This article was downloaded by: [Smithsonian Institution Libraries]On: 23 April 2012, At: 07:24Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

PalynologyPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/tpal20

Palynology of the Eocene Esmeraldas Formation,Middle Magdalena Valley Basin, ColombiaGuillermo Rodríguez-Forero a b , Francisca E. Oboh-Ikuenobe b , Carlos Jaramillo-Munoz c ,Milton J. Rueda-Serrano a d & Edwin Cadena-Rueda c ea Biostratigraphy Team, Instituto Colombiano del Petróleo-ECOPETROL S.A., Km. 7 viaPiedecuesta, Santander, Colombiab Department of Geological Sciences and Engineering, Missouri University of Science andTechnology, 129 McNutt Hall, Rolla, MO, 65409, USAc Smithsonian Tropical Research Institute, Unit 0948 APO AA 34002–0948, USAd Paleoflora LTDA, Bucaramanga, Colombiae Department of Marine, Earth and Atmospheric Sciences, North Carolina State University,Raleigh, NC, 27695, USA

Available online: 23 Feb 2012

To cite this article: Guillermo Rodríguez-Forero, Francisca E. Oboh-Ikuenobe, Carlos Jaramillo-Munoz, Milton J. Rueda-Serrano & Edwin Cadena-Rueda (2012): Palynology of the Eocene Esmeraldas Formation, Middle Magdalena Valley Basin,Colombia, Palynology, DOI:10.1080/01916122.2012.650548

To link to this article: http://dx.doi.org/10.1080/01916122.2012.650548

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form toanyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses shouldbe independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims,proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly inconnection with or arising out of the use of this material.

Palynology of the Eocene Esmeraldas Formation, Middle Magdalena

Valley Basin, Colombia

Guillermo Rodrıguez-Foreroa,b*, Francisca E. Oboh-Ikuenobeb, Carlos Jaramillo-Munozc,Milton J. Rueda-Serranoa,d and Edwin Cadena-Ruedac,e

aBiostratigraphy Team, Instituto Colombiano del Petroleo-ECOPETROL S.A., Km. 7 via Piedecuesta, Santander, Colombia;bDepartment of Geological Sciences and Engineering, Missouri University of Science and Technology, 129 McNutt Hall, Rolla, MO65409, USA; cSmithsonian Tropical Research Institute, Unit 0948 APO AA 34002–0948, USA; dPaleoflora LTDA, Bucaramanga,Colombia; eDepartment of Marine, Earth and Atmospheric Sciences, North Carolina State University, Raleigh, NC 27695, USA

The palynology of the Eocene Esmeraldas Formation in the Middle Magdalena Valley Basin, Colombia wasanalyzed in order to constrain the age of the unit. This formation is a very important oil reservoir in the MiddleMagdalena Valley Basin, which is a product of the fragmentation of a Cenozoic foreland basin by the uplift ofthe Northern Andes. The lateral continuity of the formation, as well as its correlations with lithostratigraphicunits in adjacent basins is not clearly understood. The Los Corros Fossil Horizon, a molluscan horizon in theupper part of the Esmeraldas Formation, has been used to trace the top of the formation. This horizon is notlaterally continuous over the basin and its age is still debatable. Data from 82 samples from an outcrop section inthe Nuevo Mundo Syncline area and from seven previously studied wells have been integrated with apalynological zonation of northern South America in order to date the Esmeraldas Formation. The age rangesfrom the late Early Eocene to the Late Eocene. The Esmeraldas Formation is correlative with the upper PicachoFormation and the lower part of the Concentracion Formation in the Eastern Cordillera, and the upper MiradorFormation and the base of the Carbonera Formation in the Llanos Foothills. The Los Corros Fossil Horizon isLate Eocene and is time-correlative with a marine transgression in the central Llanos Foothills. A non-metricmultidimensional scaling analysis suggests that floras from the Middle Magdalena Valley were different fromthose in the Llanos Foothills area during the Middle to Late Eocene. This is apparently due to taphonomiceffects. The results of this study will contribute to a better understanding of the overall evolution of the MiddleMagdalena Valley Basin.

Keywords:Middle Magdalena Valley Basin; Esmeraldas Formation; Eocene; palynology; biostratigraphy; Colombia

1. Introduction

The Middle Magdalena Valley Basin (MMVB) (Fig-ure 1) is an intramontane basin in Colombia. It islimited by the Eastern Cordillera to the east, theCentral Cordillera to the west, the San Lucas Ridge(Palestina fault) to the northwest and the Piedras–Girardot transpressive fold belt (Montes et al. 2005) tothe south and covers approximately 32,000 km2

(Morales et al. 1958). MMVB is part of a majorCenozoic foreland basin that has been fragmented bythe uplift of the Northern Andes (Gomez et al. 2005).Paleocene coastal to alluvial sediments (Lisama For-mation, Figure 2) were deposited in the foreland basinformed by the Central Cordillera uplift (Gomez et al.2005). Eocene–Neogene fluvial rocks (La Paz, Esmer-aldas, Mugrosa and Colorado formations and Realand Mesa Groups, Figure 2) are evidence of fragmen-tation due to uplifting of the Eastern Cordillera. TheMMVB has many stratigraphic and structural

complexities, with abundant syntectonic deposits andvery fast lateral facies and thickness changes (Gomezet al. 2005). Reliable stratigraphic models depends onbiostratigraphy for accurate lateral correlations bothat basinal and reservoir scale. Three fossil horizons,Los Corros, Mugrosa and Colorado, have beentraditionally used as basin-wide correlation elements(Pilsbry and Olsson 1935). However, these fossilhorizons are laterally discontinuous in the basin as aresult of either unconformities between tectono-sequences (Suarez 1996) or lateral facies variations.In addition, the ages of the horizons are still debatable(Nuttal 1990).

Palynology appears to be the best paleontologicaltool for correlation because of the continental natureof the Cenozoic rocks in the MMVB (Pardo-Trujilloet al. 2003). Although many research studies have beencarried out in the basin (mostly by the oil industry),

*Corresponding author. Email: guillermo.rodriguezf@ecopetrol.com.co; grvh3@mail.mst.edu

Palynology

2012, 1–20, iFirst article

ISSN 0191-6122 print/ISSN 1558-9188 online

� 2012 AASP – The Palynological Society

http://dx.doi.org/10.1080/01916122.2012.650548

http://www.tandfonline.com

Dow

nloa

ded

by [

Smith

soni

an I

nstit

utio

n L

ibra

ries

] at

07:

24 2

3 A

pril

2012

limited information has been published (Rueda et al.1996; Pardo-Trujillo et al. 2003). Some of thepublished studies have focused on the Paleocene toLower Eocene interval (Lisama and La Paz forma-tions) (Jaramillo and Dilcher 2001; Pardo-Trujilloet al. 2003) and three of them have studied theEsmeraldas Formation (Van der Hammen 1957,1961; Germeraad et al. 1968; Jaramillo et al. 2006).Germeraad et al. (1968) did not publish any distribu-tion chart or palynological counts, while the study byJaramillo et al. (2006) had low sampling resolution.Jaramillo et al. (2006) did not report the presence ofkey Eocene datums (e.g. the last appearance datum orLAD of Spinizonocolpites grandis) in the EsmeraldasFormation.

The aim of this study was to analyze the pollen andspore contents of the fluvial Eocene EsmeraldasFormation in the Nuevo Mundo Syncline area (Fig-ure 1) in order to refine the palynostratigraphy and todate the formation. Samples from the Nuevo Mundo

Syncline were compared with those from the MedinaSection in the Llanos Foothills (Jaramillo et al. 2011a)in order to determine their floristic similarities.

2. Stratigraphy of the Esmeraldas Formation

An average thickness of 1100 m has been reported forthe Esmeraldas Formation at the eastern flank of theNuevo Mundo Syncline and towards the east of thebasin in general (Morales et al. 1958). In the westernpart of the basin, the Esmeraldas does not crop out,but well data indicate that the formation thins towardthe west. Thin-bedded to laminated, fine-grainedsandstones and siltstones dominate the formation.These sediments are interbedded with dark gray shalesthat are locally mottled brown, red and purple withscattered lignite seamlets (Morales et al. 1958). At thetop of the unit, there is a 15 m thick interbedding oflayers of packstone comprising fossils of freshwaterbivalves and gastropods within shales (Los Corros

Figure 1. Location and geologic map of the Nuevo Mundo Syncline area in the Middle Magdalena Valley Basin. Outcropsections and wells used in this study are indicated. Structure map of Colombia (left) modified from Mora et al. (2010), geologicmap of Nuevo Mundo (right) modified from Ecopetrol (2007).

2 G. Rodrıguez-Forero et al.

Dow

nloa

ded

by [

Smith

soni

an I

nstit

utio

n L

ibra

ries

] at

07:

24 2

3 A

pril

2012

Fossil Horizon, Pilsbry and Olsson 1935; Nuttal 1990;Gomez et al. 2005). The Esmeraldas Formationaccumulated in an alluvial plain setting, while theLos Corros Fossil Horizon accumulated in freshwaterlakes to backswamps (Gomez et al. 2005). Caballero(2010) reported sedimentary structures such as flaserand wavy laminations along the entire unit at theeastern flank of the Nuevo Mundo Syncline (Figure 1),

suggesting some tidal influence in that area probablyrelated to a fluvial-estuarine transition to the east butwithout any saline influence in the area.

3. Previous biostratigraphic studies

Published biostratigraphic data of the Esmeraldasformation has mainly focused on the Los Corros

Figure 2. General Cenozoic stratigraphic column of the Middle Magdalena Valley Basin (modified from Gomez et al. 2005).

Palynology 3

Dow

nloa

ded

by [

Smith

soni

an I

nstit

utio

n L

ibra

ries

] at

07:

24 2

3 A

pril

2012

molluscan fossil horizons (Pilsbry and Olsson 1935;Nuttal 1990). Pilsbry and Olsson (1935) dated theLos Corros horizon as Late Eocene by comparingthe molluscan record with data from other localitiesin the Caribbean. Nuttal (1990) questioned thetaxonomy of Pilsbry and Olson (1935) and concludedthat there were no taxonomical similarities betweenthe marker types chosen in the fossil horizons andthe compared types. Based on his taxonomicalconclusions and comparison with reported Cenozoicmollusks, Nuttal (1990) dated the Los Corros as ‘notmore precise than Paleogene’. Van der Hammen(1957, 1961) used a zonation based on the periodicityof relative abundance peaks of pollen groups that herelated to climatic cycles. These studies correlate theupper part of the Esmeraldas with the Late Eoceneand with the palynological Zone A of the LowerOligocene. He also recognized acmes of Psilamono-copites medius, Mauritiidites franciscoi and Psilamo-noletes tibui (now Laevigatoporites tibuensis) in themajor part of the Esmeraldas, which were consideredcharacteristic of the Late Eocene. These approachesare however strongly affected by the ‘closed sum’statistical artifact (Moore et al. 1991; Kovach andBatten 1994; Jaramillo et al. 2011a). As the relativeabundance of one species increases the othersdecrease, even though their real abundances do notdecrease. Germeraad et al. (1968) indicate that theEsmeraldas Formation lies in the Retitricolporitesguianensis (now Rhoipites guianensis) and Verruca-tosporites usmensis Zones of Middle to Late Eoceneage.

4. Methods

GEMS LTDA (2009) measured four sections (A, B, Cand D) in the Esmeraldas Formation on the easternflank of the Nuevo Mundo Syncline area (Figure 1)and constructed a composite reference section that was1238 m thick (Figure 3). A complete section at a singlesite is difficult to find because of poor exposures.Eighty-two samples from the Esmeraldas Formationwere collected and analyzed for palynology (Table 1).Sample spacing averaged c. 10 m. The samples wereprocessed at Instituto Colombiano del Petroleo usingstandard techniques outlined by Traverse (2007).

Taxonomic identifications were carried out usingJansonius and Hills (1976 and supplements), referencesrelated to tropical palynology (e.g. Germeraad et al.1968; Jaramillo and Dilcher 2001) and the morpholo-gical database of Jaramillo et al. (2011b). Key taxa werelocated using the England FinderTM coordinate system.

The palynological data from the outcrop sectionwas correlated with seven previously studied wellsalong the basin (Figure 1) using Graphic Correlation

(Shaw 1964; Edwards 1984, 1989) and GraphCor1

(Hood 1998). Correlation was carried out to determinethe stratigraphic distribution of palynomorphs withinthe Esmeraldas Formation using the composited out-crop section as the standard reference section. Thecomposite section was correlated with the zonation ofJaramillo et al. (2011a) in order to constrain the age ofthe Esmeraldas Formation.

The palynofloras of the Esmeraldas Formationwere compared to the Middle to Late Eocene floras inthe Medina section of the Llanos Foothills (Jaramilloet al. 2011a) to evaluate their degree of similarity. TheSorensen (Sorensen 1948; Magurran 2004) and Mor-isita-Horn (Horn 1966 in Wolda 1981) indices wereused to compare presence-absence and abundances ofspecies, respectively. Only samples with counts above100 grains were used in this analysis, since sampleswith lower counts have a low probability of represent-ing the total amount of species and their relativeabundances (Hayek and Buzas 1997). A non-metricmultidimensional scale analysis (MDS, Kovach 1989)was performed using both indices as linkage betweenthe samples. The program R for Statistical Comput-ing (R-development-core-team 2005) and the packageVegan (Oksanen et al. 2005) were used for theanalysis. The diversity of the composite section wascalculated by using the rarefaction technique. Speciesdiversity (number of species, Rosenzweig 1995) wascalculated for the composite section by using therarefaction technique (Sanders 1968) and relativeabundances (Hayek and Buzas 1997) in order tounderstand the vegetation composition of the Esmer-aldas Formation. The sample size cutoff for therarefaction was 100.

5. Results

The lowest 335.3 m (1100 feet, 24 samples) of thestudied section are virtually barren of terrestrialorganic matter. In the upper 914.4 m (3000 feet), theterrestrial organic matter and palynomorph recoveryranges from poor to excellent and preservation ismoderate to good. The organic matter is fullycontinental and mainly comprising structured phyto-clasts (woody fragments, plant tissues and cuticles). Insample D 246 (38 m, 128 feet) amorphous organicmatter dominates the organic matter recovered. D 246contains fossil remains of bivalves of 2–3 cm in size,which correspond to those found in the Los CorrosFossil Horizon. This horizon was reported in the upperpart of the D section (Figure 3).

Palynological counts include 4308 spore grains,2304 pollen grains, 637 fungal spores and fungalfruiting bodies and 158 Botryococcus fragments/speci-mens (raw counts in Appendix A, see supplementary

4 G. Rodrıguez-Forero et al.

Dow

nloa

ded

by [

Smith

soni

an I

nstit

utio

n L

ibra

ries

] at

07:

24 2

3 A

pril

2012

online material). Marine palynomorphs are absent.The upper 91.4 m (300 feet) of the section containsscattered Botryococcus which were dominant at 38.1 m(128 feet). There was no record of the colonial algaPediastrum in the outcrop section, although they wererecorded in some of the previously studied wellsections. Table 2 lists the biostratigraphically impor-tant Eocene taxa identified in this study and all areillustrated in Plate 1 (figures 1–22). The otherpalynomorph taxa identified in this study are listedin Appendix B (see supplementary online material).One new pollen species, Rhoipites? perprolatus, isformally described. This grain had been calledTrilongicolpites perbonus in several unpublishedreports by Tropical Oil Company, Shell, Intercoland Ecopetrol (i.e. Rueda et al. 1996) and as

Rhoipites guianensis var. ‘perbonus’ in internalEcopetrol reports.

5.1. Systematic paleontology

Descriptive morphological terminology closely followsthat of Jaramillo and Dilcher (2001) for exinearchitecture and tectal sculpturing. The Rules of theInternational Code of Botanical Nomenclature(McNeill et al. 2006) for species names are followedherein. All figured and type specimens are stored in thepalynological collection of the Litoteca NacionalBernardo Taborda, Instituto Colombiano del Petroleo,km 7 via Piedecuesta, Santander, Colombia. TheLitoteca Nacional (National Core Library) of Colom-bia is a government institute and a public centre of

Figure 3. Stratigraphic sections used to build the outcrop composite section of the Esmeraldas Formation in the Nuevo Mundosyncline (modified from GEMS LTDA 2009).

Palynology 5

Dow

nloa

ded

by [

Smith

soni

an I

nstit

utio

n L

ibra

ries

] at

07:

24 2

3 A

pril

2012

information and research in geological sciences, whichis officially responsible for managing and preservingthe rock and microfossil collections in Colombia. TheLibrary promotes use by scientists and consultantsinterested in global geological processes and explora-tion of oil, mining and energy resources. The inventory

Table 2. List of the main Eocene biostratigraphic pollenand spore taxa identified in the Nuevo Mundo Synclineoutcrop section.

SporesCicatricosisporites dorogensis Potonie and Gelletich 1933

PollenCorsinipollenites undulatus (Gonzalez 1967) Jaramillo and

Dilcher 2001Cricotriporites minutiporus Jaramillo and Dilcher 2001Cyclusphaera scabrata Jaramillo and Dilcher 2001Foveotriporites hammenii Gonzalez 1967Lanagiopollis crassa (Van der Hammen and Wymstra 1964)

Frederiksen 1988Luminidites colombianensis Jaramillo and Dilcher 2001Perisyncolporites pokornyi Germeraad et al. 1968Racemonocolpites facilis Gonzalez 1967Retibrevitricolporites grandis Jaramillo and Dilcher 2001Retibrevitricolporites speciosus Jaramillo and Dilcher 2001Retitricolpites simplex Gonzalez 1967Retisyncolporites angularis Gonzalez 1967Rhoipites guianensis (Van der Hammen and Wymstra 1964)

Jaramillo and Dilcher 2001Rhoipites? perprolatus sp. nov.Spirosyncolpites spiralis Gonzalez 1967

AlgaeBotryococcus spp.

Table 1. Samples analyzed for palynology from the outcropcomposite section. Positions are from the base and the top ofthe Esmeraldas Formation in the composite section.

Section Number

Positionfrom thebase (m)

Positionfrom thetop (m) Lithostratigraphy

D 268 1327 –87 MugrosaD 265 1240 0 EsmeraldasD 263 1212 28 EsmeraldasD 255 1207 33 EsmeraldasD 246 1202 38 EsmeraldasD 269 1194 46 EsmeraldasD 230 1191 49 EsmeraldasD 221 1186 54 EsmeraldasD 219 1180 60 EsmeraldasD 205 1172 68 EsmeraldasD 195 1170 70 EsmeraldasD 185 1165 75 EsmeraldasD 175 1153.5 86.5 EsmeraldasD 164 1148 92 EsmeraldasD 153 1135 105 EsmeraldasD 144 1117 123 EsmeraldasD 134 1110.5 129.5 EsmeraldasD 118 1099.5 140.5 EsmeraldasD 90 1067 173 EsmeraldasD 75 1060 180 EsmeraldasD 66 1055 185 EsmeraldasD 596 1027 213 EsmeraldasD 47 1023 217 EsmeraldasD 31 1015 225 EsmeraldasB 597 1009 231 EsmeraldasB 581 964.5 275.5 EsmeraldasB 576 888.8 351.2 EsmeraldasB 567 869.8 370.2 EsmeraldasB 557 859.5 380.5 EsmeraldasA 464 803 437 EsmeraldasA 455 796.3 443.7 EsmeraldasA 444 766.2 473.8 EsmeraldasA 441 754.5 485.5 EsmeraldasA 440 753.5 486.5 EsmeraldasA 436 747.5 492.5 EsmeraldasA 427 741 499 EsmeraldasA 417 733.5 506.5 EsmeraldasA 407 726 514 EsmeraldasB 522 724.5 515.5 EsmeraldasB 517 723 517 EsmeraldasB 494 711.5 528.5 EsmeraldasB 488 708.5 531.5 EsmeraldasB 783 599 641 EsmeraldasA 395 504.5 735.5 EsmeraldasA 388 499 741 EsmeraldasB 776 448.5 791.5 EsmeraldasB 767 441 799 EsmeraldasB 749 427.3 812.7 EsmeraldasB 746 424.5 815.5 EsmeraldasB 736 417.5 822.5 EsmeraldasB 728 411.5 828.5 EsmeraldasB 710 395.3 844.7 EsmeraldasB 699 383.5 856.5 EsmeraldasB 689 376 864 EsmeraldasB 678 356 884 EsmeraldasB 672 349.5 890.5 EsmeraldasB 662 306 934 EsmeraldasA 280 289.5 950.5 EsmeraldasA 270 271.7 968.3 Esmeraldas

(continued)

Table 1. (Continued).

Section Number

Positionfrom thebase (m)

Positionfrom thetop (m) Lithostratigraphy

A 250 267.5 972.5 EsmeraldasB 660 253 987 EsmeraldasB 658 251.5 988.5 EsmeraldasA 240 250.3 989.7 EsmeraldasB 648 244 996 EsmeraldasA 230 242.5 997.5 EsmeraldasA 220 235 1005 EsmeraldasB 638 233.5 1006.5 EsmeraldasA 210 228.2 1011.8 EsmeraldasB 628 226 1014 EsmeraldasA 201 220.7 1019.3 EsmeraldasC 372 60 1180 EsmeraldasC 362 57 1183 EsmeraldasC 352 50 1190 EsmeraldasA 198 40 1200 EsmeraldasC 345 34 1206 EsmeraldasB 623 32 1208 EsmeraldasC 342 31.5 1208.5 EsmeraldasA 197 31 1209 EsmeraldasA 342 25 1215 EsmeraldasC 330 22.5 1217.5 EsmeraldasA 189 19 1221 EsmeraldasA 179 17.5 1222.5 Esmeraldas

6 G. Rodrıguez-Forero et al.

Dow

nloa

ded

by [

Smith

soni

an I

nstit

utio

n L

ibra

ries

] at

07:

24 2

3 A

pril

2012

includes public and confidential collections of cores,cuttings, outcrops, petrological rock samples andmicropaleontological collections. Holotypes and para-types can be freely consulted upon written request tothe library manager.

Rhoipites? perprolatus sp. nov.Plate 1, figures 15–19

Holotype. Lisama-10 well, 9850’, EF O23 (Plate 1,figure 15).Paratypes. Vega de Pato Creek, 144 B, EF U18(Plate 1, figure 16); Lisama-10 well, 9850’, EF G18/2(Plate 1, figure 17).Ethymology. After its perprolate shape.Diagnosis. Tricolpate, mid-sized (22–42 mm long),ellipsoidal, perprolate, exine thick (1 mm), reticulate,heterobrochate, marginate, lumina distinctly elongatedparallel to the colpi, slightly cristate and resembling afunnel.Description. Single pollen grain, monad, radial sym-metry, isopolar, perprolate; tricolpate, sometimestricolporate, colpi slightly marginate, furrow long,reaching polar area and projecting deeply inwards,margo formed by slight decrease in size of the luminatoward the aperture; exine semitectate, columellaedistinct, reticulate, heterobrochate, large in mesocol-pium becoming smaller near colpi margins andapocolpium, reticula angular, elongated toward poles,slightly cristate, lumina 1–1.5 mm, decreasing graduallytoward the base of the reticula resembling a funnel,muri thick (1–1.5 mm).Dimensions. Equatorial diameter: range 9–22 mm,mean ¼ 14 mm; polar diameter: range 22–42 mm,mean ¼ 31 mm, SD ¼ 10.3; polar/equatorial diameter:range 1.4–3.5, mean ¼ 2.2 colpi margins 2 mm thick,exine 1 mm thick, lumina 3–4 mm wide, parallel to thepolar axis. Specimens measured: 304 in several sectionsalong the MMVB and the Catatumbo Basin (Appen-dix C, supplementary online material). Measurementsare listed in Appendix D (see supplementary onlinematerial).Comments. The genus Rhoipites was defined by Wode-house (1933) to include pollen grains that areellipsoidal, tricolporate, reticulate with a long andpointed colpi; the colpi and pore have conspicuousthickenings that projects deeply inwards (Jansoniusand Hills 1976, CARD 2421). Rhoipites? perprolatushas great similarity to Rhoipites guianensis (Van derHammen and Wijmstra 1964) Jaramillo and Dilcher(2001) in having angular reticula elongated toward thepoles and decreasing toward the aperture. However,Rhoipites? perprolatus is perprolate (polar diameter/equatorial diameter mean ¼ 2.3 mm), slightly hetero-brochate and tricolpate to occasionally tricolporate,

whereas R. guianensis is prolate (polar diameter/equatorial diameter mean ¼ 1.5 mm), strongly hetero-brocate and always tricolporate.

5.2. Palynostratigraphy

The graphic correlation equations between the sec-tions are shown in Appendix E (see supplementaryonline material), and the first appearance datums(FAD) and last appearance datums (LAD) in thecomposite section can be found in Appendix F (seesupplementary online material). The most importantevents are illustrated in Figures 4 and 5. Most of theFADs and LADs in the composite section arestrongly influenced by the edge effect (Foote 2000),which is an artificial increase in FAD and LAD at thelimits of a studied section. The time represented bythe Esmeraldas Formation is shorter compared to thebiostratigraphic range of most of the recorded taxa; itwould therefore be expected to have a strong edgeeffect in the sections. Nevertheless, there are someuseful events for regional correlation purposes. TheFAD of Rhoipites guianensis, the LAD of Racemono-colpites facilis and the LAD of Rhoipites? perprolatusappear to be useful as correlation elements and arenot affected by edge effects. The Esmeraldas Forma-tion can be divided into one interval zone and oneassemblage zone that is subdivided into two subzones(Figure 5). The positions of the base and top of thezones are presented in composite units (cu). Theseunits have the polarity of a well, being smaller towardyounger rocks and higher toward older strata of thecomposite section.

5.2.1. Interval zone (4061–3242 cu)

The lowest (oldest) 243.84 m (800 feet) of the sectionhas very low recovery of pollen and spores. This zoneis recognized by the presence of Spirosyncolpitesspiralis, Cyclusphaera scabrata, Psilatriletes spp. andLaevigatosporites tibuensis. Botryococcus spp. andfungal spores are also present.

5.2.2. Rhoipites? perprolatus assemblage zone(3242–0 cu)

Key Rhoipites guianensis and Rhoipites? perprolatusevents define this assemblage zone. The base of thezone is defined by the top of the Interval Zone and theFAD of Rhoipites? perprolatus, Rhoipites guianensisand Cicatricosisporites dorogensis. The top of the zoneis the LAD in the basin of Rhoipites? perprolatus. Theassemblage comprises typical Eocene taxa, such asBrevitricolpites microechinatus, Bevitricolpites macro-exinatus, Corsinipollenites undulatus, Cricotriporites

Palynology 7

Dow

nloa

ded

by [

Smith

soni

an I

nstit

utio

n L

ibra

ries

] at

07:

24 2

3 A

pril

2012

8 G. Rodrıguez-Forero et al.

Dow

nloa

ded

by [

Smith

soni

an I

nstit

utio

n L

ibra

ries

] at

07:

24 2

3 A

pril

2012

minutiporus, Cyclusphaera scabrata, Foveotriporiteshammenii, Lanagiopollis crassa, Luminidites colombian-esis, Mauritiidites franciscoi var. franciscoi, Monocol-popollenites ovatus, Monoporopollenites annulatus,Perisyncolporites pokornyi, Polypodiaceoisporites spp.,Pseudostephanocolpites perfectus, Retibrevitricolporitesgrandis, R. speciosus, Retistephanoporites minutiporus,Retitricolpites simplex, Retisyncolporites angularis,Spirosyncolpites spiralis, Tetracolporopollenites macu-losus, T. transversalis and Verrutriletes virueloides. TheFAD of Perisyncolporites pokornyi and Luminiditescolombianesis, the LAD of Cricotriporites minutiporus,L. colombianensis and Racemonocolpites facils and thepresence of Retisyncolporites angularis and Pseudos-tephanocolpites perfectus are also recorded in this

zone. Fossil remains of molluscs were present in theupper part in the outcrop section and in the ditch-cuttings of some wells and are associated with highabundances of freshwater algae (Botryococcus spp.and Pediastrum spp.) and amorphous organic matter.These fossils likely correspond to the Los CorrosFossil Horizon.

The LAD of Racemonocolpites facilis and abun-dance of the freshwater algae divide the Rhoipites?perprolatus Zone into two informal subzones, A and B.

5.2.3. Subzone A (3242–223.04 cu)

This subzone is defined as a concurrent-rangesubzone, the top of which is defined by the LAD of

Figure 4. Graphic correlation between the Esmeraldas composite section in the Middle Magdalena Valley and the Eocene–Oligocene zonation of the Llanos and Llanos Foothills of Jaramillo et al. (2011a).

PLATE 1. Figure 1. Cicatricosisporites dorogensis. Slide D 263, EF K41/3. Figure 2. Corsinipollenites undulatus. Slide D 153,EF X43. Figure 3. Cricotriporites minutiporus. Slide B 517, EF T38/1. Figure 4. Cyclusphaera scabrata. Slide D 219, EF L2.Figure 5. Foveotriporites hammenii Slide B 517, EF Q27. Figure 6. Lanagiopollis crassa. Slide D 269, EF N39/9. Figure 7.Luminidites colombianensis. Slide A 395, EF C21. Figure 8. Perisyncolporites pokornyi. Slide D 263, EF P36. Figure 9.Racemonocolpites facilis. Slide D 195, EF W43/1. Figure 10. Retibrevitricolporites grandis. Slide D 246, EF S6. Figure 11.Retibrevitricolporites speciosus. Slide D 269, EF H51/2. Figure 12. Retitricolpites simplex. Slide D 269, EF K7. Figure 13.Retisyncolporites angularis. Slide B 783, EF L10/3. Figure 14. Rhoipites guianensis. Slide D 164, EF W26/3. Figure 15. Rhoipites?perprolatus Holotype, Slide Lisama-10 9850, EF O23. Figure 16. Rhoipites? perprolatus Paratype, Slide D 144, EF U18. Figure17. Rhoipites? perprolatus Paratype, Slide Lisama 10 9850, EF G18/2. Figure 18. Rhoipites? perprolatus SEM photograph, Lisama4 (10700’). Figure 19. Rhoipites? perprolatus SEM photograph, 2 Lisama 4 (11350’). Figure 20. SEM close-up of reticula ofRhoipites? perprolatus. Figure 21. Spirosyncolpites spiralis. Slide A 395, EF S26. Figure 22. Botryococcus spp. Slide D 269, EFP20/2.

"

Palynology 9

Dow

nloa

ded

by [

Smith

soni

an I

nstit

utio

n L

ibra

ries

] at

07:

24 2

3 A

pril

2012

Figure

5.

Palynostratigraphicdistributionschem

eoftheEsm

eraldasForm

ationin

theNuevoMundosynclinearea.Thebiostratigraphicranges

ofthemostim

portanttaxa

are

shown.

10 G. Rodrıguez-Forero et al.

Dow

nloa

ded

by [

Smith

soni

an I

nstit

utio

n L

ibra

ries

] at

07:

24 2

3 A

pril

2012

Racemonocolpites facilis. This LAD is located 140 cubelow the Los Corros Fossil Horizon and has beenrecognized in five sections. The palynological assem-blage is the same as that described for the Rhoipites?perprolatus Zone.

5.2.4. Subzone B (223.04–0 cu)

This zone has the typical palynological assemblage ofthe Rhoipites? perprolatus zone and high abundancesof the freshwater algae Botryococcus and Pediastrum.The mollusc layers of Los Corros Horizon arelocated within this subzone. It is worth noting thatalthough the acmes of Botryococcus spp. andPediastrum spp. are located within this subzone,they were not found co-occurring in the samelocations. Botryococcus spp. acmes were found onlyin the outcrop section, whereas Pediastrum spp.acmes were found in three wells. The LAD ofRhoipites? perprolatus at the Esmeraldas–Mugrosaboundary defines the top of the subzone.

5.3. Correlation and age of the Esmeraldas Formation

The correlation points used for comparison with thebiozonation of the Llanos Foothills of Colombia(Jaramillo et al. 2011a) were the FADs of Lanagiopolliscrassa, Cicatricosisporites dorogensis, Perisyncolporitespokornyi and Luminidites colombianensis and theLADs of Cricotriporites minutiporus and Racemono-colpites facilis. Figures 4 and 5 illustrate the graphiccorrelation and the palynostratigraphic distribution

scheme, respectively. Appendix E contains the correla-tion equations.

Cyclusphaera scabrata and Spirosyncolpites spiraliswere present within the Interval Zone. The FADs ofboth palynomorphs are located in the Early EoceneT04-T05 zones of Jaramillo et al. (2011a), respectively.The FAD of Cicatricosisporites dorogensis (zone T06of Jaramillo et al. (2011a), Middle Eocene) was locatedabove, within an interval with good pollen and sporerecovery, and was found in similar stratigraphicposition across the MMVB. This suggests a probableEarly Eocene age for the base of the EsmeraldasFormation (Interval Zone, Figure 4). Using thecorrelation equation, the top of the T06 Zone istentatively placed at 764 cu in the Esmeraldascomposite section; the Rhoipites? perprolatus zone istherefore Middle to Late Eocene (T06-T07 Zones). TheLAD of Racemonocolpites facilis is located between703 cu and the Los Corros in the Esmeraldascomposite section, and within the T07 Zone in thefoothills. The Los Corros Fossil Horizon is LateEocene (T07 Zone, Figure 4).

5.4. Similarities between the Nuevo Mundo synclineand the Medina sections

The Sorensen values show that the median of theNuevo Mundo versus the Medina samples is 0.18,Nuevo Mundo within-samples is 0.51 and Medinawithin-samples is 0.48. For the Morisita-Horn Index,the median of the Nuevo Mundo versus the Medinasamples is 0.27, Nuevo Mundo within-samples is 0.78

Figure 6. Non-metric multidimensional scaling analysis grouping the samples from the Nuevo Mundo Section (1–30) in theMiddle Magdalena and the Medina section of the Llanos Foothills (31–47).

Palynology 11

Dow

nloa

ded

by [

Smith

soni

an I

nstit

utio

n L

ibra

ries

] at

07:

24 2

3 A

pril

2012

and Medina within-samples is 0.6. The non-metricmultidimensional (MDS) analysis (Figure 6) alsoshows that the samples from Nuevo Mundo andMedina section form two distinct groups. Using bothsimilarity indices, the first axis explains the dissim-ilarity between Nuevo Mundo and Medina samplesand the second axis the dissimilarities of NuevoMundo within samples and Medina within samples.

6. Discussion

6.1. Palynostratigraphy and age

The age of the base of the Rhoipites? perprolatus zone isuncertain. The Middle Magdalena Valley Unconfor-mity (Gomez et al. 2005), a regional unconformity inthe MMVB, places the Esmeraldas Formation aboveeither the Paleocene Lisama Formation or Cretaceousrocks in the wells used for correlation. The base of theEsmeraldas Formation (Interval Zone) in these wellsmay not represent the true stratigraphic base of theformation compared to the most complete section ofEsmeraldas in the Nuevo Mundo Syncline. However,despite low recovery, the Interval Zone is present in allwells below the FAD of R. guianensis, suggesting thatthe zone could be regionally extensive. It is associatedwith red to purple shales and paleosols, probablyrelated to intense oxidation due to subaerial exposure.

In northern South America, the FAD of Cicatri-cosisporites dorogensis occurs during the earliestMiddle Eocene (Germeraad et al. 1968; Muller et al.1987; Jaramillo et al. 2011a). This FAD marks the baseof the T06 Zone of Jaramillo et al. (2011a). Pardo-Trujillo et al. (2003) and Jaramillo and Dilcher (2001)did not record Cicatricosisporites dorogensis, Rhoipitesguianensis and Rhoipites? perprolatus in the La PazFormation in the Nuevo Mundo Syncline area.

Cicatricosisporites dorogensis is not very abundantin the Esmeraldas Formation. In the Cenozoic of

northern South America, the FAD of C. dorogensis iswell defined and the spore becomes frequent regionallyfollowing its FAD. We observed the same pattern inthe Esmeraldas Formation, suggesting that the ob-served FAD of Cicatricosisporites dorogensis in theEsmeraldas Formation could be near the true origina-tion event. The LADs of Spinizonocolpites grandis andEchitriporites trianguliformis var. orbicularis are usefulto identify the upper Middle and Late Eocene(Jaramillo et al. 2011a). Unfortunately, they are notpresent in the Esmeraldas Formation. However,additional Middle to Late Eocene events are usefulfor correlation (Jaramillo et al. 2011a). These eventsare the FAD of Perisyncolporites pokornyi andLuminidites colombianesis and the LAD of Brevitricol-pites macroexinatus, Cricotriporites minutiporus, Lumi-nidites colombianensis and Racemonocolpites facilis.There is also a Late Eocene marine flooding in theCentral Llanos Foothills (Santos et al. 2008) thatappears to be correlative with the Los Corros FossilHorizon (Figure 4).

6.2. Regional stratigraphy: Middle Magdalena Valleyto the Llanos Foothills

Cooper et al. (1995) and Bayona et al. (2008) suggestedthat the La Paz and Esmeraldas formations werecorrelative with the Mirador and the base of theCarbonera formations (sequences 3a-3b and sequence4 of Bayona et al. 2008, respectively).

Figure 7 shows the depositional environments andcorrelation of the Eocene units across the MMVB tothe Llanos Foothills. Santos et al. (2008) documented amarine transgression during the Late Eocene innorthwestern South America. Using a Salinity Index,they established marine influence in the Central LlanosFoothills and Putumayo Basins within the T07 Zone ofJaramillo et al. (2011b). We have found that this Late

Figure 7. Correlation of the MMVB versus the Llanos Foothills. MMVB data from Pardo-Trujillo (2004). (1) This work; (2)Gomez et al. (2005); (3) Eastern Cordillera data from Pardo-Trujillo (2004) and ICP-STRI (2006); (4) Llanos and LlanosFoothills data from Fajardo et al. (2000); (5) Parra et al. (2009); (6) Jaramillo et al. (2005, 2009); (7,8) Bayona et al. (2008); (9)Cooper et al. (1995) and (10) S. catatumbus Zone (T05) of the Eocene can be defined in the MMVB by following the zonationproposed by Jaramillo et al. (2011a).

12 G. Rodrıguez-Forero et al.

Dow

nloa

ded

by [

Smith

soni

an I

nstit

utio

n L

ibra

ries

] at

07:

24 2

3 A

pril

2012

Eocene transgression is correlative with the lacustrineLos Corros Fossil Horizon (Figure 4). Santos et al.(2008) hypothesized that the sea transgressed througha corridor formed in the proto Lower MagdalenaValley Basin north of the MMVB, based on thetectonic reconstruction of the Northern Andes byMontes et al. (2005) (Figure 8). The Nuevo MundoSyncline area was located very close to this marinecorridor in the south (Figure 8). It seems plausible tointerpret that both the marine flooding in the LlanosFoothills and the extensive lacustrine system developedin the MMVB are products of the same event. Both arecoetaneous, short-lived and widespread. After therotation of the Maracaibo block (Montes et al. 2005,red polygon in Figure 8, basins 1 and 2), the marinecorridor was also rotated and placed farther northeastof Nuevo Mundo Syncline. Bayona et al. (2008)proposed that between the MMVB and the EasternCordillera, there has probably been a geographic

barrier since the Eocene. Moreno (2010) reportedpaleocurrents with a predominant NW direction eastof the Nuevo Mundo Syncline area, which suggestprovenance of sediments from the east from a probableproto-Eastern Cordillera. It is possible that there was abarrier preventing the marine flooding from reachingthe MMVB, where extensive lakes were developing. Onthe other hand, it is also possible that the inlandexpression of the marine flooding was a rapid increasein the accommodation space driving the developmentof extensive but short-lived lacustrine systems, such asthose represented by the los Corros Fossil Horizon.

6.3. Similarities between sections

The MDS analysis shows that the floral compositionsof Nuevo Mundo syncline section in the MMVB andthe Medina section in the Llanos Foothills are differentboth in terms of presence-absence and abundance of

Figure 8. Marine transgression during the Late Eocene (Santos et al. 2008). (1) Flooding and Nuevo Mundo Lacustrineenvironments during the Late Eocene. The proposed marine corridor was located north of the lacustrine area. (2) Present-dayposition of the proposed flooding corridor and the Nuevo Mundo area after the rotation of blocks of Northern Colombia(rotation after Montes et al. 2005). The rotation and translation of the blocks fragmented and displaced the marine corridortowards the northeast.

Palynology 13

Dow

nloa

ded

by [

Smith

soni

an I

nstit

utio

n L

ibra

ries

] at

07:

24 2

3 A

pril

2012

species (Figure 6). These differences could be theproduct of differences in depositional environments.The Nuevo Mundo Syncline was mainly alluvial,whereas coastal plain to slightly marine conditionsprevailed in the Llanos Foothills at the same time.Alluvial plain sediments such as those found in theNuevo Mundo Syncline tend to have lower recoveryrates than other environments, as noted by Lorente(1986). She also found that richness in the pollenrecovery in this environment is highly variable frombarren to very rich, depending on taphonomicprocesses (oxidation) resulting from environmentalfluctuations. Differences in taphonomy rather thandifferences in floristic provinces appear to be a moreplausible explanation for the observed differencesbetween MMVB and Medina.

7. Conclusions

The age of the Esmeraldas Formation ranges from lateEarly Eocene to Late Eocene. The Los Corros FossilHorizon is dated as Late Eocene and it is coetaneouswith the short-lived marine incursion recorded in thecentral Llanos Foothills. The Esmeraldas Formationcorrelates with the upper part of the Picacho Forma-tion and the lower part of the Concentracion Forma-tion in the Eastern Cordillera, the middle to upper partof the Mirador Formation and the lower part of the C8member of the Carbonera Formation in the LlanosFoothills.

Acknowledgements

We dedicate this manuscript to the memory of DougNichols, a great palynologist and a fantastic person.The authors would like to thank the InstitutoColombiano del Petroleo for financial assistance toGRF and permission to publish this study. We areindebted to Andres Mora, Felipe De La Parra,Victor Caballero, Jorge Rubiano, the BiostratigraphyTeam at ICP and the Paleontology Laboratorygroup at Missouri University of Science and Tech-nology for discussions and help at different stages ofthis study. GRF would like to thank his parents andbrothers for their support and his wife Angelica forher love, for being there always and for keeping himfocused.

Author biographies

GUILLERMO RODRIGUEZ-FORERO is a palynologistat the Biostratigraphy Team of the Colombian PetroleumInstitute of ECOPETROL, the Colombian state-owned oil

company. He got his M.Sc. degree from Missouri Universityof Science and Technology in 2010. He has experience in the

palynostratigraphy of Cenozoic of northern South America

and applications of biostatigraphy in the oil industry andgeological problems.

FRANCISCA E. OBOH-IKUENOBE is Professor and

Head of the Geology and Geophysics Program at MissouriUniversity of Science and Technology. Her research interestsin palynology and sedimentology span Quaternary, Neogene,

Paleogene, Cretaceous and Cambro-Ordovician sedimentarysequences in diverse localities in Western Australia, SouthAfrica, Nigeria, offshore West Africa, Colombia, Egypt, U.S.

Gulf Coast, Missouri and the Western Interior Basin. Shebegan palynological studies at the University of Ife (nowObafemi Awolowo University) in Nigeria where she receivedBSc (Honours) and M.Sc. degrees in geology. She attended

the University of Cambridge as a Commonwealth Scholarand was awarded a Ph.D. degree for a thesis on Miocenesediments of the Niger Delta. She is currently President of

AASP – The Palynological Society and is a Fellow of theGeological Society of America (GSA). She is a member-at-large of the GSA Diversity in the Geosciences and Research

Grant committees, served as Director of the Association forWomen Geoscientists Foundation in 2005–2009, and repre-sented AASP as Councillor of the International Federationof Palynological Societies from 1996 to 2004.

CARLOS JARAMILLO-MUNOZ is a staff scientist withthe Smithsonian Tropical Research Institute in Panama. Hisresearch investigates the causes, patterns, and processes of

tropical biodiversity at diverse scales of time and space. He isalso interested in Cretaceous–Cenozoic biostratigraphy oflow latitudes, developing methods for high-resolution bio-

stratigraphy and the paleobiogeography of Tethys.

MILTON J. RUEDA-SERRANO is a consultant geologistfor the oil industry in Colombia, with more than 25 years of

experience in the palynostratigraphy of Cenozoic andCretaceous sequences from northern South America, taxon-omy and applications of palynology in the solution of

geological problems.

EDWIN A. CADENA-RUEDA is a Colombian geologist–

paleontologist, he obtained an M.Sc. degree in Geology in2009 at the University of Florida, and currently is a Ph.D.candidate at the Marine Earth and Atmospheric SciencesDepartment, North Carolina State University. His bachelor

degree thesis was focused in the morphological descriptionusing SEM of Trilongicolpites perbonus (Rhoipites? perprola-tus) from the Paleogene of Colombia. Currently, his research

topics are in evolution and paleobiogeography of tropicalturtles, as well as tracking the preservation of bone cells andproteins in turtles from Present to Jurassic.

References

Bayona G, Cortes M, Jaramillo C, Ojeda G, Aristizabal JJ,Reyes-Harker A. 2008. An integrated analysis of anorogen–sedimentary basin pair: Latest Cretaceous–Cen-ozoic evolution of the linked Eastern Cordillera orogenand the Llanos foreland basin of Colombia. GeologicalSociety of America Bulletin 120, 1171–1191.

14 G. Rodrıguez-Forero et al.

Dow

nloa

ded

by [

Smith

soni

an I

nstit

utio

n L

ibra

ries

] at

07:

24 2

3 A

pril

2012

Caballero V. 2010. Evolucion tectono - sedimentaria delSinclinal de Nuevo Mundo, cuenca sedimentaria ValleMedio del Magdalena Colombia, durante el Oligoceno –Mioceno [Unpublished M.Sc. thesis]. [Colombia]: Uni-versidad Industrial de Santander. 149 p.

Cooper MA, Addison FT, Alvarez R, Coral M, GrahamRH, Hayward AB, Howe S, Martınez J, Naar J, Penas R,Pulham AJ, Taborda A. 1995. Basin development andtectonic history of the Llanos Basin, Eastern Cordilleraand Middle Magdalena Valley, Colombia. AmericanAssociation of Petroleum Geologists Bulletin 79, 1421–1443.

Duenas H. 1980. Palynology of Oligocene-Miocene strata ofborehole Q-E-22, Planeta Rica, Northern Colombia.Review of Palaeobotany and Palynology 30, 313–328.

Ecopetrol SA. 2007. Mapa geologico Ecp-Ariana CuencaVMM. Internal Report.

Ecopetrol SA. - Instituto Colombiano Del Petroleo (ICP),and Smithsonian Tropical Research Institute(STRI)(ICP-STRI), 2006. Palinologıa Formacion Con-centracion. Ecopetrol S.A. - Instituto Colombiano delPetroleo, Internal Report. p. 40–50.

Edwards LE. 1984. Insights on why graphic correlation(Shaw’s method) works. Journal of Geology 92, 583–597.

Edwards LE. 1989. Supplemented graphic correlation: apowerful tool for paleontologists and nonpaleontolo-gists. Palaios 4, 127–143.

Fajardo A, Rojas L, Cristancho J. 2000. Definicion delmodelo estratigrafico en el intervalo Cretaceo tardıo aMioceno medio en la cuenca Llanos orientales ypiedemonte llanero. Bucaramanga. Instituto Colombia-no del Petroleo, Internal Report. p. 200.

Foote M. 2000. Origination and extinction components oftaxonomic diversity: Paleozoic and post-Paleozoic dy-namics. Paleobiology 26, 578–605.

Frederiksen N. 1988. Sporomorph biostratigraphy, floralchanges, and paleoclimatology, Eocene and earliestOligocene of the eastern Gulf Coast. United StatesGeological Survey Professional Paper 1448, 1–68.

GEMS LTDA – Instituto Colombiano Del Petroleo (ICP).2009. Adquisicion y consolidacion de informacion para elValle Medio del Magdalena. Ecopetrol S.A. - InstitutoColombiano del Petroleo, Internal Report. 49 p.

Germeraad JH, Hopping CA, Muller J. 1968. Palynology ofTertiary sediments from tropical areas. Review ofPalaeobotany and Palynology 6, 189–348.

Gomez E, Jordan T, Allmendinger R, Hegarty K, Kelley S.2005. Syntectonic Cenozoic sedimentation in the north-ern middle Magdalena Valley Basin of Colombia andimplications for exhumation of the northern Andes.Geological Society of America Bulletin 117, 547–569.

Gonzalez AE. 1967. A palynologic study on the upperLos Cuervos and Mirador formations (lower and middleEocene), Tibu Area, Colombia. Leiden: E.J. Brill. 68 p.

Hayek LC, Buzas MA. 1997. Surveying natural populations.New York: Columbia University Press. 563 p.

Horn H. 1966. Measurement of ‘overlap’ in comparativeecological studies. Am Nat 100, 419–424.

Hood KC. 1998. GraphCor Interactive Graphic CorrelationSoftware. Houston, TX: Hood, K.C.

Jansonius J, Hills LV. 1976. Genera file of fossil spores andpollen [with supplements]. Special Publication, Depart-ment of Geology, University of Calgary, Canada.

Jaramillo C. 2008. Five useful techniques to analyzepalynological data. The Palaeobotanist 57, 529–537.

Jaramillo C, Dilcher DL. 2001. Middle Paleogene palynologyof Central Colombia, South America: a study of pollenand spores from tropical latitudes. PalaeontographicaAbteilung B 285, 87–213.

Jaramillo C, Munoz F, Cogollo M, De La Parra F. 2005.Quantitative biostratigraphy for the Paleocene of theLlanos Foothills, Colombia: improving palynologicalresolution for oil exploration. In: Powell AJ, Riding J,editors. Recent developments in applied biostratigra-phy. The Micropaleontological Society, Special Pub-lication. London: The Geological Society. p. 145–159.

Jaramillo C, Rueda M, Mora G. 2006. Cenozoicplant diversity in the Neotropics. Science 311, 1893–1896.

Jaramillo C, Pardo-Trujillo A, Rueda M, Harrington G,Bayona G, Torres V, Mora G. 2007. Palynology of theUpper Paleocene Cerrejon Formation, northern Colom-bia. Palynology 31, 153–189.

Jaramillo C, Rueda M, Bayona G, Santos C, Florez P, ParraF. 2009. Biostratigraphy breaking paradigms: dating theMirador Formation in the Llanos Basin of Colombia. InDemchuk T, editor. Biostratigraphic applications.SEPM Special Publication 93, 29–40.

Jaramillo C, Rueda M, Torres V. 2011a. A palynologicalzonation for the Cenozoic of the Llanos Foothills ofColombia. Palynology 35, 46–84.

Jaramillo C, Rueda M, de la Parra F. 2011b. A Morpholo-gical Electronic Database of Cretaceous-Tertiary FossilPollen and Spores from Northern South America.Colombian Petroleum Institute & Smithsonian TropicalResearch. Available from: http://biogeodb.stri. si.edu/jaramillo/palynomorph/.

Kovach WL. 1989. Comparisons of multivariate analyticaltechniques for use in Pre-Quaternary plant paleoecol-ogy: Review of Palaeobotany and Palynolology 60, 255–282.

Kovach WL, Batten DJ. 1994. Association of palynomorphsand palynodebris with depositional environments: qua-titative approaches. In: Traverse A, editor. Sedimenta-tion of organic matter. Cambridge: CambridgeUniversity Press. p. 391–408.

Lorente M. 1986. Palynology and Palynofacies of the UpperTertiary in Venezuela. Berlin: J. Cramer. 222 p.

Magurran AE. 2004. Measuring biological diversity. Malden(MA): Blackwell Publishing. 256 p.

McNeill J, Barrie FR, Burdet HM, Demoulin V, Hawks-worth DL, Marhold K, Nicolson DH, Prado J SilvaPC Skog JE, Wiersema JH. 2006. International Codeof Botanical Nomenclature (Vienna Code) 2006.Ruggell (Liechtenstein): A.R.G. Gantner Verlag KG.568 p.

Montes C, Hatcher R, Restrepo-Pace P. 2005. Tectonicreconstruction of the Northern Andean blocks: obliqueconvergence and rotations derived from the kinematics ofthe Piedras-Girardot Area. Colombia: Tectonophysics399, 221–250.

Moore PD, Webb, JA, Collinson, ME. 1991. Pollen analysis.London: Blackwell Scientific Publications. 216 p.

Mora A, Horton BK, Mesa A, Rubiano J, Richard, A,Ketcham, RA, Parra M, Blanco V, Garcia D, StockliDF. 2010. Migration of Cenozoic deformation in theeastern Cordillera of Colombia interpreted from fissiontrack results and structural relationships: implications forpetroleum systems: American Association of PetroleumGeologists Bulletin 94, 1–37.

Palynology 15

Dow

nloa

ded

by [

Smith

soni

an I

nstit

utio

n L

ibra

ries

] at

07:

24 2

3 A

pril

2012

Morales LG, Podesta DJ, Hatfield WC, Tanner H, Jones SH,Barker MH, O’Donoghue DJ, Mohler E, Dubois EP,Jacobs C, Goss CR. 1958. General geology and oiloccurrences of the Middle Magdalena Valley, Colombia.Habitat of Oil Symposium, American Association ofPetroleum Geologists. p. 641–695.

Moreno C. 2010. Paleogene sedimentation patterns and basinevolution during Andean orogenesis, Middle MagdalenaValley basin, Colombia [Unpublished M.Sc. thesis].[Austin, TX]: University of Texas at Austin. 59 p.

Muller J, Di Giacomo E, Van Erve A. 1987. A palynologiczonation for the Cretaceous, Tertiary and Quaternary ofnorthern South America. American Association ofStratigraphic Palynologists Contribution Series 19, 7–76.

Nuttal CP. 1990. A review of the Tertiary non marinemolluscan faunas of the Pebasian and other inlandbasins of north-western South America. Bulletin of theBritishMuseum of Natural History (Geology) 45, 165–371.

Oksanen J, Kindt R, O’Hara B. 2005. Community EcologyPackage, Package VEGAN, R for Statistical Computing.

Pardo-Trujillo A, Jaramillo CA, Oboh-Ikuenobe F. 2003.Paleogene palynostratigraphy of the Eastern MiddleMagdalena Valley, Colombia. Palynology 27, 151–174.

Pardo-Trujillo A. 2004. Paleocene-Eocene palynology andpalynofacies from northeastern Colombia and westernVenezuela [Unpublished thesis]. [Belgium]: University ofLiege. 103 p.

Parra M, Mora A, Jaramillo C, Strecker M.R, Sobel ER,Quiroz L, Rueda M, Torres V. 2009. Orogenic wedgeadvance in the northern Andes: evidence from the Oligo-Miocene sedimentary record of the Medina Basin,Eastern Cordillera, Colombia. Geological Society ofAmerica Bulletin 121, 780–800.

Parra M, Mora A, Jaramillo C, Torres V, Zeilingern G,Streckern M. in press. Tectonic controls on Cenozoicforeland basin development in the north-eastern Andes,Colombia. Basin Research.

Pilsbry HA, Olsson AA. 1935. Tertiary fresh-water mollusksof the Magdalena embayment. Proceedings of theAcademy of Natural Sciences of Philadelphia 87, 7–21.

R-Development-Core-Team 2005. R: A language andenvironment for statistical computing. R Foundationfor Statistical Computing: Vienna, Austria. ISBN 3–900051–07–0. Available from: http://www.R-project.org.

Rosenzweig ML. 1995. Species diversity in space and time.Cambridge: Cambridge University Press. 433 p.

Rueda M, Zegarra M, Reyes JP. 1996. Estratigrafıa ypalinoestratigrafıa del Valle Medio del Magdalena.Volumen 4. Instituto Colombiano del Petroleo, InternalReport. 52 p.

Sanders HL. 1968. Marine benthic diversity: a comparativestudy. American Naturalist 102, 243–282.

Santos C, Jaramillo C, Bayona G, Rueda M, Torres V. 2008.Late Eocene marine incursion in Northwestern SouthAmerica. Palaeogeography, Palaeoclimatology, Palaeoe-cology 264, 140–146.

Shaw AB. 1964. Time in stratigraphy. New York: McGraw-Hill Book Co. 365 p.

Sorensen T. 1948. A method of establishing groups of equalamplitude in plant sociology based on similarity ofspecies content and its application to analyses of thevegetation on Danish commons. Biologsike Skrifter 5, 1–34.

Suarez M. 1996. Facies analysis of the upper Eocene La PazFormation, and regional evaluation of post-middleEocene stratigraphy, northern Middle Magdalena ValleyBasin [Unpublished M.Sc. thesis]. [Boulder, CO]: Uni-versity of Colorado. 88 p.

Traverse A. 2007. Paleopalynology. 2nd ed. Dordrecht:Springer. 817 p.

Van Der Hammen T. 1957. Climatic periodicity andevolution of South American Maestrichtian and Tertiaryfloras: a study based on pollen analysis in Colombia.Boletın Geologico (Bogota) 5, 49–91.

Van Der Hammen T. 1961. Late Cretaceous and Tertiarystratigraphy and tectogenesis of the Colombian Andes.Geologie en Mijnbouw. Netherlands Journal of Geos-ciences 40(5), 181–188.

Van Der Hammen T, Wymstra TA. 1964. A palynologicalstudy of the Tertiary and Upper Cretaceous of BritishGuayana. Leidse Geologische Mededelingen 30, 183–241.

Wodehouse RP. 1933. Tertiary pollen II. The oil shales of theEocene green River Formation. Bull Torrey Bot. Club4(60), 479–524.

Wolda H. 1981. Similarity indices, sample size and diversity.Oecologia 50, 296–302.

16 G. Rodrıguez-Forero et al.

Dow

nloa

ded

by [

Smith

soni

an I

nstit

utio

n L

ibra

ries

] at

07:

24 2

3 A

pril

2012