudal rami, dorsal view; D, antennule, ventral view; E, antenna, postantennal process, and maxillule, ventral view; F, mandible. Scale bars: 1 mm (A, B); 0.1 mm (C–F).

dryad40/100

Characterizing juvenile dispersal dynamics of invasive barred owls: implications for management

Characterizing natal dispersal can help manage the spread of invasive species expanding their ranges in response to land use and climate change. The Barred Owl (Strix varia) is a prominent example of an apex predator undergoing a rapid range expansion, having spread from eastern to western North America where it is now hyperabundant—threatening the Northern Spotted Owl (S. occidentalis caurina) with extinction and potentially endangering many other native species. We attached satellite tags to 31 Barred Owl juveniles at the southern leading edge of the Barred Owl's expanding range in California to characterize natal dispersal patterns and inform management. Juveniles traveled up to 100km from natal territories and experienced high mortality (annual survival = 0.204). At landscape scales, juveniles preferentially used forests, shrublands, and lower elevations during dispersal and avoided grasslands and burned areas. At finer scales, juveniles preferred shorter (younger) forests, lower elevations, and drainages, and avoided unforested areas. Our results suggest the Barred Owl range expansion is being driven primarily by high reproductive rates and densities despite low juvenile survival rates and dispersal through putatively suboptimal younger forests as a result of exclusion from high-quality habitat by territorial individuals. These findings also point to several strategies for conserving Spotted Owls and other native species in the Barred Owl's expanded range, including: (1) creating and maintaining Barred Owl-free reserves bounded by open or high-elevation areas; (2) creating reserves large enough to reduce immigration by long distance dispersers; and (3) removing Barred Owls from large riparian corridors.

opencc-zeroNov 2023View details →

Stochastic Modelling of Thin Mud Drapes inside Point Bar Reservoirs with ALLUVSIM-GANSim

Here is the dataset and code for the paper by Hu, X et al. (2023, under review). Stochastic Modelling of Thin Mud Drapes inside Point Bar Reservoirs with ALLUVSIM-GANSim, Water Resources Research.

Text-fig. 8. Ulmaceae a–e, Rosaceae f–l. a: Cedrelospermum, leaflet missing base but showing simple teeth, UAPC-ALTA S 25748. b: Ulmus with prominent teeth, asymmetric base, BBM-PAL-P000015. c: Ulmus leaf with prominent teeth, UAPC-ALTA S 59506. d: Ulmus fruit, UAPC-ALTA S 6590. e: Basal half of Ulmus leaf showing the obtuse teeth with subbasal entry of veins to the teeth, UAPC-ALTA S 59517. f: cf. Rubus, UAPC-ALTA S 67691. g. Prunus leaf, UAPC-ALTA S 25746. h: cf. Hesperomeles UAPC-ALTA S 67692. i: Photinia pagae, UAPC-ALTA S 67693. j: cf. Prunus, detail from g. k: Rosaceous leaf, UAPC-ALTA S 59501. l: Rosaceae, possibly Kerriae. BBM-PAL-P000047. Scale bars: a, f, i, k, l = 2 cm; b–e, g, h, j = 1 cm. in The Early Eocene Flora Of Horsefly, British Columbia, Canada And Its Phytogeographic Significance

Text-fig. 8. Ulmaceae a–e, Rosaceae f–l. a: Cedrelospermum, leaflet missing base but showing simple teeth, UAPC-ALTA S 25748. b: Ulmus with prominent teeth, asymmetric base, BBM-PAL-P000015. c: Ulmus leaf with prominent teeth, UAPC-ALTA S 59506. d: Ulmus fruit, UAPC-ALTA S 6590. e: Basal half of Ulmus leaf showing the obtuse teeth with subbasal entry of veins to the teeth, UAPC-ALTA S 59517. f: cf. Rubus, UAPC-ALTA S 67691. g. Prunus leaf, UAPC-ALTA S 25746. h: cf. Hesperomeles UAPC-ALTA S 67692. i: Photinia pagae, UAPC-ALTA S 67693. j: cf. Prunus, detail from g. k: Rosaceous leaf, UAPC-ALTA S 59501. l: Rosaceae, possibly Kerriae. BBM-PAL-P000047. Scale bars: a, f, i, k, l = 2 cm; b–e, g, h, j = 1 cm.

Text-fig. 5. Lauraceae, Platanaceae, Cercidiphyllaceae/Trochodendraceae. a: Sassafras hespera with 2 lobes, UAPC-ALTA S6556. b: cf. Lindera leaf. UAPC-ALTA S 67687. c: Macginitiea gracilis, UAPC-ALTA S 25748. d: Macginicarpa capitulum showing florets grouped in fives, UAPC-ALTA S 59507. e, g: Platanaceous fruitlets with basal tufts of dispersal hairs, UAPC-ALTA S 25748B, S S275238. f: Macginicarpa infructesence with five attached capitula, UAPC-ALTA S 59507A. h: Leaf similar to Populus and Trochodendroides, BBM-PAL-P000010. i: Leaf similar to Populus and Trochodendroides, UAPC-ALTA S 59516. j: cf. Trochodendroides, UAPC-ALTA S 59516. k: Jenkinsella infructesence; Figured in Penhallow 1908, plate 33. l: cf. Leaf similar to Cercidiphyllum and Trochodendroides, BBM-PAL-P000010. Scale bars: a–c, f, h, j, l = 2 cm, d, i, k = 1 cm, e, g = 0.5 cm. in The Early Eocene Flora Of Horsefly, British Columbia, Canada And Its Phytogeographic Significance

Text-fig. 5. Lauraceae, Platanaceae, Cercidiphyllaceae/Trochodendraceae. a: Sassafras hespera with 2 lobes, UAPC-ALTA S6556. b: cf. Lindera leaf. UAPC-ALTA S 67687. c: Macginitiea gracilis, UAPC-ALTA S 25748. d: Macginicarpa capitulum showing florets grouped in fives, UAPC-ALTA S 59507. e, g: Platanaceous fruitlets with basal tufts of dispersal hairs, UAPC-ALTA S 25748B, S S275238. f: Macginicarpa infructesence with five attached capitula, UAPC-ALTA S 59507A. h: Leaf similar to Populus and Trochodendroides, BBM-PAL-P000010. i: Leaf similar to Populus and Trochodendroides, UAPC-ALTA S 59516. j: cf. Trochodendroides, UAPC-ALTA S 59516. k: Jenkinsella infructesence; Figured in Penhallow 1908, plate 33. l: cf. Leaf similar to Cercidiphyllum and Trochodendroides, BBM-PAL-P000010. Scale bars: a–c, f, h, j, l = 2 cm, d, i, k = 1 cm, e, g = 0.5 cm.