Am. J. Bot. Join the BSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (12)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Graham, A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Graham, A.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Graham, A.
(American Journal of Botany. 1998;85:1409-1425.)
© 1998 Botanical Society of America, Inc.

Studies in Neotropical paleobotany. XI. Late Tertiaryvegetation and environments of southeastern Guatemala: palynofloras fromthe Mio-Pliocene Padre Miguel Group and the Pliocene HerreríaFormation1

Alan Grahama

a Department of Biological Sciences, Kent State University,Kent, Ohio 44242


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 LOCALITIES
 AGE AND STRATIGRAPHY
 COMPOSITION
 DISCUSSION
 REFERENCES
 
Plant microfossils are reported from the late Tertiary Padre MiguelGroup and the Pliocene Herrería Formation of southeasternGuatemala. The most abundant palynomorphs are cf. Acrostichum(maximum 45%), Cyperaceae (29%), cf. Antrophyum(27%), monolete fern spores (16%), and Pinus(11%). Pollen grains of Picea, Juglans,Quercus, and Ulmus, primarily from the Padre Miguelflora, reveal the presence of a northern cool-temperate element. Theyounger Herrería flora is more lowland and warm temperate. Inthe absence of evidence for substantially higher elevations, thedifferences between the Neogene and modern vegetation are attributedmostly to climate. MAT (mean annual temperature) is estimated2°–3°C cooler than at present for the Padre Miguel flora,and ~3.5°C warmer for the Herrería flora. There islittle evidence for arid vegetation, and the tropical rain forest wasabsent or poorly represented. These data are consistent with those ofother fossil floras in the region and with trends suggested by isotopicpaleotemperature analysis and global sea-level changes. The Guatemalaassemblages further provide evidence that the current version of thetropical rain forest is recent in origin and has undergone considerablechange in its range and composition throughout the lateCenozoic.

Key Words: Guatemala • Neogene • paleocommunities • paleoenvironment


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 LOCALITIES
 AGE AND STRATIGRAPHY
 COMPOSITION
 DISCUSSION
 REFERENCES
 
The Cenozoic history of the area between southern Mexico and northernSouth America is of considerable evolutionary and biogeographicimportance. The lands that constituted much of Central America duringthe Tertiary, particularly in southern Central America, providedhabitats for many terrestrial taxa that were geographically andreproductively isolated. Beginning ~3.5 Ma (million years ago)uplift and volcanic accumulations formed a low-lying, essentiallycontinuous land connection, and after ~2.5 Ma corridors suitable forthe exchange of upland biotas were established (Stehli and Webb, 1985; Coates et al., 1992; Graham, 1992a, inpress; Jackson, Budd, and Coates,1996; Webb and Rancy, 1996;Burnham and Graham, in press; 1.9 Mafide T. Donnelly, State University of New York, Binghamton; personalcommunication, 1997). In addition to physical connections, otherimportant factors influencing evolutionary history and migrationsthrough the region were climate and physiography. However, there are noknown plant macrofossil assemblages suitable for reconstructing Tertiarypaleoenvironments in the region. Some information is available fromplant microfossil studies for the Quaternary (Costa Rica: Martin, 1961; Horn,1985; Hooghiemstra et al.,1992; Islebe, Hooghiemstra, and van derBorg, 1995; Guatemala: Cowgill andHutchinson, 1966; Cowgill et al.,1966; Tsukada, 1966;Tsukada and Deevey, 1967; Deevey, Vaughan, and Deevey, 1977; Deevey, 1978; Leyden,1984; Leyden et al., 1993;Panama: Bartlett and Barghoorn, 1973;Bush et al., 1992). Other palynologicaldata are available for the Tertiary of Mexico, Costa Rica, Panama, andthe Antilles (see references in Table1). Missing has been information on the Tertiary plantcommunities, migrations, and paleoenvironments of northern CentralAmerica— sizable and geographically intermediate region thatincludes Belize, Guatemala, El Salvador, Honduras, and Nicaragua. In1989, samples of lignites, lignitic shales, and organic clays wereobtained from late Tertiary strata in southeastern Guatemala. Three ofthese contained well-preserved spores and pollen grains, and slides froma fourth locality were obtained from a previous study. Collectively,they provide the first insight into the late Tertiary vegetation,environments, and physiography of the region between ~3 and 5million years ago.


View this table:
[in this window]
[in a new window]
  		Table 1. Unit, age, location, and references to current report (this paper) and previous records cited in the text.

 

   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
LOCALITIES
 AGE AND STRATIGRAPHY
 COMPOSITION
 DISCUSSION
 REFERENCES
 
Small samples (10–15 g) were macerated in a mortar and pestle,transferred to polystyrene beakers, covered with 10% HCl, andallowed to stand overnight to remove carbonate minerals. The sampleswere centrifuged at 1500 rpm for 5 min, washed 4 times with distilledwater, and covered with HF in polystyrene beakers for 1–4 h toremove silicates. The residues were washed and placed inHNO3 overnight to dissolve lignin and other organic debris. The samples were then washed, rinsed with glacial acetic acid,acetolyzed with nine parts of acetic anhydride to one part ofH2SO4 for 5 min at 85°C, rinsed in glacialacetic acid, washed, and drained overnight. Melted glycerine jelly wasadded to the residues in the centrifuge tubes, 2–3 drops wereplaced on a slide, and the slides were sealed with CoverBond. Thepalynomorphs were examined with a Leitz Orthoplan Photomicroscope andphotographed at 400x using Panatomic X black and white film. Collections were made at seven sites (A-G), and well-preservedpalynomorphs were recovered from three (B, F, and G); slides were alsoreceived from a site designated as our locality 4. The samples andslides from each locality are identified by numbers (e.g., Locality B-1,slide 1). The location of the specimens on the slides is designated byEngland Slide Finder (ESF) coordinates. Identifications were made bycomparisons to a spore and pollen reference collection of modern speciesnumbering ~24 000 slides, and from illustations anddescriptions in the literature. Slides, residues, unprocessed samples,negatives, and duplicate prints are in the palynological collections atKent State University.


   LOCALITIES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 LOCALITIES
AGE AND STRATIGRAPHY
 COMPOSITION
 DISCUSSION
 REFERENCES
 
Locality B (14°43'N, 89°29'W; Figs. 1, 2) is on the eastern edge ofthe village of Vado Hondo toward Jocotán. It is shown on thequadrangle map, Jocotán 2360 III, mapped by David Crane andpublished by the Instituto Geografico (now Militar), and it is mapped asTQss (Tertiary/Quaternary sandstone). The exposure is ~2 m invertical height, and the sample yielding the best-preserved palynomorphs(B-1) is from a lignite layer near the base of the section. It belongsto the Padre Miguel Group, which is probably late Miocene toMio-Pliocene in age (see discussion on Age and Stratigraphy).



View larger version (183K):
[in this window]
[in a new window]
  		Figs. 1–4. Collection localities in southeastern Guatemala. 1,2. Locality B. 3,4. Locality F. Photograph by Shirley A. Graham.

 
Locality F (15°45'N, 88°42'W; Figs. 3, 4) is a mining pit at SanPedro La Cocona. The best sample (F-12) came from a dark clay layernear the top of the section. The sediments are from the base of theHerrería Formation and are probably Pliocene in age. Locality Gis along a roadside ditch ~1 km east of locality F at Frontera de laPavas and just across the road from an aboveground section of thepipeline that brings oil from the Petén to the port of Barrios. The total section is ~1 m in height and consists of thin layers ofgray clay, black organic clay, and lignite, capped by tuff (water-lainvolcanic ash). The best sample (G-3) is from a black organic clay nearthe top of the section. The sediments are also part of theHerrería Formation.

The fourth locality is along the landing-strip road 0.5 km west ofRio Olopa in the Esquipulas Valley. The area was mapped by BurkeBurkhart, and samples were collected in the 1960s by Burkart and sent toEsso Research Laboratory (now Exxon Company, USA) in Houston forprocessing. Preliminary identifications of the fossil spores and pollenwere made by William Elsik (cited as by Dan Jones in Burkart, Clemons, and Crane, 1973; W. Elsik,Exxon Company, USA; personal communication, 1990), and includeJuglans, Laevigatosporites, Latisporites,Punctatosporites, Deltoidospora,Pityosporites, Triporopollenites,Triatriopollenites, Tricolpites, andTricolopopollenites. The slides were provided to ourlaboratory by Elsik in 1990 for further study. The samples belong tothe Padre Miguel Group.


   AGE AND STRATIGRAPHY
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 LOCALITIES
 AGE AND STRATIGRAPHY
COMPOSITION
 DISCUSSION
 REFERENCES
 
The geology of Guatemala is complex, and the age and stratigraphy ofthe various formations are mostly poorly known. The collectionlocalities are in the vicinity of the Polochic-Motagua Fault Zone whichseparates the North American and Caribbean plates (Fig. 5). The region to the north ofthe fault zone belongs to the Maya or Yucatán Block, whichincludes Guatemala north of the zone, Belize, the YucatánPeninsula, and Mexico west to the Isthmus of Tehuantepec. TheGuatemala-Belize portion of the Maya Block has been almost completelyemergent since the Eocene. The region to the south of the fault zonebelongs to the Chortis Block, which includes southern Guatemala, ElSalvador, Honduras, and part of northern Nicaragua. The origin of thetwo blocks is uncertain, but it is thought that the Maya Blockoriginated in the present Gulf of Mexico prior to the fragmentation ofwestern Gondwana, while the Chortis Block came from the Pacific Coast ofMexico. The two blocks were sutured in the late Cretaceous (Donnelly et al., 1990), and the resultingdeformation created uplands that have existed throughout the Tertiary(Sierra de Chuacús, Sierra de las Minas, Montanas del Mico;Fig. 5).



View larger version (33K):
[in this window]
[in a new window]
  		Fig. 5. Index map of place names and localities.

 
The extent of displacement along the Polochic-Motagua Fault Zone isunsettled, and estimates range from up to 1000 km (Hess and Maxwell, 1953; Wadge and Burke, 1983; Burke et al., 1984) to only several kilometres(Anderson, Erdlac, and Sandstrom,1985). The estimate favored at present is that the Caribbeanplate has moved at least 130 km eastward on the Polochic Fault relativeto the North American plate (Burkart,1978, 1983, 1994; Burkart andMoreno, 1983; Burkart and Self,1985). Displacement began ~10 Ma (middle Miocene) andmost of the movement probably occurred during an interval of ~3.7 Ma(Deaton, 1982; Deaton and Burkart, 1984). Activity along theMotagua fault zone continues today (Fig.6) and produces devastating earthquakes in the region,including the notable one around El Progreso on 4 February 1976, whichinstantaneously produced 2 m of offset followed by another 2 m ofpostquake slippage (Espinosa, 1976;Plafker, 1976). Another consequenceof the intense tectonic activity has been to fragment the strata into amosaic of local fault-bound exposures whose ages are difficult todetermine and, in the absence of radiometric dates, difficult tocorrelate regionally with other sequences.



View larger version (172K):
[in this window]
[in a new window]
  		Fig. 6. Movement of the Caribbean plate (in the distance) eastward relative to the North American plate (foreground) as shown by displacement of the white highway line at the bridge. Photograph by Shirley A. Graham.

 
The Padre Miguel Group has a large surface outcrop in southeasternGuatemala, and it extends into western Honduras and northwestern ElSalvador (Burkart, Clemons, and Crane,1973). It was earlier thought to range in age from middleMiocene to Pleistocene (Burkart, Clemons, andCrane, 1973) because an upper unit, the San Jacinto Formation(Crane, 1965) was included. However,this formation is distinctly younger (~4 Ma; Reynolds 1977, 1980), making it Pliocene, and it is nowregarded as separate from the underlying Padre Miguel Group. Followingthe recent summary by Donnelly et al.(1990), locality B and the Esquipulas Valley site (ourlocality 4) in the Padre Miguel Group are considered late Miocene toMio-Pliocene in age. The number of types and the stratigraphic range ofpalynomorphs from the Padre Miguel Group are not sufficient to date thesediments more precisely. However, the modern aspect of the assemblagefavors an age toward the younger end of the estimate.

The Herrería Formation is probably slightly younger than thePadre Miguel Group (excluding the San Jacinto Formation). According toDonnelly et al. (1990), "Thecoast of Guatemala and southern Belize, continuing westward around mostof the shoreline of Lago de Izabal, is the locus of a nearly continuousbelt of varicolored claystones, subordinate marl and sandstone, andscattered lignite beds. It is evidently Pliocene, has a reportedthickness of 240 m, and was named the Herrería Formation byVinson (1962)." The general ageand stratigraphic relations of these units are summarized in Fig. 7.



View larger version (52K):
[in this window]
[in a new window]
  		Fig. 7. Generalized stratigraphic column for Tertiary strata in southeastern Guatemala (relative ages are not well established). Based on Donnelly et al. (1990) .

 

   COMPOSITION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 LOCALITIES
 AGE AND STRATIGRAPHY
 COMPOSITION
DISCUSSION
 REFERENCES
 
Twenty-three identified and 14 unknown spore and pollen types arerecognized from the Padre Miguel Group and the HerreríaFormations. In the following descriptions, the phrase "previousrecords" refers to other reports of the taxon from Tertiarypalynofloras in northern Latin America, and further information on theserecords is given in Table1. A summary of the composition, distribution of the taxaamong the localities, and numerical representations are given inTable 2.


View this table:
[in this window]
[in a new window]
  		Table 2. Composition and numerical representation of palynomorphs among collection localities in the Padre Miguel Group and Herrer;aaia Formation, southeastern Guatemala. Figures are percentages based on counts of 200. Dashes indicate presence at <0.5%.

 
Operculodinium centrocarpus (Pyrrophyta; Fig. 8). This dinoflagellate cyst isproduced by the modern Gonyaulax grindleyi (=Protoceratium reticulatum), and it is common in estuarine andother nearshore marine sediments of Tertiary age. Locality G-3, slide1, ESF N-33,1–2, Herrería Formation. Previous record: LaBoca Formation.



View larger version (136K):
[in this window]
[in a new window]
  		Figs. 8–18. Plant microfossils from the late Tertiary Padre Miguel Group and the Herrería Formation of southeastern Guatemala. See text for descriptions and measurements, and Table 2 for occurrences among the localities and numerical representations. 8. Operculodinium . 9,10. Lycopodium . 11. Selaginella . 12. Monolete fern spore type 2. 13. Monolete fern spore type 1. 14. Monolete fern spore type 3. 15. Monolete fern spore type 4. 16. cf. Acrostichum . 17. cf. Antrophyum . 18. Pinus .

 
Lycopodium (Lycopodiaceae; Figs. 9, 10). Oblate, amboval-triangular to nearly circular, trilete, arms of laesura straight,inner margin entire, 15–18 µm long, extending to spore margin;proximal face laevigate, distal face rugulate with spaces betweenrugulae appearing as narrow, elongated, slit-like openings, wall2–3 µm thick; size 30–34 µm. Locality 4, slide 2, ESFV-52,3–4, Padre Miguel Group. Previous records: Culebra, Gatun,La Boca, Paraje Solo, San Sebastian, and Uscari formations/sequence.

The spores are similar to those of such modern species as L.dichotomum, distributed from the United States to South America. The modern species occurs in a wide variety of habitats, includingtropical moist and premontane wet forests, and in damp sites withinother drier vegetation types.

Selaginella (Selaginellaceae; Fig. 11). Oblate-spheroidal; trilete,laesura occasionally obscured by folding or dense sculpture elements(specimen split along lasura); echinate, spines 4–6 µm inlength; wall 2 µm thick; size (specimen compressed) 27 x 18µm excluding spines. Locality 4, slide 2, ESF P-37,2, Padre MiguelGroup. Previous records: Cucaracha, Culebra, Gatun, Gatuncillo, LaBoca, Paraje Solo, San Sebastian, and Uscari formations/sequence.

Selaginella is widespread in northern Latin America andgrows primarily in wet shaded habitats, although it ranges into moistsites in pine-oak woods and into even drier communities. Selaginella horizontalis is an example of one modern specieshaving microspores similar to the specimen (cf. Roubik and Moreno P., 1991, p. 153, fig. 7). Only microspores were recovered, and these are frequent but not abundantin the Tertiary of northern Latin America.

Monolete fern spore type 1 (Fig.13). Reniform; monolete, laesura straight, narrow, situatedalong concave margin of spore, 28–30 µm long, extendingapproximately three-quarters spore length, inner margin entire;laevigate; size 42–45 x 27–30 µm. Locality 4,slide 1, ESF P-31,2–4, Padre Miguel Group.

Variations on this spore type include transitions to faintly granularand distinctly smaller ones (type 2, 21–12 µm; Fig. 12, Locality 4, slide 1, ESF X-54,1–3), verrucate forms (type 3, Fig. 14, Locality B-1, slide 1, ESFO-42,1–2), and those distinctly verrucate (type 4, Fig. 15, Locality B-1, slide 1, ESFR-58,4). These types are produced by several members of theBlechnaceae, Polypodiaceae, and Pteridaceae (Tryonand Lugardon, 1991), and the microfossils range widelystratigraphically and geographically throughout the Tertiary of theCaribbean region.

cf. Acrostichum (Pteridaceae; Fig. 16). Oblate to oblate-spheroidal,amb triangular to oval-triangular, apices rounded; trilete, laesurasmall in relation to spore diameter, straight, narrow, 12–15 µmlong, extending one-half to two-thirds distance to spore margin, innermargin entire; granular; wall ~1.5 µm thick; size 45–50µm. Locality F-12, slide 1, ESF S-32,2, HerreríaFormation.

The spores are characterized by their relatively large size, alaesura that is small in relation to the size of the spore, and agranular surface. They are similar to those of Acrostichumaureum, which is widespread but grows along coasts in brackishwater. The presence of this habitat is further indicated by theRhizophora and marine dinoflagellates found at locality F.

cf. Antrophyum (Vittariaceae; Fig. 17). Oblate to oblate-spheroidal,amb oval-triangular; trilete, laesura straight, narrow, 20–22µm long, extending two-thirds to nearly to spore margin, inner marginentire; laevigate; wall ~2 µm thick; size 60 µm. LocalityG-3, slide 1, ESF J-25,3–4, Herrería Formation. Previousrecords: Artibonite, Cucaracha, Culebra, Gatuncillo, and La Bocaformations/group.

Antrophyum commonly grows in rain forests and in cloudforests extending up to ~1500 m in elevation.

Picea (Pinaceae; Fig.19). Monocolpate, colpus extending between air sacs on lowerside of body; vesiculate, air sacs 2, approximately spherical,reticulate and grading into sculpture pattern of body, ~24 x18 µm, body scabrate, 55–62 x 28–30 µm; noreentrant angle, no marginal flange; wall ~2 µm thick; size62–68 x 41–45 µm. Locality B-1, slide 1, ESF K-57,Padre Miguel Group. Previous record: Paraje Solo Formation (withmicrofossils of Abies).



View larger version (149K):
[in this window]
[in a new window]
  		Figs. 19–38. Plant microfossils from the late Tertiary Padre Miguel Group and the Herrería Formation of southeastern Guatemala. See text for descriptions and measurements; Table 2 for occurrences among the localities and numerical representations. 19. Picea . 20. Cyperaceae. 21. Palmae. 22. Gramineae. 23. Iresine . 24,25. Compositae. 26,27. Quercus . 28. Alfaroa -Oreomunnea . 29. Juglans , 30. Ulmus . 31. Caesalpinia . 32. Ludwigia . 33. Unknown 1. 34. Unknown 2. 35. Unknown 3. 36. Unknown 4. 37. Unknown 5. 38. Unknown 6.

 
Picea does not presently occur in Guatemala, although itscommon associate and the ecologically similar Abiesguatemalensis grows in the coniferous forest at high elevationswith pine (Pinus ayacahuite, P. hartwegii, P.montezumae). The southernmost sizable stands of spruce are now inthe mountains of northern Mexico (P. mexicana, Coahuila, NuevoLeon; P. chihuahuana, Chihuahua, Durango). A few fragmentedand/or poorly oriented winged conifer pollen at locality G(Herrería Formation) may include some Picea, but themost abundant and well-preserved specimens were from the Padre MiguelGroup.

Pinus (Pinaceae; Fig.18). Monocolpate, colpus extending between air sacs on lowerside of body; vesiculate, air sacs 2, approximately spherical,reticulate, and distinct from sculpture pattern of body, 27–30µm in diameter, body scabrate, 50–55 x 25–27 µm,reentrant angle present, marginal flange present to obscure; wall ~2µm thick; size 62–67 x 35–38 µm. Locality G-3,slide 2, ESF G-37,1–2, Herrería Formation. Previousrecords: Artibonite, Paraje Solo, and Simojovel (La Quinta Formation)groups/formation.

Pinus ayacahuite, P. hartwegii, P.montezumae, P. oocarpa, and P. strobiliformis aremembers of the subalpine coniferous forests of Guatemala, and pine-oakforests are presently a prominent vegetation type on drier sites atmidelevations in much of Mesoamerica. The occurrence of Piceaand Quercus microfossils also suggests the presence of bothcommunities in the Late Tertiary of Guatemala. Pinus is alsopart of the low savanna vegetation of Izabal and Petén, butpollen of the defining components of that community (e.g.,Byrsonima, Curatella) was not present.

Cyperaceae (Fig. 20). Cuneiform; triporate, lateral pores often indistinct, principal porecircular, granular, margin diffuse, located at apex of wedge-shapedgrain, ~8 µm in diameter; scabrate; tectate, wall 0.5–1µm thick, homogeneous (400x magnification); size 30 x 24µm (widest part). Locality 4, slide 1, ESF N-45,3, Padre MiguelGroup. Previous record: Paraje Solo Formation.

Gramineae (Fig. 22). Spherical, amb circular; monoporate, pore circular, 3–4 µm indiameter, inner margin entire, surrounded by annulus 3–4 µmwide; psilate to scabrate; tectate, wall ~2 µm thick, homogeneous(400x magnification); size 28–30 µm. Locality 4, slide1, ESF G-59,2–4, Padre Miguel Group. Previous records: Culebra,Gatun, La Boca, Paraje Solo Formations.

Palmae (Fig. 21). Prolate;monocolpate, colpus straight, 32–34 µm long, extending nearlyentire length of grain, inner margin entire to finely dentate; scabrate;tectate, wall ~2 µm thick, columellae evident (400xmagnification); size 38–40 x 17–19 µm (widest partjust off equator of grain). Locality B-1, slide 2, ESF O-37,2, PadreMiguel Group. Previous records (this type): Artibonite, Cucaracha,Culebra, Gatun, Gatuncillo, La Boca, Paraje Solo formations/group.

Iresine (Amaranthaceae; Fig.23). Spherical, amb polygonal; periporate, pores circular toslightly elongate, margins entire to minutely dentate, numerous(~16), equidistant, ~4 µm apart; scabrate; tectate, wall~1.5 µm thick, columellae just evident (400xmagnification); size 17–19 µm. Locality B-1, slide 1, ESFV-49,3, Padre Miguel Group. Previous record: Paraje Solo Formation.

Iresine is a widespread, tall herb (to 3 m), most common inopen habitats, and found in vegetation types ranging from tropical dryto tropical wet forests. Roubik and Moreno P.(1991, p. 182, fig. 351) illustrate pollen of I.celosia-I. diffusa from Barro Colorado Island similar tothe fossil specimens.

Compositae (Figs. 24, 25). Prolate-spheroidal to nearly spherical, amb circular; tricolporate,colpi equatorially arranged, meridionally elongated, equidistant,straight, ~15 µm long, extending nearly the full length of thegrain, narrow costi colpi, pore small (~1 µm in diameter),circular, situated at midpoint of colpus; echinate, spines 2–3µm long, dense, slightly curved, broadening toward the base; tectate,wall ~3 µm thick, columellae just evident (400x magnification);size 22–24 µm. Locality 4, slide 2, ESF V-35 (Fig. 24), Locality B-1, slide 1, ESFE-58 (Fig. 25), Padre MiguelGroup. Previous records (this type): Paraje Solo, Uscariformation/sequence.

Compositae pollen is frequent and often abundant in late Tertiarydeposits of the Gulf/Caribbean region. The small, mid- to high-spineforms are produced by many members of the tribesAstereae-Heliantheae-Helineae. The geologic record of the family issummarized by Graham (1996).

Quercus (Fagaceae; Figs. 26,27). Prolate to prolate-spheroidal, amb circular; tricolpate,colpi equatorially arranged, meridionally elongated, equidistant,straight, inner margin entire to minutely dentate, 20–22 µmlong, extending two-thirds to nearly entire length of grain; scabrate;tectate, wall ~1.5 µm thick, columellae just evident (400xmagnification); size 26–28 x 20–22 µm. LocalityB-1, slide 1, ESF E-61,2 (Fig.26), Locality 4, slide 1, N-58,2 (Fig. 27), Padre Miguel Group. Previousrecords: Gatun, Paraje Solo Formations.

Quercus is presently widespread in temperate towarm-temperate habitats at midelevations in Mesoamerica, includingGuatemala (Q. acatenangensis, Q. borucasana), andalong with Pinus, it forms the extensive pine-oak forest ofMexico and Guatemala. Quercus is also a prominent member ofthe upland mixed broad-leaved forest at midelevations in moisterhabitats. Both genera appear late in northern Latin America based onthe plant microfossil floras presently known from the region. Quercus is common in the middle Pliocene Paraje Solo Formationof Veracruz, Mexico (maximum 34%), it is less abundant in thelate Miocene-Pliocene Padre Miguel Group of Guatemala (12%), andfirst appears in small amounts (<0.5%) in Panama in theMio-Pliocene Gatun Formation; viz., it arrives progressively later andis less abundant toward the south. Quercus does not appear innorthern South America until the Quaternary (Hooghiemstra, 1989, 1994; Hooghiemstra and Sarmiento, 1991; Hooghiemstra and Ran, 1994).

Alfaroa/Oreomunnea (Juglandaceae; Fig. 28). Oblate, amb oval-triangular;triporate, pores equatorially arranged, equidistant, circular, 3–4µm in diameter, inner margin entire; psilate to finely scabrate;tectate, wall 2 µm thick, homogeneous (400x magnification);size 22–24 µm. Locality G-3, slide 1, ESF G-30,2–4,Herrería Formation. Previous records: Artibonite, Cucaracha,Gatun, Gatuncillo, La Boca, Paraje Solo, San Sebastian, Simojovelgroup/formations.

Following the treatment of Stone(1972; Stone and Broome,1975), the New World species of this complex belong toAlfaroa (seven species) and Oreomunnea (two species)scattered in wet mountain and premontane cloud forest from southernMexico to northern South America. By this treatment,Engelhardia is an Old World genus. In the modern vegetation ofnorthern Latin America Alfaroa/Oreomunnea often occursin association with Podocarpus in habitats more moist thanthose supporting the other prominent midaltitude temperate community ofPinus and Quercus. Alfaroa/Oreomunnea appears earliest among the Tertiarypalynofloras from northern Latin America. It is present in the lateEocene Gatuncillo flora of Panama, while pollen of Podocarpus(not recovered from the Guatemala samples) is found first in theOligocene San Sebastian Formation of Puerto Rico, and later in theSimojovel, Uscari, Paraje Solo, and Gatun formations/sequence. However,it was likely present earlier because macrofossils ofPodocarpus occur in the Eocene of southeastern United States(Dilcher, 1969).

Juglans (Juglandaceae; Fig.29). Oblate-spheroidal, amb circular; multiporate, heteropolar(pores in one hemisphere), circular, 2–3 µm in diameter, innermargin entire, equidistant (10–12 µm apart), annulus 3–4µm wide; psilate to faintly scabrate; tectate, wall ~2 µmthick, homogeneous (400x magnification); size 27–31 µm. Locality 4, slide 1, ESF H-56, Padre Miguel Group. Previous record:Paraje Solo Formation.

In the New World, Juglans is presently distributed fromeastern Canada to Argentina (Manning1957, 1960). In GuatemalaJuglans olanchana and J. steyermarkii grow in thehumid, temperate, wet mountain forest at midelevations between 1000 and1300 m. Juglans was likely associated withAlfaroa-Oreomunnea and Ulmus in this foresttype. Juglans is not known in the modern vegetationimmediately to the south in El Salvador, Costa Rica, or Panama, andthere are no fossil records to document its presence in those areasduring the Tertiary. However, Juglans is now relativelywidespread in South America, there is a fossil seed from presumed lateMiocene deposits in Ecuador (Brown,1946), and it becomes a regular component of the vegetationaround Bogotá, Colombia ~2.2 Ma (van der Hammen and Hooghiemstra, 1997). How and when it dispersed from northern Central America to South Americais not known.

Caesalpinia (Leguminosae, Caesalpinioideae; Fig. 31). Oblate-spheroidal, ambcircular; tricolporate, colpi equatorially arranged, meridionallyelongated, equidistant, straight, short, 8–10 µm, inner marginminutely dentate, surrounded by broad granular membrane (margocolpus)nearly equal in width (12–14 µm) to mesocolpal area, poresituated at midpoint of colpus, circular, 4–5 µm in diameter,inner margin entire; reticulate, diameter of lumen 0.5–1 µm,approximately equal to width of muri of reticulum, muri surface smooth;tectate-perforate, wall ~2 µm thick, columellae evident(400x magnification); size 36 µm. Locality G-3, slide 1, ESFF-16,4, Herrería Formation. Previous records: none.

Caesalpinia is currently being revised by G. Lewis (HerbarioLOJA, Ecuador, personal communication, 1997). In the broadest sense itis a pantropical genus of ~120–130 species of which70–75% are strictly neotropical. In Mesoamerica, ittypically occurs in dry open habitats as a member of dry tropical, opendeciduous, and thorn forests on calcareous or sandy soils, although itranges into other vegetation types on dry sites. Many species grow incoastal habitats, and most do not occur at elevations above 1100 m. However, Caesalpinia nicaraguensis (a new species) is anarrowly restricted endemic in central Nicaragua, and grows in oak ormixed pine and oak forest between 1100 and 1400 m, while C.laxa in eastern Mexico is known from 1400 and 2000 m.

In northern South America fossil pollen of Caesalpinia isknown from the middle Eocene to the Recent (Germeraad, Hopping, and Muller, 1968). Ithas been reported as leaflets from the Miocene of Cuba (Berry, 1939), and as pollen from Holocene(essentially modern) sediments of Haiti (Higuera-Gundy, 1989). Caesalpinia pollen has not been reported previously from theTertiary of northern Latin America. The fossil record of theLeguminosae (Fabaceae) for the Gulf-Caribbean region has been summarizedby Graham (1992b, see also Graham, 1993; Grahamand Dilcher, 1995), and the pollen morphology of theCaesalpinioideae is described by Graham andBarker (1981).

Ludwigia (Onagraceae; Fig.32). Oblate, amb oval-triangular; triporate, poresequatorially arranged, equidistant, circular, 10–12 µm indiameter, surrounded by conspicuous costae pori 4–6 µm wide;scabrate; tectate, wall 2–3 µm thick, columellae just evident(400x magnification); size 45 µm. Locality B-1, slide 1, ESFH-30,1, Padre Miguel Group. Previous record: Paraje Solo Formation.

In Mesoamericana, Ludwigia is a genus of 21 subtropicalherbaceous species (P. Hoch, Missouri Botanical Garden, personalcommunication, 1997). In addition to being a common roadside weed, itgrows in open habitats within fern-palm swamps and marshes. Thepresence of fern spores and palm pollen in the fossil assemblagesindicates the availability of these habitats in the late Tertiary ofGuatemala.

Rhizophora (Rhizophoraceae; for descriptions andillustrations see Graham and Jarzen,1969; Graham, 1976a, 1985, 1988a;Graham and Palacios Chávez,1996). In one sample (F-12) the sediments were separatedduring processing into <20 µm and >20 µm fractions, and the<20 µm part contained virtually all Rhizophora pollen. This was the only sample having Rhizophora pollen. The >20µm fraction was almost all cf. Acrostichum.

Ulmus (Ulmaceae; Fig.30). Oblate, amb approximately square; tetraporate, poresequatorially arranged, situated at corners of grain, small, ~2 µmin diameter, inner margin entire, surrounded by annulus 2–3 µmwide; rugulate; tectate, wall 2–3 µm thick, homogeneous(400x magnification); size 23–25 µm. Locality 4, slide2, ESF C-50,4, Padre Miguel Group. Previous record: Paraje SoloFormation.

Ulmus is represented in Guatemala by U. mexicana,which is distributed from Chiapas, Mexico, to Panama. It is part of thetemperate subcloud forest and grows between 175 and 1500 m elevation. It was likely associated with Alfaroa-Oreomunnea andJuglans in the midelevation temperate forest on sites moremoist than those supporting the drier pine-oak communities at similarelevations.

Unknown 1 (Fig. 33). Oblate, amb triangular, apices rounded; laesura not evident (trilete?);reticulate, diameter of lumen ~6 µm, muri narrow, ~2 µmwide, smooth, straight to slightly sinuous; wall 2–3 µm thick,some supporting columellae 4–6 µm long, narrow, ~2 µmwide; size 54 µm. Locality G-3, slide 1, ESF R-32,4, HerreríaFormation. Previous records: none.

Unknown 2 (Fig. 34). Prolate, tricolpate, colpi equatorially arranged, meridionallyelongated, equidistant, straight, inner margin entire to finely lobate,12–15 µm long; verrucate, verrucae low, smooth, mound-likestructures, dense (~1 µm apart); tectate, wall 2–3 µmthick, homogeneous (400x magnification); size 22–24 x16–18 µm. Locality G-3, slide 1, ESF L-30, 3–4 (see alsoR-41; slide 2, Q-31,3); Herrería Formation. Previous records:cf. La Boca Formation (Graham, 1989, p.62, figs. 55, 56); Paraje Solo Formation (Graham,1976a, p. 840, figs. 244–247).

Unknown 3 (Fig. 35). Prolate; tricolporate, colpi equatorially arranged, meridionallyelongated, equidistant, straight, 16–18 µm long, inner marginentire to minutely dentate, pore faint, circular, situated at midpointof colpus; finely reticulate; tectate-perforate, wall 2–3 µmthick, columellae evident (400x magnification), size 23–25x 17–19 µm. Locality G-3, slide 2, Q-21,3,Herrería Formation.

Unknown 4 (Fig. 36). Prolate; tricolpate, colpi equatorially arranged, meridionallyelongated, equidistant, straight, 10–12 µm long, inner marginentire to minutely dentate, narrow costae colpi ~2 µm wide;finely reticulate, diameter of lumen approximately equal to width ofmuri of reticulum, muri surface smooth; tectate-perforate, wall 2 µmthick, columellae evident (400x magnification); size 15 x 12µm. Locality 4, slide 1, ESF L-47,2, Padre Miguel Group.

Unknown 5 (Fig. 37). Prolate; tricolporate, colpi equatorially arranged, meridonallyelongated, equidistant, straight, 19–21 µm long, inner marginentire, costae colpi 3–4 µm wide, pore circular, situated atmidpoint of colpus, 2–3 µm in diameter; finely reticulate,diameter of lumen approximately equal to width of muri of reticulum,muri surface smooth; tectate-perforate, wall 2 µm thick, columellaeevident (400x magnification); size 26–28 x 17–19µm. Locality 4, slide 2, ESF F-45,2–4, Padre Miguel Group.

Unknown 6 (Fig. 38). Prolate; tricolporate, colpi equatorially arranged, meridionallyelongated, equidistant, straight, 14–16 µm long, inner marginentire, costae colpi ~3 µm wide, pore circular, situated atmidpoint of colpus, 2–3 µm in diameter; very finely reticulate,diameter of lumen equal to width of muri of reticulum, muri surfacesmooth; tectate-perforate, wall ~2 µm thick, columellae evident(400x magnification); size 22–24 x 12–14 µm. Locality 4, slide 1, ESF G-28, Padre Miguel Group.

Unknown 7 (Fig. 39). Spherical, amb circular; triporate, pores circular, equatoriallyarranged(?), margin diffuse, granular membrane, 4–5 µm indiameter; finely reticulate, diameter of lumen equal to or slightlygreater than width of muri of reticulum, muri surface smooth;tectate-perforate, wall ~3 µm thick, columellae evident(400x magnification); size 39–42 µm. Locality G-3, slide1, ESF E-18,2–4, Herrería Formation.



View larger version (79K):
[in this window]
[in a new window]
  		Figs. 39–46. Plant microfossils from the late Tertiary Padre Miguel Group and the Herrería Formation of southeastern Guatemala. See text for descriptions and measurements; Table 2 for occurrences among the localities and numerical representations. 39. Unknown 7. 40. Unknown 8. 41. Unknown 9. 42. Unknown 10. 43. Unknown 13. 44. Unknown 11. 45. Unknown 12. 46. Unknown 14.

 
Unknown 8 (Fig. 40). Spherical, amb circular; apertures obscure (triporate?); reticulate,diameter of lumen 4–5 µm, polygonal, muri narrow (~2µm), straight to slightly sinuous, muri surface smooth;tectate-perforate, wall ~4 µm thick, columellae evident(400x magnification), some long (5–6 µm); size34–36 µm. Locality B-1, slide 1, ESF E-50,1–3, PadreMiguel Group.

Unknown 9 (Fig. 41). Oblate, amb circular; tetracolpate, colpi equatorially arranged,meridionally elongated, equidistant, short (~6 µm), conspicuouscostae colpi 3–4 µm wide; reticulate, diameter of lumen large(4–5 µm) in relation to width of muri (~2 µm), muristraight to slightly sinuous, surface smooth; tectate-perforate, wall~2 µm thick, columellae evident (400x magnification); size36 µm. Locality B-1, slide 1, ESF Q-67,4, Padre Miguel Group.

Unknown 10 (Fig. 42). Oblate-spheroidal, amb circular; tricolpate, colpi equatoriallyarranged, meridionally elongated, equidistant, straight, inner marginentire to minutely dentate, 8–9 µm long (apex to equator);finely reticulate, diameter of lumen approximately equal to width ofmuri of reticulum, muri surface smooth; tectate-perforate, wall ~2µm thick, columellae evident (400x magnification); size23–25 µm. Locality G-3, slide 1, ESF Q-38,1–3,Herrería Formation.

Unknown 11 (Fig. 44). Spherical, amb circular; nonaperturate (apertures obscure?); coarselyreticulate, lumina polygonal, diameter 2–8 µm, muri ~2µm wide, straight, surface smooth; tectate-perforate, wall 4–5µm thick, columellae evident (400x magnification); size26–28 µm. Locality B-1, slide 2, ESF G-42,1–2, PadreMiguel Group. Previous records: Culebra, Gatun, La Boca Formations.

These specimens are similar to the pollen of several Rubiaceae,including Anisomeris, Chomelia, Guettarda,and Terebaria.

Unknown 12 (Fig. 45). Oblate, amb oval-triangular to circular; tricolpate, colpi equatoriallyarranged, meridionally elongated, equidistant, short (3–4 µmequator to apex), inner margin entire, apices rounded, conspicuouscostae colpi ~2 µm wide; finely reticulate, diameter of lumenapproximately equal to width of muri of reticulum, muri surface smooth;tectate-perforate, wall ~2 µm thick, columellae evident(400x magnification); size 18 µm. Locality B-1, slide 2, ESFS-67, Padre Miguel Group. Previous records: Gatun, Gatuncillo, ParajeSolo Formations.

The specimen is of the same general type as those recovered from theParaje Solo Formation (Graham, 1976a,p. 840, figs. 248–255). It probably belongs to theSterculiaceae-Tiliaceae complex and may be Mortoniodendron(Graham, 1979).

Unknown 13 (Fig. 43). Prolate-spheroidal to spherical, amb circular; periporate, poresapproximately uniformly distributed (2–4 µm apart), circular,2–3 µm in diameter, inner margin entire; psilate; tectate(?),wall 1–2 µm thick, homogeneous (400x magnification); size30–32 µm. Locality 4, slide 2, ESF V-35,4, Padre MiguelGroup.

The specimens differ in wall structure (columellae not evident) andin sculpture (psilate) from the periporate Chenopodiaceae-Amaranthaceaein which columellae are evident in median optical section, and whichhave a scabrate sculpture pattern (or appear scabrate due to thesupporting columellae that are evident through the thin sexine). It issimilar to the zygospores/aplanospores of Zygnema (Chlorophyta;Hooghiemstra, 1984, fig. 401;van Geel, 1976; van Geel and van der Hammen, 1978; see alsoGraham, 1971).

Unknown 14 (Fig. 46). Thespecimen is similar to Unknown 13 except that it is smaller (18–20µm). Locality 4, slide 1, ESF K-54, Padre Miguel Group. It issimilar to the ascospore Gelasinospora (Sordariaceae; Hooghiemstra, 1984, fig.359).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 LOCALITIES
 AGE AND STRATIGRAPHY
 COMPOSITION
 DISCUSSION
REFERENCES
 
Modern vegetation
The modern plant communities of Guatemala have not been characterizedin detail with respect to composition, distribution, and ecologicalparameters. The Flora Mesoamericana (Davidse,Sousa S., and Chater, 1994 et seq.) is providing an inventoryof the species, and studies have been made on alpine and subalpinevegetation in Guatemala and Costa Rica (Islebeand Kappelle, 1994; Islebe,1996), and on submontane forests in El Salvador (Berendshon,1991). Descriptions of the extant vegetation as part of modern pollenrain studies are available for parts of highland Guatemala (Islebe and Hooghiemstra, 1995), and forsea-level mangrove to paramo communities of Costa Rica (Rodgers and Horn, 1996). However, the onlygeneral summaries of the modern communities of Guatemala are those ofLundell (1937), Standley and Steyermark (1945), and Steyermark (1950; Fig. 47).



View larger version (85K):
[in this window]
[in a new window]
  		Fig. 47. Vegetation map of Guatemala. Based on Standley and Steyermark (1945) and Steyermark (1950) .

 
The vegetation of Guatemala may be arranged into the following 13generalized categories (with prominent genera listed) modified fromStandley and Steyermark (1945).

1) Mangrove swamp—Acrostichum, Avicennia,Conocarpus, Laguncularia, and Rhizophora.

2) Freshwater swamps and marshes—ferns, palmae, Cyperaceae,Gramineae, Pachira, Thalia, and Typha.

3) Rain forest of the Atlantic region—Andira, Araceae,Bauhinia, Bromeliaceae, Calocarpum,Calyptogyne, Cohune, Chamaedorea,Chrysophila, Entada, Ficus,Hexopetion, Orbignya, Orchidaceae, andPeperomia.

4) Limestone plain of Petén—Achras,Brosimum, Calophyllum, Crysophila,Ficus, Piper, Psychotria, Rheedia,and Swietenia.

5) Low savanna of Izabal and Petén—open stands ofPinus caribaea, Byrsonima, Curatella,Compositae, Cyperaceae, Gramineae, and Leguminosae.

6) Mixed forest of the Pacific plains (scrub, pronounced wet-dryseason, high temperatures)—Bursera, Ceiba,Cochlospermum, Gliricidia, Sterculia, andTabebuia.

7) Arid desert plains (chaparral of the western plateaus, Valley ofthe Rio Motagua)—Acacia, Acanthocereus,Aristida, Bacopia, Bouteloua,Bursera, Cassia, Cephalocereus,Conobea, Crescentia, Crumenaria,Cyperus, Diodia, Echinodorus,Eichhornia, Erythroxylon, Evolvulus,Fimbristylis, Gomphrena, Jacquinia,Juliania, Karwinskia, Lemairocereus,Limnobium, Pacourina, Panicum,Pereskia, Phoradendron, Portulaca,Ruprechtia, Scleria, Sida, andTalinum.

8) Wet montane forest of Alta Verapaz—Anthophytum,Podocarpus, Palmae, Alfaroa-Oreomunnea,Berchemia, Bromeliaceae, Calocarpum,Carpinus, Gelseminum, Hymenaea,Juglans, Liquidambar, Lycaste (L.skinneri, monja blanca- the national flower of Guatemala),Magnolia, Orchidaceae, Persea, Rhus,Ulmus, and Vochysia.

9) Mixed montane forest of the Pacific bocacosta [middle humidslopes (to ~1000 m), southwestern Guatemala]—ferns, treeferns, Billia, Dussia, Erblichia,Heisteria, Louteridium, Luania,Mollinedia, and Sloanea.

10) Upland mixed broad-leaved forest of temperate and cold regions(1500–2300 m)—tree ferns, Alnus, Araceae,Begoniaceae, Bromeliaceae, Cestrum, Cheiranthodendron,Cornus, Dahlia, Garrya, Geranium,Heliocarpus, Ilex, Olmediella, Orchidaceae,Oreopanax, Ostrya, Prunus, Quercus,Rojasianthe, Roupala, Sambucus, andTurpinia.

11) Pine-oak forest—This community is not described or mappedas such on the older vegetation maps (Standleyand Steyermark, 1945; Steyermark,1950), and it has been severely reduced in extent in recenttimes. In general, pine-oak communities occur at the mid- to lowerelevations of areas mapped as coniferous forest in Fig. 47. Thus, slight range extensionsfrom the west and south would bring them into the immediate vicinity ofcollection localities B and 4.

12) Coniferous forest (1600–4200 m)—Abies,Cupressus, Juniperus, Pinus,Taxodium, Acaena, Alchemilla,Arracacia, Buddleia, Fuchsia,Lycianthes, Oreopanax, Oxylobus,Pernettya, and Rubus.

13) Alpine regions (usually above 3300 m)—Alchemilla,Aplopappus, Arctostaphylos, Arenaria,Draba, many Gramineae (e.g., Festuca and othernorthern genera), Geranium, Gnaphalium,Luzula, Muehlenbeckia, Potentilla,Selerianum, Vaccinium, Weldenia, andWerneria.

Paleocommunities andphysical setting
One of the differences between the assemblages of this study and manyof the other plant microfossil floras listed in Table 1 is the absence or rarity ofmangrove pollen in most of the Guatemala samples (Avicennia,Conocarpus, Laguncularia, Pelliceria, andRhizophora). With the exception of the Artibonite flora,deposited in an upland pine forest, and the middle Eocene Chapeltonflora, predating the arrival of modern mangroves in the region, theother fossil floras include Rhizophora as an importantcomponent. There are relatively low maximum amounts in the middle(?) tolate Eocene Gatuncillo flora (10%) when Rhizophora wasjust entering the region (Graham,1995), and in the Culebra flora (10.5%) with highpercentages of ferns and palms suggesting fresh water marsh and swamphabitats. Otherwise, maximum amounts range up to 96% in samplesfrom the San Sebastian and Paraje Solo floras, and to 95% atSimojovel. These assemblages were deposited in coastal environmentsunder sustained brackish water conditions.

Sediments at localities F and G (Herrería Formation) weremostly deposited in freshwater swamps, marshes, and estuaries withabundant ferns, and the sites were periodically innundated by brackishwater. At Locality F, fern spores constitute 63% of thepalynomorphs, and at Locality G 44% of the assemblage, comparedto 16 and 4% at localities B and 4, respectively (Padre MiguelGroup; Table 2). Theoccasional occurrence of brackish water is indicated by the presence ofcf. Acrostichum (45%) and Rhizophora (virtually100% sample 12) only at Locality F, and the dinoflagellateOperculodinium (6%) at Locality G, but with theexception of sample F-12, these incursions generally did not allow forthe establishment of widespread or sustained mangrove communities.

In addition to tectonics (movement of the land), higher sea levels inlate Tertiary may have extended the influence of brackish water fartherinland, especially along estuaries, which would account for the highrepresentation of cf. Acrostichum, and Rhizophora inone sample, in the Herrería Formation. According to Cronin and Dowsett (1991), the middle Pliocene(3–4 Ma; Planktic Foraminiferal Zone 19) was the last time theearth was substantially warmer than at present (by ~3.5°C). At3.1 Ma sea level was 35 ± 18 m higher along the Atlantic coastthan at present (Dowsett and Cronin,1990), and the shoreline extended farther inland there than atany time since the Eocene (Cronin,1991; Krantz, 1991). To theextent that eustatically higher sea level partly accounts for theabundance of cf. Acrostichum spores and Rhizophorapollen in the Pliocene Herrería Formation, this affords anopportunity to estimate the age of the formation more precisely. In theabsence of radiometric dates, if the Herrería Formation wasdeposited during a time of substantially higher sea level, this suggestsan age of middle Pliocene (Fig.48). The estimate will be assessed as new information becomesavailable.



View larger version (19K):
[in this window]
[in a new window]
  		Fig. 48. Sea-level fluctuations derived from deep-ocean O records with inferred age of the Padre Miguel Group and Herrería Formation. The two curves define the range of highstands and lowstands. Based on Krantz (1991) .

 
At the more inland and older localities B and 4 there was little orno influence of brackish water. Spores of cf.Acrostichum and pollen of Rhizophora wereabsent, and only a single specimen of dinoflagellate was found atLocality 4. The most abundant pollen type at Locality 4 was Cyperaceae(29%) suggesting a sedge marsh, with Compositae (9%),Gramineae (6%), and Palmae (4%) contributing pollen fromplants growing in the marsh or in adjacent communities. Surrounding themarsh on drier upland sites were forests of pine (11%) and oak(7%), and on more moist sites there was a temperate forest ofAlfaroa-Oreomunnea (9%), Juglans(2%), and Ulmus (2%). At locality B there isless sedge pollen (7%) than at locality 4, but otherwise thepaleocommunities are similar.

Another feature of these late Tertiary paleocommunities fromsoutheastern Guatemala is the absence of spores and pollen from rainforest plants. There were no spores of tree ferns, which are present inmany of the floras listed in Table2, and there was no pollen of the dominant rain forest generaof the present Guatemala rain forest [see Modernvegetation—3) Rain forest of the Atlantic region]. This issimilar to results from the middle Pliocene Paraje Solo flora ofsoutheastern Veracruz, Mexico, where although pollen of mangroves wasabundant, pollen of lowland tropical rain forest plants was absent. Pollen of Picea was also present in the Paraje Solo flora, asin the Padre Miguel assemblage, even though it is absent from the modernvegetation of both regions.

There is little or no evidence for paleocommunities of dry to aridaspect in the vicinity of the localities. Genera presentlycharacterizing the limestone plain of Petén, low savanna ofIzabal and Petén, arid desert plains (Rio Motagua Valley), andthe distant mixed forest of the Pacific plains are virtually absent fromthe fossil floras. Only Iresine and Caesalpinia aresuggestive of dry habitats, and some species of the latter grow undermoist conditions.

The presence of Alfaroa-Oreomunnea,Juglans, and Ulmus pollen, especially at inlandlocalities B and 4 (Padre Miguel Group), suggests that a version of thewet montane forest occupied moist sites at midelevations in Guatemaladuring the late Tertiary. The related upland mixed broad-leaved forestprobably grew in colder environments at slightly higher elevations, andsome pollen of oak may have come from that community. Other species ofoak, mixed with Pinus, formed the pine-oak forest that is sowidespread at present. There is no evidence for an extension of themixed montane forest of the Pacific bocacosta into southeasternGuatemala in the Mio-Pliocene.

Evidence for a cool coniferous forest is equivocal. Although pollenof Abies, Cupressus, Juniperus, andTaxodium was not recovered, some pollen of Pinus mayhave come from that community. The strongest indication that someversion of the coniferous forest may have been present is the fossilpollen of Picea. It occurs in the Pliocene of Veracruz Mexico(Paraje Solo flora) where it was associated with Abies. Sprucepollen has been recovered from Quaternary sediments in central Texas(Graham and Heimsch, 1960), from lateTertiary sediments from southeastern Mexico (Graham, 1976a), and from Guatemala (thisreport). It no longer occurs in any of these areas. Apparently spruceis quite sensitive to changes in climate (compared, e.g., toAbies), and migrated rapidly in response to the cool-moist andwarmer-drier cycles of the late Tertiary and Quaternary in the high tomiddle latitudes. No pollen from the distant alpine community isrepresented in the Guatemala material.

In summary, the paleocommunities of southeastern Guatemala in thevicinity of the collection localities during the late Tertiary includedversions of (1) freshwater swamps, marshes, and estuaries (ferns,Cyperaceae, Gramineae, Palmae), periodically inundated by brackishwater, (2) local mangrove swamps (cf. Acrostichum,Rhizophora), (3) wet montane forest (cf. Anthophytum,Palmae, Alfaroa-Oreomunnea, Juglans,Ulmus), (4) upland mixed broad-leaved forest(Quercus), (5) pine-oak forest (Pinus,Quercus), and (6) coniferous forest (Picea,Pinus). It is likely that arid habitats developed and becamemore extensive during glacial–integlacial transitions, asdemonstrated by Leyden (1984, 1985) for Guatemala and Lake Valencia,Venezuela.

Paleoclimates
The two palynofloras from Guatemala represent virtually the entirepaleobotanical database for the Tertiary of northern Central America. This is insufficient for precise quantitative assessment of thepaleoclimate, but when considered within the context of otherinformation from independent lines of inquiry, some trends are evident. The late Miocene and Mio-Pliocene was a time of global cooling afterearly Miocene warmth. This is the interval during which the Arcticglaciations began and the Greenland ice sheet developed. It alsoencompasses the time when northern temperate elements first appear innorthern Latin America. These elements are rare (Picea,Pinus?) in the early Miocene Simojovel flora (La QuintaFormation) of Chiapas, Mexico, but many are represented in the middlePliocene Paraje Solo flora of Veracruz, Mexico. Quercus firstreached present-day central Panama in the late Miocene-Pliocene (Gatunflora).

In the Guatemalan material, cool-temperate components(Picea, Quercus, Alfaroa-Oreomunnea,Juglans, Ulmus) occur only in the lateMiocene-Pliocene Padre Miguel flora or are best represented in thatassemblage (Table 2). There is no geologic evidence that elevations in southeastern Guatemalain the Mio-Pliocene were substantially higher than at present, sodifferences between the modern and paleovegetation can likely beattributed to climate. At Puerto Barrios (elevation ~2 m) the MATis 26.9°C with an annual range of 23.0°C. The average annualrainfall is 2888 mm, but may reach 6000 mm in the vicinity, and rangesmonthly from 76 to 475 mm (Committee for theWorld Atlas of Agriculture, 1969). A general estimate for theMio-Pliocene is a MAT 2°–3°C cooler than at present. Bycontrast, the Pliocene Herrería flora lacks most of thesecool-temperate elements, suggesting warmer conditions that may havecorresponded to the interval of mid-Pliocene warmth and higher sealevels. A general estimate is for a MAT during that interval 3.5°Cwarmer than at present.

Although these two fossil floras from southeastern Guatemala provideonly an approximation of the vegetation and paleoenvironments, they dolend support to the developing consensus that the low to middlelatitudes were not immune to the Late Tertiary and Quaternary climaticchanges that affected the northern biotas. This dynamic nature of thetropical rain forest was first documented by study of the Paraje Soloflora where, in an area presently characterized by tropical rain forest,that community was absent or poorly represented in the past (Graham, 1976a, b). As information from fossil floras andfaunas, isotopes (Guilderson, Fairbanks, andRubenstone, 1994), snow-line depressions (Rind and Peteet, 1985), noble gases (Stute et al., 1995), and other sourcesaccumulates, it is becoming clear that tropical communities areephemeral, delicately balanced, assemblages that have undergoneconsiderable alteration in range and composition throughout theCenozoic.


   FOOTNOTES
 
1 The author thanks Gabriel Dengo for providing information on collection sites and valuable suggestions for doing field work in Guatemala. The samples were collected in 1989 with Burke Burkart and Shirley A. Graham. The manuscript was read by Burke Burkart, Thomas Donnelly, and Shirley A. Graham, and Peter Hoch and Gwilym Lewis shared unpublished information on Ludwigia and Caesalpinia . Some samples were processed courtesy Gary Barker at Amoco, William Elsik at Exxon, and Kenneth Piel at Unocal. This project was initiated in 1980 under NSF grant DEB-8007312, but conditions in Guatemala at that time made it imprudent to attempt field work. Instead, related studies were carried out in Panama and at other localities in the Caribbean region. Near the end of the Panama study difficulties developed there, but had eased in Guatemala; hence the extended interval between the start (1980) and the completion (1997) of the Guatemala project. The study was further supported by NSF grant DEB-9206743. Back


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 LOCALITIES
 AGE AND STRATIGRAPHY
 COMPOSITION
 DISCUSSION
 REFERENCES
 
Anderson, T. H., R. J. Erdlac, and M. A.Sandstrom. 1985 Late Cretaceous allochthons andpost-Cretaceous strike-slip displacement along theCuilco-Chixoy-Polochic fault, Guatemala. Tectonics 5:453–475.

Bartlett, A. S., and E. S.Barghoorn. 1973 Phytogeographic history of the Isthmusof Panama during the past 12,000 years (a history of vegetation,climate, and sea-level change). In A. Graham [ed.],Vegetation and vegetational history of northern Latin America,203–299. Elsevier, Amsterdam.

Berendsohn, W. 1991 Thearboreal vegetation of the Laderas de La Laguna, a neotropical forestfragment in El Salvador, C.A. Dissertationes Botanicae 165:1–190.

Berry, E. W. 1939 A Mioceneflora from the Gorge of the Yumuri River, Matanzas, Cuba. JohnsHopkins University Studies in Geology 13: 95–136.

Brown, R. W. 1946 Walnutsfrom the late Tertiary of Ecuador. American Journal of Science244: 554–556. [ISI]

Burkart, B. 1978 Offsetacross the Polochic fault of Guatemala and Chiapas, Mexico. Geology 6: 328–332. [Abstract]

———. 1983 NeogeneNorth American-Caribbean plate boundary across northern Central America:offset along the Popochic fault. Tectonophysics 99:251–270.[CrossRef][ISI]

———. 1994 NorthernCentral America. In S. K. Donovan and T. A. Jackson[eds.], Caribbean geology, an introduction, 265–284. University of the West Indies Publishers' Association, Kingston,Jamaica.

———, R. E. Clemons, andD. C. Crane. 1973 Mesozoic and Cenozoic stratigraphy ofsoutheastern Guatemala. American Association of PetroleumGeologists Bulletin 57: 63–73.

———, and G.Moreno. 1983 North American-Caribbean plate boundary insouthern Chiapas, Mexico: further evidence of Neogene activity. Programa 10th Conferencia Geológica del Caribe(Cartagena) 28 (Abstract).

———, and B.Self. 1985 Extension and rotation of crustal blocks innorthern Central America and its effect upon the volcanic arc. Geology 13: 22–26. [Abstract]

Burke, K., C. Cooper, J. F. Dewey, P. Mann,and J. L. Pindell. 1984 Caribbean tectonics andrelative plate motions. Geological Society of America Memoir162: 31–63.

Burnham, R. J., and A. Graham. Inpress. The history of neotropical vegetation: new developments andstatus. Annals of the Missouri Botanical Garden.

Bush, M. B., D. R. Piperno, P. A. Colinvaux,P. E. De Oliveira,L. A. Krissek, M. C. Miller, and W. E.Rowe. 1992 A 14,300-yr paleoecological profile of alowland tropical lake in Panama. Ecological Monographs 62:251–275.[CrossRef][ISI]

Coates, A. G., J. B. C. Jackson, L. S.Collins, T. M. Cronin, H. J. Dowsett, L. M. Bybell, P. Jung, and J. A.Obando. 1992 Closure of the Isthmus of Panama: thenear-shore marine record of Costa Rica and western Panama. Geological Society of America Bulletin 104:814–828.[Abstract/Free Full Text]

Committee for the World Atlas ofAgriculture. 1969 World Atlas of Agriculture. Instituto Geografico de Agostini, Novara, Italy.

Cowgill, U. M., and G. E.Hutchinson. 1966 La Aguda de la Santa Ana Vieja: thehistory of a pond in Guatemala. Archiv für Hydrobiologie62: 335–372.

———, ———,A. A. Racek, C. E. Goulden, R. Patrick, and M.Tsukada. 1966 The history of Laguna de Petenxil, asmall lake in northern Guatemala. Memoirs of the Connecticut Academy of Arts and Sciences 17:1–126.

Cronin, T. M. 1991 Plioceneshallow water paleoceanography of the North Atlantic Ocean based onmarine ostracodes. In T. M. Cronin and H. J. Dowsett[eds.], Pliocene climates. Quaternary Science Reviews10: 175–188.

———, and H. J. Dowsett[eds.]. 1991 Pliocene climates.Quaternary Science Reviews 10: 1–296. [CrossRef][ISI]

Crane, D. C. 1965 Geology ofthe Jocotán and Timushán quadrangles, southeasternGuatemala. Ph.D. dissertation, Rice University, Houston, TX.

Davidse, G., M. Sousa S., and A. O. Chater[eds.]. 1994 [et seq.]. FloraMesoamericana. Universidad Nacional Autonoma de Mexico, Mexico, D.F.

Deaton, B.C. 1982 Relationship of the Colotenango Conglomerate ofGuatemala to the motion of the Polochic Fault during the Tertiary. Master's thesis, University of Texas, Arlington, TX.

———, and B.Burkart. 1984 Time of sinistral slip along the Polochicfault of Guatemala. Tectonophysics 102:297–313.[CrossRef][ISI]

Deevey, E. S. 1978 Holoceneforests and Maya disturbance near Quexil Lake, Petén, Guatemala. Polish Archives of Hydrobiology 25:117–129.

———, H. H. Vaughan, and G.B. Deevey. 1977 Lakes Yaxha and Sacnab, Petén,Guatemala. In H. L. Golterman [ed.], Interactionsbetween sediments and fresh water, 189–196. Junk, TheHague.

Dilcher, D.L. 1969 Podocarpus from the Eocene of NorthAmerica. Science 164: 299–301. [Abstract/Free Full Text]

Donnelly, T. W., G. S. Horne, R. C. Finch,and E. Lópen-Ramos. 1990 Northern CentralAmerica; the Maya and Chortis blocks. In G. Dengoand J. E. Case [eds.], The Caribbean region, 37–76. Geological Society of America, Boulder, CO.

Dowsett, H. J., and T. M.Cronin. 1990 High eustatic sea level during the middlePliocene: evidence from the southeastern U.S. Atlantic Coastal Plain. Geology 18: 435–438. [Abstract/Free Full Text]

Espinosa, A. F.[ed.]. 1976 The Guatemalan earthquake ofFebruary 4, 1976, a preliminary report. United States Geological Survey Professional Paper 1002:1–90.

Germeraad, J. H., C. A. Hopping, and J.Muller. 1968 Palynology of Tertiary sediments fromtropical areas. Review of Palaeobotany and Palynology 6:189–348.

Graham, A. 1971 The role ofmyxomyceta spores in palynology (with a brief note on the morphology ofcertain algal zygospores). Review of Palaeobotany andPalynology 11: 89–99.

———. 1976a Studiesin neotropical paleobotany, II. The Miocene communities of Veracruz,Mexico. Annals of the Missouri Botanical Garden 63:787–842.[CrossRef][ISI]

———. 1976b LateCenozoic evolution of tropical lowland vegetation in Veracruz, Mexico. Evolution 29: 723–735.

———. 1979 Mortoniodendron(Tiliaceae) and Sphaeropteris/Trichipteris (Cyatheaceae) inCenozoic deposits of the Gulf-Caribbean region. Annals of theMissouri Botanical Garden 66: 572–576. [CrossRef][ISI]

———. 1985 Studies inneotropical paleobotany. IV. The Eocene communities of Panama. Annals of the Missouri Botanical Garden 72:504–534.

———. 1987 Miocenecommunities and paleoenvironments of southern Costa Rica. AmericanJournal of Botany 74: 1501–1518. [CrossRef][ISI]

———. 1988a Studiesin neotropical paleobotany. V. The lower Miocene communities ofPanama—the Culebra Formation. Annals of the Missouri BotanicalGarden 75: 1440–1466. [CrossRef][ISI]

———. 1988b Studiesin neotropical paleobotany. VI. The lower Miocene communities ofPanama—the Cucaracha Formation. Annals of the MissouriBotanical Garden 75: 1467–1479. [CrossRef]