Sedimentary Geology 191 (2006) 227 – 254 www.elsevier.com/locate/sedgeo

Sedimentology and paleoecology of an Eocene–Oligocene alluvial–lacustrine arid system, Southern Mexico Hugo Beraldi-Campesi a,⁎, Sergio R.S. Cevallos-Ferriz a,1 , Elena Centeno-García a,2 , Concepción Arenas-Abad b,3 , Luis Pedro Fernández c,4 a

Institute of Geology, UNAM, Ciudad Universitaria, Coyoacán, 04510, DF México Area of Stratigraphy, Department of Earth Sciences, University of Zaragoza, E-50009 Zaragoza, Spain Area of Stratigraphy, Department of Geology, University of Oviedo, C/ Jesús Arias de Velasco s/n, E-33005 Oviedo, Spain b

c

Received 18 May 2005; received in revised form 24 January 2006; accepted 23 March 2006

Abstract A depositional model of the Eocene–Oligocene Coatzingo Formation in Tepexi de Rodríguez (Puebla, Mexico) is proposed, based on facies analysis of one of the best-preserved sections, the Axamilpa Section. The sedimentary evolution is interpreted as the retrogradation of an alluvial system, followed by the progressive expansion of an alkaline lake system, with deltaic, palustrine, and evaporitic environments. The analysis suggests a change towards more arid conditions with time. Fossils from this region, such as fossil tracks of artiodactyls, aquatic birds and cat-like mammals, suggest that these animals traversed the area, ostracods populated the lake waters, and plants grew on incipient soils and riparian environments many times throughout the history of the basin. The inferred habitat for some fossil plants coincides with the sedimentological interpretation of an arid to semiarid climate for that epoch. This combined sedimentological–paleontological study of the Axamilpa Section provides an environmental context in which fossils can be placed and brings into attention important biotic episodes, like bird and camelid migrations or the origin of endemic but extinct plants in this area. © 2006 Elsevier B.V. All rights reserved. Keywords: Tepexi–Coatzingo; Eocene–Oligocene; Sedimentary evolution; Fossil tracks; Fossil plants; Riparian ambients

1. Introduction ⁎ Corresponding author. Present address: School of Life Sciences, LSE 418, Arizona State University, Tempe, AZ 85287-4601, USA. Tel.: +1 480 727 7762; fax: +1 480 965 7599. E-mail addresses: [email protected] (H. Beraldi-Campesi), [email protected] (S.R.S. Cevallos-Ferriz), [email protected] (E. Centeno-García), [email protected] (C. Arenas-Abad), [email protected] (L.P. Fernández). 1 Tel.: +52 55 5622 4312. 2 Tel.: +52 55 5622 4310. 3 Tel.: +34 976 762 129. 4 Fax: +34 98 510 3103. 0037-0738/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.sedgeo.2006.03.018

Extensive magmatism and tectonic activity have contributed to the complex geological history of central Mexico during the Cenozoic. Paleogene inland basins there are poorly known, mainly because of the extensive cover by younger rocks and imprecise correlations (e.g. Ferrari et al., 1999; Morán-Zenteno et al., 1999). Paleontological studies have provided references for dating rocks and have contributed to a more comprehensive understanding of the biotic

228

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254

communities, but yet such studies do not always provide an environmental context for fossils and they rarely involve sedimentological studies in their paleoenvironmental reconstructions. Therefore, Paleogene basins with well-preserved outcrops and fossils

can be extremely valuable components in facilitating reconstruction studies. The Eocene–Oligocene Coatzingo Formation, located in the Central–South area of Puebla, Mexico (Fig. 1A), is known for its plant and animal fossili-

Fig. 1. (A) Map showing the study area and fossil localities from the Tepexi–Coatzingo basin. (B) Local geology (solid line-box in map A) of the Tepexi de Rodriguez area. Pz = Paleozoic; Mz = Mesozoic; K = Cretaceous; E = Eocene; E–O = Eocene–Oligocene; P = Pliocene–Pleistocene; Q = Quaternary.

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254

ferous localities (e.g. Buitrón and Malpica-Cruz, 1987), particularly artiodactyl (camel-like and cervid) fossil tracks, together with those of cat-like mammals, proboscideans, birds, reptiles and the skeleton of a flamingo (Cabral-Perdomo, 1996). Palynomorph assemblages described from several localities in the Formation (Carranza-Sierra, 2001; Carranza-Sierra and Martínez-Hernández, 2002; Martínez-Hernández and Ramírez-Arriaga, 1999) have served as tools for dating, biostratigraphical correlations and floral descriptions. An ever-growing collection of fossil plants from the Ahuehuetes locality, one of the most wellpreserved Oligocene records in Mexico, has been used to estimate ages, paleotemperatures and to improve the understanding of biogeographical patterns of North Americas' flora (Calvillo-Canadell and Cevallos-Ferriz, 2002; Magallón-Puebla and Cevallos-Ferriz, 1994a; Velasco de León and Cevallos-Ferriz, 2000; Ramírez-Garduño and Cevallos-Ferriz, 2002). Other fossils from the Formation include fossil wood (currently being studied), stromatolites, and ostracods. Outcrops of the Coatzingo Fm in the Pie de Vaca and nearby localities have been interpreted as lacustrine in origin (Buitrón and Malpica-Cruz, 1987; Pantoja-Alor et al., 1988) based on field observations, but no sedimentological studies that describe the type of paleolake, its evolution or the overall paleoenvironmental conditions that existed have previously been undertaken. In this paper, we provide a paleoenvironmental reconstruction of part of the Coatzingo Formation, based on facies analysis of one of its most complete sections, termed here the “Axamilpa Section” (AS). An overall integration of the sedimentological and paleontological data from the region is taken into consideration to further discuss relevant biotic events, such as interchanges and endemisms, that may have occurred in this area, some of them with profound evolutionary implications. 2. Local geology The geology of the Tepexi de Rodriguez area (Fig. 1B) has not been studied completely. In general, a Cenozoic succession is underlain by Cretaceous rocks, e.g. the Tlayua Fm (Kashiyama et al., 2003; PantojaAlor, 1990) and the Rosario Fm, and by a Paleozoic basement, the metamorphic Tecomate Formation, which represents the northernmost part of the Acatlán Complex (Ortega-Gutierréz et al., 1999). The lower rocks of the Cenozoic succession correspond to a white, calcareous, clast-supported conglomerate that

229

crops out in many areas of the Tepexi–Coatzingo basin, which has been informally linked with the Eocene Balsas Conglomerate (Martínez-Hernández and Ramírez-Arriaga, 1999; Pantoja-Alor et al., 1988). It crops out West of the AS (although their contact is not clearly visible), and unconformably underlies the eastern part of the Ahuehuetes section, being then at least part of the local basement of the Coatzingo Fm. Finally, the Cenozoic succession is covered with the Plio-Pleistocene Agua de Luna Formation (Pantoja-Alor et al., 1988) in the SE part of the Tepexi area (Fig. 2). 2.1. The Coatzingo Formation The Coatzingo Formation, previously named Pie de Vaca Fm (Pantoja-Alor et al., 1988), correspond to Paleogene rocks within the Tepexi–Coatzingo basin. That Formation has been locally divided into two units: the lower Pie de Vaca Unit, composed mainly of limestones and cherty and sandy limestones, and the upper Ahuehuetes Unit, composed mainly of tuff and tuffaceous sandstones (Silva-Romo and GonzálezTorres, in: Calvillo-Canadell and Cevallos-Ferriz, 2002; Fig. 2), both interpreted as fluvial to lacustrine low-energy environments. The geographical limit of the Coatzingo Fm is at present uncertain; however, several localities (e.g. Chigmecatitlán, Zaragoza, and Cuayuca) have been correlated biostratigraphically (Martínez-Hernández and Ramírez-Arriaga, 1999), extending its area further to the west and south of the study site (Fig. 1A). 2.3. The Axamilpa Section The Axamilpa Section (AS) is located along the Axamilpa River, close to the Ahuehuetes locality, 3 km N of the town of Tepexi de Rodríguez, in the central– southern part of Puebla (97°55′48ʺW, 18°36′42ʺN; 1680 masl). This section, which represents the Pie de Vaca Unit (Fig. 2), is thought to underlie the Ahuehuetes Unit, although they are geographically separated. Initially, the Pie de Vaca Unit was interpreted to be of Miocene to Pliocene–Pleistocene age (Buitrón and Malpica-Cruz, 1987; Cabral-Perdomo, 1995; Rodríguez de la Rosa et al., 2005), based upon fossil mammal tracks from the Pie de Vaca locality. Later on, extensive palynologic and paleobotanic studies have pointed to an Eocene–Oligocene age for both the Pie de Vaca and Ahuehuetes Units (Calvillo-Canadell and CevallosFerriz, 2002; Carranza-Sierra and Martínez-Hernández,

230

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254

Fig. 2. General stratigraphy of the regional lithologies within the Tepexi–Coatzingo basin and the position of the Coatzingo Formation Units. The localities Axamilpa Section, Ahuehuetes, and Pie de Vaca (right) are shown in a stratigraphical context.

2003; Martínez-Hernández and Ramírez-Arriaga, 1999), interpretation also supported by regional geological studies (Silva-Romo, 1998 in Calvillo-Canadell and Cevallos-Ferriz, 2002). 3. Stratigraphy and facies of the Axamilpa Section Well-exposed horizontal strata of the AS crop out mainly in a vertical cliff (Fig. 3A) that shows no evidence of major deformation. Three stratigraphic sections were measured at sites A, B, and C (Fig. 3B), then correlated and combined in a single log (Fig. 4, Log 1). A second section (Log 2) was measured at site D. Three lithological groups are recognized in the AS: a) detrital rocks (conglomerates, coarse sandstones and minor limestone, from the base of the section to ∼ 20 m), b) detrital and chemical rocks (sandstones, minor limestones and marls, from ∼ 20 to ∼ 36 m), which commonly appear interbedded, and c) rocks mainly of chemical origin (limestones and evaporites, marls and scarce terrigenous, from ∼ 36 m to the top). Log 2 contains only groups b) and c). Facies descriptions and interpretations are summarized in Table 1. The vertical distribution of facies and their associations is shown in Fig. 5. Microscopic features of the facies are shown in Fig. 6.

Conglomeratic facies Matrix-supported conglomerates (Cgm, Fig. 7A) This facies comprises polymictic, matrix-supported conglomerates, which form irregular beds up to 2 m thick. The matrix is a sand–silt mixture with a high percentage of carbonate. Clasts are angular to subrounded, 3 to 7 cm in diameter, and composed of schists, white or reddish limestone, claystone and quartz. Some parts are replaced and cemented with hematite. Interpretation: These conglomerates represent cohesive debris flow deposits. Matrix strength and buoyancy prevented large clasts to sink, keeping them floating and favoring a disorganized fabric. These conglomerates always form on slopes, and are typical of the proximal or mid parts of alluvial fans (e.g. Colombo, 1992; Reading, 1986). Clast-supported conglomerates (Cgc, Fig. 7B–C) These consist of poorly sorted clast-supported conglomerates. Clasts are angular to subrounded, up to 14 cm long, consisting mainly of schist, although some quartz and limestone clasts are also present. Pore spaces between clasts are commonly filled by calcite cements. This facies forms tabular, dm- to m-thick, strata with erosive bases. Beds display planar and trough cross-stratification, with

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254 Fig. 3. (A) Panoramic view, facing NE, of the Axamilpa Section and surroundings. (B) Topographic map of the Axamilpa Section showing the sites where logs were measured. Log 1 was measured in sites A, B, and C; Log 2 was measured in site D (logs shown in Fig. 5). 231

232

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254

normal or inverse grading. Some beds pass laterally into coarse sandstone. Clasts are regularly imbricated, showing N40° paleocurrent directions. Interpretation: This kind of deposits would form in braided channels and bars in middle parts of alluvial systems (Miall, 1996). Some coarsening-upward examples of this facies may record progradation of minor deltaic-type bars, given their close relationship to calcareous lacustrine facies, or merely reflect migration of the coarser head of longitudinal bars on the finer-grained downcurrent tail. Sandstone facies Cross-stratified sandstones (Sc, Fig. 7D) These consist of very coarse to fine-grained sandstones, arranged in dm to m thick tabular or lenticular strata, which display normal or inverse grading and cross stratification. These beds can contain conglomeratic lenses, and may pass laterally into fine-grained conglomerates. Fluid escape structures are locally present. Interpretation: This facies is common in fluvial systems of medium to low energy (Miall, 1996), where gravel deposits may accumulate at the base of the channels (Davis, 1983). It is interpreted to have formed from channeled and unconfined flows, in the middle-to-distal parts of alluvial fans. Fluid escape structures may be evidence of contemporary seismic movements or other factors that led to sediment disturbance. Horizontal-laminated and cross-stratified sandstones (Shc, Fig. 7E) This facies is formed by fine to medium (rarely coarse) sandstones arranged in cm- to dm-thick beds, which are amalgamated or, more rarely, separated by pale-green marl layers, containing clay, quartz and limestone grains. Sandstone beds, which may display load structures and marl intraclasts in their base, show internal horizontal lamination, sometimes with cross and flaser stratification and ripple lamination. Sandstones contain large amounts of schist fragments and also scattered angular quartz and other rock fragments (9–30 mm in diameter). Calcareous cement and hematized zones are common in this facies. The marls are porous and poorly indurated, and may contain ‘algal lumps’, peloids, ostracod remains, oolites, bioturbation galleries. Brecciation and nodulation, as well as gypsum crystals and chalcedony, may also be present. The abundance of oolites and ostracods in this facies varies along the section.

Interpretation: This facies may represent deposits laid down on the floodplains of a distal alluvial-fan setting or even in the delta plain of a lacustrine delta system (cf. Miall, 1992, 1996; Rust, 1980). In such settings, unconfined floods can deposit sand sheets with a variety of structures (e.g. parallel, flaser, and ripple lamination) that reflect variable energy conditions. The common normal grading in the sandstones reflects the waning behavior of these flows, although turbulent episodes capable of eroding the substratum could have occurred, as implied by the many marly rip-up clasts. During quiet periods or the weakest flows, alternated clay and carbonate accumulation would account for the formation of marls among the clay-rich layers. The presence of oolites, which are common in large saline and calcareous lakes fringes (e.g. Kowalewska and Cohen, 1998; Swirydezuk et al., 1979), indicates currents or oscillation produced by wind and swell during periods of absence of terrigenous. The bioturbation suggests the presence of interstitial organisms and, therefore, the existence of a water layer. During drought periods, the substrate would be exposed to desiccation and oxidation, resulting in brecciation and the precipitation of hematite. Sandstone and marl alternations (Scs, Fig. 7F–G) This facies is composed of sandstone and marl alternations, cm- to dm-thick beds, which typically form overall coarsening-upward units. The separation of both lithologies is not abrupt but continuous. Marl layer thickness decrease toward the top, while thickness of sandstones increase. Loading structures are observable at the base of the beds, and ripple- and low-angle stratification are present in the upper sandstones. Interpretation: The coarsening upward trend and the thinner beds towards the top are common indicators of deltaic progradation (e.g. Bhattacharya and Giosan, 2003). Given the thin nature of this facies and its stratigraphical position between rocks of chemical origin, it is interpreted as the progradation of small deltaic lobes in distal flat areas. Massive or graded sandstones (Sm, Fig. 7H) This facies comprises mostly fine, but also medium grain-size sandstones, mainly formed of schist, quartz and carbonate grains extensively cemented with calcite and locally with hematite. Sorting is

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254

Fig. 4. Stratigraphic logs from the Axamilpa Section.

pp. 233–236

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254

237

Table 1 General description and interpretation of the Axamilpa Section facies (see text for details) Facies

Description

Interpretation

Conglomeratic Cgm

Polymictic, matrix-supported conglomerates.

Debris-flow deposits in proximal or middle parts of alluvial fans. Braided channel and bar deposition in middle to distal parts of alluvial systems. Channelled and unconfined flows in the middle to distal parts of alluvial fans. Deposits on floodplains of distal alluvial-fan or lacustrine delta systems.

Cgc Sandstone

Sc Shc

Scs

Sm

Marl

Mm

Mh

Calcareous

Lm

Lo

Ll Ls Evaporitic

Gy

Clast-supported conglomerates with erosive bases. Planar cross-stratification, normal or inverse grading. Coarse- to fine-grained sandstones in tabular or lenticular strata. Cross-stratification and conglomeratic lenses. Fine- to medium-grained sandstones with horizontal and cross-stratification, and cm-thick interbedded marl layers. Sandstone and marl alternations. Net coarsening-upward intervals. Load structures, ripple and low-angle cross stratification. Fine to medium massive sandstone; inverse or normally graded, internal horizontal lamination and rare fluid-escape structures. Marl layers, massive or with horizontal lamination. Chert bands and gypsum nodules, brecciation and bioturbation are present. Alternation of calcareous marls and clay-rich layers. Load structures, fluid escape structures, desiccation cracks, ripples, peloids and bioturbation galleries are found. Fossil plants found at one level. Limestones in massive or stratified layers containing desiccation cracks, load structures, vertebrate tracks, rhizocretions, pores, microgranular and nodular gypsum, chert bands and nodules, chalcedony, brecciation, bioturbation, microbial-like lamination, nodulation, ostracod remains, oolites, bacterial or algal microorganisms, and peloids. Massive oolitic limestones with gypsum nodules and chert bands, vertebrate ichnofossils, peloids, and ostracods. Coated grains and pisoids, bioturbation and algal lumps are present. Varve-like laminated limestones with flaser stratification, silicified oncolites, chert and gypsum nodules. Stromatolites capped with gypsum.

Progradation of small deltaic lobes in flat marginal lacustrine areas. Suspended deposition (flash floods) in flat alluvial or deltaic settings, near lake margins. Periodic or continuous settling of carbonate muds in offshore, low-energy lacustrine areas, with reduced terrigenous input. Subaerial exposure events. Palustrine setting with soil formation and carbonate deposition.

Carbonate deposition in shallow lacustrine to palustrine settings, where herbaceous vegetation became established.

Margins of saline carbonate lakes with agitated shallow waters.

Rhythmic carbonate deposition in relatively deep, low-energy offshore lacustrine areas. Marginal shallow and low-energy lacustrine environments, with prolonged evaporation and depletion of water. Gypsum and calcite laminae. Nodular gypsum. Halite Intense and prolonged lake evaporation events. also present.

good, although the basal part of some beds contains scattered grains up to 8 mm. Beds are massive or inversely or normally graded, although, in some cases, they show horizontal lamination and scarce fluid escape structures. Interpretation: This facies suggest direct sedimentation of sand from waning, turbulent sediment-laden flows with no (or little) traction at the bed, perhaps with an eolian contribution. The differences in grading would reflect sedimentation from high to low density fluids (e.g. Lowe, 1979). The presence of fluid escape structures also indicates liquefied flows that resulted from rapid flow deceleration

and quick deposition. These probably arose from flow expansion at the mouth of confined conduits (channel mouths?), in a wide and flattened alluvial or deltaic setting. The flows loaded with sand were perhaps generated by sudden floods during heavy rainstorms (e.g. Marzo, 1992; Mutti et al., 1996). Neighboring carbonate strata suggest deposition in a distal floodplain to marginal lacustrine setting. Marl facies Massive or laminated marls (Mm, Fig. 7I) These marls form tabular layers, cm- to dm-thick, and are either massive or have horizontal lamination. Chert bands and microcrystalline gypsum nodules up

238

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254

to 1 cm are present. Some marls show brecciation and bioturbation. Interpretation: These marls represent periodic or continuous settling of carbonate mud in offshore, low-energy lacustrine areas with little terrigenous input. Gypsum nodules may have formed during periods of evaporation. Marls and claystones (Mh, Fig. 7J–K) Alternation of cm- to dm-thick layers of calcareous marls with cm-thick greenish clay-rich vermiculite layers. Thin, mm-thick laminae of carbonatecemented sandstone may be present in the marl levels. Ripples may be present locally, as well as peloids and bioturbation. Fossil plants were found at one level. Load structures, fluid escape structures, and desiccation cracks are very common (Fig. 7K). The vermiculite forms layers 1–3 cm thick that rhythmically interrupt the marls. Interpretation: Vermiculite can be found in soils (McCarthy and Plint, 2003; Schroeder et al., 1997) and in paleosols (Davies and Gibling, 2003), indicating that soils may have formed in a palustrine setting where carbonate deposition occurred during rainy seasons with high lake levels. The poor preservation of the plant fragments implies an early diagenetic degradation. Limestone facies Massive limestones (Lm, Fig. 7L) This facies consists of white-yellowish, beige or green mudstones that form cm- to dm-thick tabular layers. Desiccation cracks, load structures, vertebrate fossil tracks, rhizocretions, and pores (1 to 3 mm in diameter) are present. Replaced and fragmentary ostracod valves, complete and broken oolites, and peloids occur but are uncommon. Fungal or algal filaments are also present. Some beds contain microgranular and microcrystalline nodular gypsum, radial calcite crystals, bands and nodules of chert, and chalcedony. Some horizons are brecciated, bioturbated, and show microbial-like lamination (with bacterial or algal remains preserved, although rarely). The surfaces of these rocks are commonly replaced by hematite. Interpretation: This facies formed from carbonate deposition in shallow lacustrine to palustrine areas that underwent periodic subaerial exposure. The presence of rhizocretions indicates that herbaceous vegetation covered parts of the surface, possibly grass-like plants adapted to basic conditions.

Oolitic limestones (Lo, Fig. 8A) Beige oolitic grainstones to packstones represent this facies. They are typically partially silicified, with large amounts of oolites, peloids, ostracods, coated grains and pisoids, as well as algal-like aggregates. In a few places, they show partial dolomitization, bioturbation and hematite replacements. Some are porous and contain microcrystalline gypsum nodules, mm to cm in diameter, and chert bands, mm to cm thick. Vertebrate fossil tracks were observed at one level. Interpretation: These deposits are typical of those that form along the edges of saline carbonate lakes (e.g. Kowalewska and Cohen, 1998; Swirydezuk et al., 1979). A shallow or fringing lacustrine environment with agitated water that promoted the formation of oolites is inferred for this facies. Laminated limestones (Ll, Fig. 8B–D) This facies is represented by cm- to dm-thick tabular strata of mudstones. Laterally they have mm-thick varve-like horizontal and wavy laminae and no bioturbation is observable. Locally they show flaser stratification and contain scarce silicified oncolites, chert and gypsum nodules, 0.5 mm to 10 cm in diameter. Porosity is conspicuous with pores up to 4 mm in diameter. In thin section, small anhedral gypsum crystals, microgranular chert, clays and quartz grains are seen. Interpretation: These deposits may have formed in relatively deep and low-energy lacustrine offshore areas by carbonate deposition and variable terrigenous inputs. The varve-like appearance suggests that the carbonate versus terrigenous sedimentation was influenced by seasonal factors. This may imply also changes in the lake level, and thus the lake area. The scarcity of oncolites suggests that they were formed in shallower zones and transported to deeper zones. Likewise, scattered terrigenous suggest eolian inputs during high level stages. Stromatolitic limestones (Ls, Fig. 8E–F) Stromatolites appear associated with facies Ll and Lm, in strata varying from 60 to 70 cm in thickness. Two types of stromatolites were found at two different levels: a) large, columnar and domal stromatolites, 30 to 40 cm high, which form abundant, continuous laterally coalescent bioherms; and b) small, hemispherical stromatolites, 10 to 12 cm high, which occur alone or in clusters of two or

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254

Fig. 5. Facies distribution in the Axamilpa Section. See text for explanation of facies associations.

pp. 239–242

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254

243

Fig. 6. Microscopic features in thin section. (A) Pellets within a spar matrix. (B) Nucleated oolites with chalcedony crystals (C). (C) Large gypsum crystal (G) within a micritic matrix. (D) Radial growing of diagenetic calcite. (E) Rhizoidal pores (P, black) and nodules (N, dark grey). (F) Rhizoid pore with inner concentric layers of calcite. (G) Brecciated zone with intraclasts (I). (H) Bioturbated microbial-like lamination. (I) Partially dolomitized limestone clast, with a miliolid foraminifer (arrow). (J) Echinoderm fragment (E) along with quartz grains (Q) and pelloidal limestone fragments (P).

244

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254

245

Fig. 8. (A) General view of the Facies Lm and Lo next to a rupestrian painting. (B) Facies Ll containing gypsum nodules (arrows). (C) Close-up of the facies Ll showing compacted mm lamina. (D) Sequence of the facies Mm, Lm, Lo, Lm, and Ll (∼25 m in log 1). (E) Large stromatolites (facies Ls). (F) Small isolated stromatolites (facies Ls). (G) Transversal cut of a small stromatolite showing the outside and its internal lamination. (H) Partially dissolved and collapsed stratum of gypsum (facies Gy) with calcite laminae. (I) Intricate growth of gypsum crystals.

three domes. The stromatolitic lamination shows the characteristic dark/light-lamina alternation (Reid et al., 2000; Seong-Joo et al., 2000), with the lighter laminae being thicker and dominant. Partial silicification is present toward the centre in both types. Both stromatolitic intervals are noteworthy capped by gypsum in honeycomb-like frameworks. The matrix

between the stromatolites is massive or laminated carbonate. Interpretation: These facies formed in low energy, shallow and marginal lake environments. Their continuous and smooth lamination, and the regular morphology of the structures suggest that relatively stable conditions

Fig. 7. (A) Close-up of the facies Cgm, showing poor sorted floating clasts. (B) Erosive contact between facies Cgc (above) and facies Cgm (below). (C) Facies association A in log 1 (∼ 12 m) showing two cycles (corks). Hammer –encircled – is scale. (D) Cross-stratification of the facies Sc. (E) Fine lamination of facies Shc. (F) Facies Scs lamination. (G) Load structures of the facies Scs. (H) Facies Sm. (I) Facies Mm (arrows) in Log 1 at ∼ 25 m (scale = 1 m). (J) Facies Mh showing marl (M, light) and claystone (C, dark). (K) Flame structures of the facies Mh where thin sandstone layers occur. (L) General view of Lm facies in site D (∼11 m; hammer – encircled – is scale).

246

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254

prevailed during their formation. The extensive gypsum on their upper surfaces suggest alkaline conditions with periods of prolonged evaporation. The limited presence of stromatolites in the AS suggests that favorable conditions for their growth were not frequent. Evaporitic facies Gypsum (Gy, Fig. 8G–H) This facies consists of dm-thick, tabular to very irregular beds of dissolved and collapsed gypsum. Thin and broken carbonate layers and halite casts are also present. Microgranular, microcrystalline and macrocrystalline textures are observable, although macrocrystalline gypsum is more common in irregular beds. Interpretation: This facies represents increasing solute concentration in the lake and sediment pore water due to lowering of the water table and intense evaporation. 3.1. Silicification processes Many of the facies above described contain chert nodules and bands and chalcedony. They are clearly diagenetic features and are found replacing some of the primary components of the deposits (e.g. oncolites, stromatolites, limestones). The source of silica, as in many silicified deposits elsewhere, can be debatable. Diatoms have not been detected and thus they can likely be ruled out as a major source of silica. Although microbial mats have also been claimed as a source of silica in lacustrine deposits (Bustillo et al., 2003), where silicification takes place in littoral and eulittoral sequences of lacustrine carbonates, the paucity of microbial deposits in the AS argues against microbial mats being a major source for silica. Thus, it is more likely that the extensive magmatic activity that prevailed in this region during the Paleogene (Martiny et al., 2000; Morán-Zenteno et al., 1999) would have accounted for the silica input, as it has been seen in other basins (e.g. Dunagan and Turner, 2004; Eugster, 1980; Pirajno and Grey, 2002). In some cases, hydrothermal pulses may have loaded the pores of the preexistent rocks with the silica that would later precipitate as chert. In other cases, silicification may have occurred during early diagenesis by the mixing of meteoric waters with the evaporite pore waters, causing dissolution and collapse of the evaporite layers, calcite cementation and silicification, probably as a complex and recurrent sequence of processes (Arenas et al., 1999).

3.2. Facies associations The facies described above are associated in vertical sequences that represent the superposition of different subenvironments through time. Facies that are inferred to be genetically linked have been grouped, and their vertical associations (Fig. 6) can be interpreted as a sequence of events that describe the sedimentary evolution of the section. A association: Cgc → Sc, Sm → Shc, Lo, Mm → Lm This fining upward sequence begins with erosive or planar contacts. It comprises mainly clastic facies with thicknesses from 30 to 250 cm, followed by massive limestones that are up to 35 cm thick. This association occurs cyclically and suggests gravel and sand deposition in braided bar and channel systems of alluvial fans that reached lacustrine areas, in which sheet-like flows deposited part of the detrital sediments as facies Shc. Finer sediment reached distal zones of the fan. Decreasing terrigenous inputs allowed deposition of massive limestone facies in shallow or even fringing lacustrine conditions. The sequence represents the retreat of an alluvial system and the establishment of lacustrine and palustrine environments. Features of the alluvial-fan conglomerates, such as the clast size (8–14 cm), the local debris-flow events, and their close relationship with lacustrine facies, indicate alluvial fans of small size, mostly with active fluvial-dominated sectors in relatively steep areas, with their apices close to the palustrine and lacustrine areas that existed down stream. Paleocurrents inferred from the conglomerate clast imbrication suggest that the source of alluvial sediments was located to the SW of the study area. B association: Scs → Shc, Sm → Sc →→ Cgc This comprises a clastic facies with thickness from 50 to 250 cm and coarsening upward evolution, beginning at the base with sandstone facies, either massive or interbedded with marls, overlain by coarser sandstones that grade into conglomerates towards the top. Marls are evidence of a water body with carbonate deposition. This association records the progradation of the alluvial system by channeled flows that entered a shallow lake area, giving rise to small deltaic systems. C association: (Shc), Sm, Sc → Mm → Lm, Lo, Ll This consists of clastic and carbonate facies, varying from 4 to 70 cm thick, with planar contacts and a fining-upward pattern. It begins with fine detrital inputs that induced a subsequent rise of the water table, and caused offshore marl deposition (facies Mm). Later, shallower lacustrine conditions gave place to

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254

carbonate deposition in marginal areas with or without agitation (facies Lo and Lm, respectively). Facies Ll could represent a slightly deeper lake setting with episodic carbonate deposition, and would mark the establishment of lacustrine, probably shallow, carbonate depositional conditions after an initial deepening event. This association reflects the end of the alluvial– deltaic dominance, followed by deepening–shallowing lacustrine events. D association: Mm, Mh → Ll, Lm → Lo, Ls → Gy This is a complex association formed of cm- to dmthick claystone, marl, calcareous and evaporitic facies. Commonly the strata are tabular, except for the gypsum, which is found in deformed, collapsed and partially dissolved strata. It begins with deposition in low-energy offshore areas as a consequence of a lake deepening and expansion. It continues with carbonate precipitation in either relatively deep (facies Ll) or shallower waters (facies Lo, Lm) in which palustrine conditions were present. Two of the most common sequences of this shallowing process are D1: Mm → Ll → Lm; and D2: Mh → Ll, Lm. Facies Lm and Ll can be found alternating through time, which indicates cyclic water level changes attributed to deepening–shallowing events, for example, D3: Mm, Mh → Lm → Ll → Lm (Fig. 6). Facies Lo and Ls would also imply shallow, but probably more saline, carbonate lacustrine conditions in marginal areas affected by waves (Lo) or in calm areas (Ls), followed by high rates of evaporation and desiccation that formed facies Gy in very shallow ponds, and also caused brecciation, nodulation, and gypsum nodules to form in earlier carbonate facies. Two of the associations that reflect that evolution are: Mm → Ll, Ls → Gy, and Mh → Lm → (Ll) → Lo → Gy. The sequence represents an overall shallowing that implies the change from lacustrine–palustrine to evaporitic conditions, due to supersaturation of the lake by increased evaporation that caused sulfate deposition. Rises in the lake water level due to freshwater inputs, may have caused dilution of the lake waters, a relative deepening and the beginning of a new sequence. 3.3. Fossil record of the Axamilpa Section Stratigraphic appearance of the Axamilpa Section fossils is indicated in Fig. 4. Representative specimens are shown in Fig. 9. Vertebrate fossil tracks: (Fig. 9A–B) All the tracks observed in the AS were found on top of exposed bedding planes at various levels of the section.

247

Although footprints are not always well marked, tracks of at least 4 steps can be easily distinguished. In some strata, the tracks are better defined than in others, suggesting different consistency and water saturation of the substrate at the time of the imprint. They are usually found displaying alternated right and left steps. Foot lengths vary from 10 to 18 cm. The form of the digits is typical of ungulates (artiodactyls). Fossil tracks with similar characteristics have been described from other localities and related to the Camelidae group (Cabral-Perdomo, 1995). Bioturbation ‘galleries’: (Fig. 9C) Galleries are observed in thin laminae, in several levels of the section. The traces are usually oriented in a same direction, although they may appear randomly arranged. Sometimes ostracods are found in the same sediments, and it is possible that they contributed to the bioturbation, given the benthic and excavator habits of some species (Henderson, 1990). However, there are many other possible burrowing candidates. Leaves: (Fig. 9D–I) At least six different types of leaves were recognized in the facies Mh. They are moderately well preserved carbonaceous imprints, both fragmental and complete specimens. Recognized genera include Pseudosmodingium and Pistacia from the Anacardiaceae and Cedrelospermum from the Ulmaceae. The remaining material has not been identified due to the absence of diagnostic characters, although some plant remains resemble legumes and aquatic plants in morphology. Oncolites: (Fig. 9J–K) Scarce ellipsoidal oncolites, 3 to 5 cm in diameter, were found in limestones (Facies Ll). Their concentric lamination may be obliterated by silicification, which is most intense toward the center, whereas the external surface remains unsilicified. Rhizocretions: (Fig. 9L) In some strata, relicts of roots (rhizocretions) were observed. These are small (up to 10 mm in length), oxidized (inferred by the color), and mostly appearing as vertical traces. Ostracods: (Fig. 9M–N) These are present in several horizons of the succession, always in low-energy lacustrine marls and limestones. The shape and size of the valves, between 400 and 550 μm in length, is constant along the stratigraphic section, suggesting stability in diversity of species. They do not display ornamentation, being quite smooth, which is common in non-marine ostracods (Henderson, 1990), as assumed for those of the AS. Ostracods are present either as conjoined or disarticulated valves, complete or fragmented. They are abundant regionally. Coquinas with silicified ostracods have been observed at the Zaragoza locality (Fig. 1) and some 3 km E from

248

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254

249

Fig. 10. Sedimentary evolution proposed for the Axamilpa Section (see text for explanation). (A) Alluvial–fluvial stage. (B) Transitional stage. (C) Lacustrine stage. (D) Evaporitic stage.

it, suggesting that one or more lakes existed at that time. In that case, the similarity between faunas may indicate a relationship between the water bodies. Microorganisms: (Fig. 9O–P) Permineralized microfossils were observed in thin sections of chert samples. They appear in groups of four cells, wrapped by a mucilage-like sheath. The diameter of each cell varies from 5 to 7 μm. The diameter of the groups varies from 10 to 15 μm and they are arranged in small colonies (6 to 20 individuals per colony). Stromatolites: (see Stromatolitic limestones, Facies Ls). 4. Evolution of the sedimentary system The Axamilpa Section clearly shows a fining-upward trend from alluvial conglomerates and sandstones to lacustrine carbonates and evaporites. This evolution reflects an overall retrogradation of an alluvial–fluvial system, followed by expansion of a carbonate deposi-

tional lacustrine system that eventually evolved to a more shallower and alkaline, intermittent lacustrine settings. In the studied succession this can be conceived as four main depositional stages: 1) alluvial–fluvial (Fig. 10A), characterized mainly by detrital facies (conglomerates, sandstones and mudstones, 0–20 m); 2) transitional (Fig. 10B), characterized by marls, sandstones and limestones (alternating distal alluvial and carbonate lacustrine facies, 20–36 m); 3) lacustrine (Fig. 10C), characterized by carbonates; and finally 4) evaporitic (Fig. 10D, 36–55 m). The alluvial–fluvial stage, characterized by braided channel and bar deposits with rare debris flows, experienced a general retrogradation through time (associations A and B), that gave rise to the expansion of palustrine and lake environments northeastward, overlapping the alluvial domain, as indicated by limestones and marls that cap the retrograde cycles. The transitional stage records the influence of both the alluvial and the lacustrine depositional environments,

Fig. 9. (A) Track of vertebrate ichnites. (B) Shape of an ichnite delimited by a draw. (C) Parallel bioturbation galleries. (D) Fossil leaves of Cedrelospermum sp. (E) Carbonaceous impression of an unidentified leaflet. (F) Poorly preserved fossil leaf of Pistacia sp. (G) Fossil leaf of Pseudosmodingium sp. (H) Fossil impression of a legume. (I) Fossil leaf of a possible aquatic plant. (J) Oncolites in the field. (K) Transversal cut of a partially silicified oncolite (white calcite coat). (L) Rhizoid impressions (scale = 1 cm). (M) Transversal view of an ostracod showing substituted internal remains. (N) Transversal cut of an ostracod with non-symmetric valves. (O) Permineralized microorganisms. (P) Individual microbial cells.

250

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254

with alluvial inputs in the form of lacustrine deltaic lobes and distal-fan sheet-like deposits (associations B and C). Although a timeframe for these events is difficult to assess, the frequency of flow-generated deposits appears to decrease starting from the transitional stage, favoring the development of palustrine conditions around a lacustrine water body, as inferred from desiccation cracks, rhizocretions, nodulation, brecciation, and fenestral porosity present in facies Lm. Finally, with the cessation of the alluvial influence, carbonate lacustrine environments became extensive, and facies Lm, Ll, Lo, Ls, Mm, and Mh developed (sequence D). The common, but discontinuous, appearance of ostracods along the Axamilpa Section supports the existence of intermittent water bodies, whose intermittency appears to be more frequent as the environment became dryer. This would imply a progressive lowering of the lake level, giving rise to ephemeral and shallow saline lakes. While subjected to intense evaporation, sulfates (e.g. gypsum) would have concentrated in the lake and formed early diagenetic features (e.g., gypsum nodules) in the exposed mud flats. During early diagenesis, the influence of meteoric waters may have caused the dissolution and collapse of the evaporitic layers. The depth, salinity, and size of the water bodies must have changed substantially over time, as shown by the different lithofacies, fossil content, and sedimentary features of the AS rocks. The decreasing depth, increasing salinity, and ever-smaller water bodies that represent the lacustrine and evaporitic stages, seem to reflect a change towards more arid conditions. Some modern hypersaline terminal lakes and playas show sedimentary processes similar to those inferred for the Axamilpa Section and could serve as models to better understand environmental factors. Extensive plains are characteristic of these environments, with high slopes or mountains relatively close to the lake environment. These lakes have very shallow and even intermittent waters, scarce rainfalls and high evaporation, carbonate and evaporitic deposits. For example, Great Salt Lake, Utah, is a shallow calcareous-evaporitic lake (∼ 10 m) influenced along its margins by alluvial systems. Oolites, stromatolites, ostracods and migratory birds are common, and the vegetation is low and open (Kowalewska and Cohen, 1998). The Salton Sea in southern California (Arnal, 1961; Gilmore and Castle, 1983) is a saline lake with scarce peripheral vegetation. It has hydrothermal activity associated with rifting (Harmon, 1966) and combines fluvial and lacustrine sedimentary processes (Arnal, 1961). As in modern analogs, regional events that were taking place simultaneously in central Mexico during

the Paleogene, such as extrusive and intrusive magmatism, extensive faulting and block displacements (e.g. Martiny et al., 2000; Morán-Zenteno et al., 1999), must have determined sedimentation and biota distribution in the Coatzingo Fm to a great extent, along with local factors, such as rain frequency and intensity, changing topography and drainage patterns. The fining-upward evolution of the sequence, inferred as the retrogradation of an alluvial system, is probably the result of the basin extension. Hydrothermal pulses, deduced from the occurrence of magnesite and cherts in the AS and in nearby localities, could represent the existence of magmatic activity in the subsurface. Both ancient and present-day hydrothermalism processes have been reported for this area (Carballido-Sánchez and Delgado-Argote, 1989; Jiménez-Suárez et al., 2001). Similarly, the cyclical deposition of some facies (e.g., associations A, B, D) was probably related either to tectonic and faulting episodes, or to climatic variations, maybe related to the global climatic changes that occurred in that epoch (Frakes et al., 1992; Wolfe, 1994). 5. Paleoecology The fossil record known from the region correlates well with the sedimentological data, thus lending support to paleoecological inferences. For example, the geologic setting indicates the existence of shallow lacustrine environments, which is reinforced by the presence of a flamingo skeleton in the Pie de Vaca locality (CabralPerdomo, 1996). This suggests that environmental conditions (shallow and saline waters) similar to those known from where these birds live today (Mascitti and Bonaventura, 2002) prevailed in this area of the Coatzingo Formation. Flamingos (Phoenicopteridae) have been described from many Eocene and Oligocene strata (Martin, 1983; Olson, 1985). Because extant flamingos are gregarious and migratory (Nager et al., 1996), it is possible that the discovery of this single flamingo skeleton represents a larger population. This notion may also be supported by the fact that the fossil record of aquatic birds in Tepexi de Rodríguez is composed mainly of ichnofossils of their tracks (Cabral-Perdomo, 1996). Further work is needed to correlate precisely all the localities where avian ichnofossils are present, but their presence in the Coatzingo Formation contributes to the understanding of the past habitats and ecological requirements of these birds, as well as their temporal and spatial distribution during the Eocene–Oligocene. Fossil plants from the Axamilpa Section, such as Cedrelospermum, Pistacia and Pseudosmodingium, are

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254

also found in the Ahuehuetes Unit. The Pseudosmodingium species, which today exists in the Tepexi de Rodríguez area, is thought to be endemic to this region and, along with Cedrelospermum, is considered a pioneer in harsh environments (Ramírez and CevallosFerriz, 2002). Cedrelospermum has been referred to subhumid climates in communities of desert scrub (Magallón-Puebla and Cevallos-Ferris, 1994b; Velasco de León and Cevallos-Ferriz, 2000), which suggests similar climatic conditions prevailed in this region during the Oligocene. Furthermore, thermometric inferences from the Ahuehuetes locality fossil plants, estimate annual average temperatures of 10–18 °C (Velasco de León, 1999), and deciduous scrub or chaparral-like communities are thought to have established there in temperate to semiarid areas (i.e. where annual precipitation was less than annual evaporation, and where there were long dry periods) (Ramírez and Cevallos-Ferriz, 2002). Plant-bearing strata in the Ahuehuetes Unit are composed of volcanic ash, and although the Axamilpa Section lacks volcanic components, the major volcanism that occurred in this epoch (Martiny et al., 2000; Morán-Zenteno et al., 1999; Silva-Romo et al., 2000) might have caused environmental disruption in the surrounding areas, which in turn probably influenced the origin and distribution of plants in this region. Pistacia fossils from the Ahuehuetes locality have close relatives that are found today only in restricted points of Asia and one Oligocene locality in Germany, suggesting a long history of exchange between the North American and Eurasian floras (Ramírez and Cevallos-Ferriz, 2002). Given the stratigraphical position of the two localities (Fig. 2), it is possible that some floral elements found in the Ahuehuetes locality evolved earlier in the Eocene and lived around the dry and saline environments represented in the Axamilpa Section. The presence of riparian-related plants from the Ahuehuetes Unit, such as Salix, Populus and some Anacardiaceae (Ramírez-Garduño, 1999), which are also represented in the AS, may indicate that near-river environments became established more than once in the basin. In the Axamilpa Section, however, the presence of Graminidites sp. (Carranza-Sierra, personal communication, 2003) and fossil roots, together with the evidence for high salinity and aridity, would indicate a grass-like vegetation relatively far from the waterways where the riparian communities became established, as in modern saline-lake environments (e.g. Kowalewska and Cohen, 1998; Soria et al., 2000). This fact agrees with previous assumptions of grass-abundant biomes for this area (Martínez-Hernández and Ramírez-Arriaga,

251

2003), and may narrow the existence or riparian environments to only a few places. Pollen assemblages also suggest that the mountain ranges that surrounded the basin were populated mostly by conifers (MartínezHernández and Ramírez-Arriaga, 1999), contrasting lowland xeric and upland mesic vegetations. The distinction between riparian and non-riparian plants may have implications for their distribution and taphonomy. Perhaps near-river plants had a higher potential for fossilization due to their susceptibility of being quickly buried after flooding events (Stromberg et al., 1991), so riparian settings would be suitable places for taphonomical processes. Moreover, higher plant diversity in arid lands is found around the streams (Ali et al., 2000; Stromberg et al., 1991), so diversity of fossil plants in the Coatzingo Formation may be an indicative of riparian environments. The fact that riparian ecosystems may have existed in this region would have profound paleobiological implications. Riparian zones favor plant dispersal and mixing of seeds (Stromberg et al., 1991), and high number of endemisms can be found there (Stohlgren et al., 2005), especially in desert springs. These systems may also serve as dispersal ways for plant and animal species (Baker, 1986). Furthermore, resources such as drinking water, shade, food, and shelter, would have been an attraction for animals and may have helped as corridors for their dispersal within arid zones. In that sense, these environments could have had an influence on the biodiversity and distribution of plants and animals, and serve as spots for endemisms. The presence of this resource rich environment is further supported by the abundance of fossil tracks related to camel-like animals at various stratigraphic levels of the AS and their extensive occurrence in the Tepexi de Rodríguez region (Fig. 1A). The tracks indicate that these animals traversed the area regularly; furthermore, they are perhaps an indication of their dispersion process. Camelids originated in North America during the Paleocene–Eocene in North America (Stearn and Carrol, 1989), and 2 events of dispersion, one to Eurasia and Africa, and one to South America, have been proposed for the Miocene and Pliocene–Pleistocene respectively (van der Made et al., 2002); however, the fossil record of this group is incomplete and the starting points of these migrations are unknown. Their association with desiccation cracks and rhizocretions in the AS suggests animal movement across emergent areas, although resources from streams would have been their main attraction. Perhaps the Tepexi–Coatzingo basin served as a migratory path for these animals, as well as a place to drink, feed or breed.

252

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254

Finally, the recurrence of some of the taxa (e.g. fossil tracks, ostracods, plants) in various levels of the Coatzingo Fm indicate their permanence over time, and suggests that either the environmental conditions necessary for their development remained stable or that they adapted to the changing environments. 6. Conclusions While a wide variety of habitats exist today in this region of the globe, where life forms have existed and evolved, little is known about these diverse scenarios during the Cenozoic. Interpreting past life and the places where they lived opens the opportunity to understand particular biological or geological phenomena, and to compare them at different moments, which offers a more dynamic understanding of the natural processes. The depositional model of the Axamilpa Section provides a conception of the nature and evolution of a dry paleoenvironment over time, and gives organisms, or communities, a context in which they were able to evolve. Providing an environmental context for fossils allows a better understanding of their ecology and distribution over time and opens the opportunity for discussions with a timeline vision. Moreover, fossils and paleoenvironments from the Axamilpa Section reinforce the importance of understanding the reciprocal biotic and abiotic interactions and influences. While arid and semiarid areas in Mexico are widely distributed and assumed to be of a more recent origin, the new evidence provided by the well-preserved sediments of the Axamilpa Section expands our view on how arid environments may have looked like at that time, their depositional evolution, and the influence of physical factors on the local biota and the environment. This study may encourage more detailed studies in the Coatzingo Formation and other Cenozoic basins in Mexico that can provide data to further precise timeframes of sediment deposition, reconstruct paleoenvironments, and to understand the evolution, biogeography and adaptation of organisms in tropical North America. Acknowledgments We thank Dr. Ana Luisa Carreño, M.Sc. Claudia Carranza-Sierra, Ing. Ciro Díaz, Dr. Jerjes Pantoja Alor, Ing. Diego Aparicio from the Institute of Geology, UNAM. Sr. Félix Aranguti and his family from the

Museum of Paleontology (Tepexi de Rodríguez); Dr. Gonzalo Pardo from the University of Zaragoza, Spain; Dr. José Carlos García Ramos from the University of Oviedo, Spain; the staff of the Biological Sciences and Earth Sciences Postgraduate Department of UNAM; Dr. Scott Bates, Dr. Robin Renaut, Dr. Blas Valero Garcés, and Dr. Bruce Sellwood for their comments. This research was supported by Project DGAPA (IN 208500) to Sergio R.S. Cevallos Ferriz. References Ali, M.M, Dickinson, G., Murphy, K.J., 2000. Predictors of plant diversity in a hyperarid desert wadi ecosystem. Journal of Arid Environments 45, 215–230. Arenas, C., Alonso-Zarza, A.M., Pardo, G., 1999. Dedolomitization and other early diagenetic processes in Miocene lacustrine deposits, Ebro Basin (Spain). Sedimentary Geology 125, 23–45. Arnal, R.E., 1961. Limnology, sedimentation, and microorganisms of the Salton Sea, California. Geological Society of America Bulletin 72 (3), 427–478. Baker, H.G., 1986. Patterns of plant invasion in North America. In: Mooney, H.A., Drake, J.A. (Eds.), Ecology of Biological Invasions of North America and Hawaii. Springer-Verlag, New York, pp. 44–57. Bhattacharya, J., Giosan, L., 2003. Wave-influenced deltas: geomorphological implications for facies reconstruction. Sedimentology 50, 187–210. Bustillo, M.A., Arribas, M.E., Bustillo, M., 2003. Dolomitization and silicification in low-energy lacustrine carbonates (Paleogene, Madrid Basin, Spain). Sedimentary Geology 151, 107–126. Buitrón, B.E., Malpica-Cruz, R., 1987. Tepexi de Rodríguez, Puebla, una localidad fosilífera famosa de México: I. Congreso Nacional de Paleontología; Libreto Guía de la Excursión. Sociedad Mexicana de Paleontología. 24 pp. Cabral-Perdomo, M.A., 1995. Los icnofósiles de vertebrados terrestres del Terciario Tardío del área de Tepexi de Rodríguez, Estado de Puebla. Bachelor's Thesis. Universidad Nacional Autónoma de México, Facultad de Ciencias, Mexico. Cabral-Perdomo, M.A., 1996. A flamenco's tafogliph from the Pie de Vaca Formation (late Cenozoic), near Tepexi de Rodríguez, Central Mexico: an unusual paleontological phenomenon. Journal of Vertebrate Paleontology 16 (3), 25A. Calvillo-Canadell, L., Cevallos-Ferriz, S.R.S., 2002. Bauhcis moranii gen. et nov. (Cercideae, Caesalpinieae), an Oligocene plant from Tepexi de Rodríguez, Puebla, Mex., with leaf architecture similar to Bauhinia and Cercis. Review of Palaeobotany and Palynology 122 (3–4), 171–184. Carballido-Sánchez, E.A., Delgado-Argote, L.A., 1989. Geología del cuerpo serpentinítico de Tehuitzingo, Estado de Puebla – interpretación preliminar de su emplazamiento. Revista del Instituto de Geología 8 (2), 134–148. Carranza-Sierra, C., 2001. Palinoestratigrafía del Grupo Balsas. Implicaciones paleoambientales, climáticas y cronoestratigráficas (Municipios de Zacapala and Coatzingo, Estado de Puebla). Bachelor's Thesis. Facultad de Ciencias, UNAM. México. Carranza-Sierra, C., Martínez-Hernández, E., 2002. Balsas Group palynostratigraphy: palaeoenvironmental and chronostratigraphic implications, Zacapala and Coatzingo municipalities, Puebla State, Mexico. Palynology 26, 262.

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254 Carranza-Sierra, C., Martínez-Hernández, E., 2003. Paleogene palynostratigraphy of the paleolake (Tepexi Coatzingo) at the Puebla State, Mexico. Palynology 27, 219. Colombo, F., 1992. Abanicos aluviales. In: Arche, A. (Ed.), Sedimentología, vol. 1. Consejo Superior de Investigaciones Científicas, Madrid. Davis Jr., R.A., 1983. Depositional Systems: A Genetic Approach to Sedimentary Geology. Prentice-Hall, New Jersey. Davies, S.J., Gibling, M.R., 2003. Architecture of coastal and alluvial deposits in an extensional basin: the Carboniferous Joggins Formation of eastern Canada. Sedimentology 50, 415–439. Dunagan, S.P., Turner, C.E., 2004. Regional paleohydrologic and paleoclimatic settings of wetland/lacustrine depositional systems in the Morrison Formation (Upper Jurassic); Western Interior, USA. Sedimentary Geology 167, 269–296. Eugster, H.P., 1980. Lakes Magadi, Kenya, and its precursors. In: Nissenbaum, A. (Ed.), Hypersaline Brines and Evaporitic Environments. Developments in Sedimentology, vol. 28. Elsevier, pp. 195–232. Ferrari, L., López-Martínez, M., Aguirre-Díaz, G., Carrasco-Núñez, G., 1999. Space-time patterns of Cenozoic arc volcanism in central Mexico: from the Sierra Madre Occidental to the Mexican Volcanic Belt. Geology 27 (4), 303–306. Frakes, L.A., Francis, J.E., Syktus, J.I., 1992. Climate Modes of the Phanerozoic. The History of the Earth's Climate Over the Past 600 Million Years. Cambridge University Press, Cambridge. Gilmore, T.D., Castle, R.O., 1983. Tectonic preservation of the divide between the Salton basin and the Gulf of California. Geology 11, 474–477. Harmon, C., 1966. Isotopic composition and origin of the Red Sea and Salton Sea geothermal brines. Science 154 (3756), 1544–1548. Henderson, P.A., 1990. Fresh Water Ostracods. Universal Book Services. Avon, UK. Jiménez-Suárez, G., Flores Márquez, L., Chávez, R., Garfias, J., Campos-Enríquez, O., 2001. La tectónica and la calidad del agua subterránea en el valle de Puebla, 21(3). Unión Geofísica Mexicana (GEOS), p. 182. Kashiyama, A., Fastovsky, D.E., Rutherford, S., King, J., Montellano, M., 2003. Genesis of a locality of exceptional fossil preservation: paleoenvironments of Tepexi de Rodríguez (mid-Cretaceous, Puebla, Mexico). Cretaceous Research 24, 407–431. Kowalewska, A., Cohen, A.S., 1998. Reconstructions of paleoenvironments of the Great Salt Lake Basin during the late Cenozoic. Journal of Paleolimnology 20, 381–407. Lowe, D.R., 1979. Sediment gravity flows: their classification and some problems of application to natural flows. In: Doylle, L.J., Pilkey, O.H. (Eds.), Geology of Continental Slopes. (SEPM) Special Publication, vol. 27, pp. 75–82. Magallón-Puebla, S., Cevallos-Ferriz, S.R.S., 1994a. Fossil legume fruits from tertiary strata of Puebla, Mexico. Canadian Journal of Botany 72, 1027–1038. Magallón-Puebla, S., Cevallos-Ferris, S.R.S., 1994b. Latest occurrence of the extinct genus Cedrelospermum (Ulmaceae) in North America; Cedrelospermum manchesteri from Mexico. Review of Palaeobotany and Palynology 81, 115–128. Martin, L.D., 1983. The origin and early radiation of birds. In: Brush, A.H., Clark, G.A. (Eds.), Perspectives in Ornithology. Cambridge University Press, Cambridge, pp. 291–338. Martínez-Hernández, E., Ramírez-Arriaga, E., 1999. Palinoestratigrafía de la región de Tepexi de Rodríguez, Puebla, México. Implicaciones cronoestratigráficas. Revista Mexicana de Ciencias

253

Geológicas, UNAM, vol. 16. Instituto de Geología and Sociedad Geológica Mexicana, México, pp. 187–207. Martínez-Hernández, E., Ramírez-Arriaga, E., 2003. Early abundance of grasses (Poaceae) during the Oligocene in southern North America. Palynology (Abstracts) 28 (1), 233–269. Martiny, B., Martínez-Serrano, R.G., Morán-Zenteno, D., MacíasRomo, C., Ayuso, R.A., 2000. Stratigraphy, geochemistry and tectonic significance of the Oligocene magmatic rocks of western Oaxaca, Southern Mexico. Tectonophysics 318, 71–98. Marzo, M., 1992. Sistemas fluviales de alta sinuosidad. In: Arche, A. (Ed.), Sedimentología, vol. 1. Consejo Superior de Investigaciones Científicas, Madrid. Mascitti, V., Bonaventura, S.M., 2002. Patterns of abundance, distribution and habitat use of flamingos in the High Andes, South America. Waterbirds 25 (3), 358–365. McCarthy, J.P., Plint, G.A., 2003. Spatial variability of palaeosols across Cretaceous interfluves in the Dunvegan Formation, NE British Columbia, Canada: palaeohydrological, palaeogeomorphological and stratigraphic implications. Sedimentology 50, 1187–1220. Miall, A.D., 1992. Alluvial deposits. In: Walker, R.G. (Ed.), Facies Models: Response to Sea Level Change. Geological Association of Canada, pp. 119–142. Miall, A.D., 1996. The Geology of Fluvial Deposits: Sedimentary Facies, Basin Analysis and Petroleum Geology. Springer-Verlag Inc., Berlin. Morán-Zenteno, D.J., Tolson, G., Martínez-Serrano, R., Martiny, B., Schaaf, P., Silva-Romo, G., Macías-Romo, C., Alba-Aldave, L.A., Hernández-Bernal, M.S., Solís Pichardo, G.N., 1999. The Tertiary arc-magmatism of the Sierra Madre del sur, Mexico, and the transition to the volcanic activity of the Trans-Mexican Volcanic Belt. Journal of South American Earth Sciences 12, 513–535. Mutti, E., Davoli, G., Tinterri, R., Zavala, C., 1996. The importance of fluvio–deltaic systems dominated by catastrophic flooding in tectonically active basins. Memorie di Scienze Geologiche 48, 233–291. Nager, R.G., Johnson, A.R., Boy, V., Rendón-Martos, M., Calderón, J., Cezilly, F., 1996. Temporal and spatial variation in dispersal of the great flamingo (Phoenicopterus ruber roseus). Oecologia 107, 204–211. Olson, S.L., 1985. The fossil record of birds. In: Farner, D.S., King, J.R., Parkes, K.C. (Eds.), Avian Biology, vol. 8. Academic Press, New York, pp. 79–256. Ortega-Gutierréz, F., Elías-Herrera, M., Reyes, S.M., Macías, R.C., López, R., 1999. Late Ordovician–Early Silurian continental collisional orogeny in southern Mexico and its bearing on Gondwana–Laurentia connections. Geology 27, 719–722. Pantoja-Alor, J., 1990. Geología and paleoambiente de la cantera de Tlayúa, Tepexi de Rodríguez, Puebla. Revista del Instituto de Geología, UNAM 9 (2), 156–169. Pantoja-Alor, J., Malpica-Cruz, V.M., Galguera, A.G., 1988. Geología de los alrededores de la cantera de Tlayúa, Tepexi de Rodríguez, Puebla. UNAM Instituto de Geología, 3er Simposio de Geologia Regional, México. Memoria, pp. 61–69. Pirajno, F., Grey, K., 2002. Chert in the Palaeoproterozoic Bartle member, Killara Formation, Yerrida Basin, Western Australia: a rift-related playa lake and termal spring environment? Precambrian Research 113, 169–192. Ramírez, J.L., Cevallos-Ferriz, S.R.S., 2002. A diverse assemblage of Anacardiaceae from Oligocene sediments, Tepexi de Rodríguez, Puebla, Mexico. American Journal of Botany 89 (3), 535–545. Ramírez-Garduño, J.L., 1999. Análisis foliar de Anacardiaceae, Berberidaceae and Salicaceae en Los Ahuehuetes (Oligoceno),

254

H. Beraldi-Campesi et al. / Sedimentary Geology 191 (2006) 227–254

Tepexi de Rodríguez, Puebla. Mesters Thesis. Facultad de Ciencias, UNAM. México. Reading, G.G., 1986. Sedimentary Environments and Facies, 2nd ed. Blackwell Scientific Publications, Oxford. Reid, R.P., Visscher, P.T., Decho, A.W., Stolz, J.F., Bebout, B.M., Dupraz, C., Macintyre, I.G., Paerl, H.W., Pinckney, J.L., PrufertBebout, L., Steppe, T.F., Desmarais, D.J., 2000. The role of microbes in accretion, lamination and early lithification of modern marine stromatolites. Nature 406, 989–992. Rodríguez de la Rosa, R., Aguillón-Martínez, M.C., López-Espinoza, J.L., 2005. The fossil record of vertebrate tracks in Mexico. Ichnos 11, 27–37. Rust, B.R., 1980. Coarse alluvial deposits. In: Walker, R.G. (Ed.), Facies Models. Geological Association of Canada, pp. 9–21. Schroeder, P.A., Kim, J.G., Melear, N.D., 1997. Mineralogical and textural criteria for recognizing remnant Cenozoic deposits on the Piedmont: evidence from Sparta and Greene County, Georgia, U.S. A. Sedimentary Geology 108, 195–206. Seong-Joo, L., Browne, K.M., Golubic, S., 2000. On stromatolite lamination. In: Riding, R.E., Awramik, S.M. (Eds.), Microbial Sediments. Springer-Verlag, Berlin-Heidelberg, pp. 16–24. Silva-Romo, G., Mendoza-Rosales, R.C., Martiny, B., 2000. Acerca del origen de las cuencas cenozoicas de la zona cortical extendida del sur de México: un ejemplo en la región Mixteca. Boletín informativo. Época II, vol. 20(3). Unión Geofísica Mexicana (GEOS), pp. 326–327. Soria, A.R., Luzón, A., Pérez, A., Roc, A.C., 2000. Evolución vertical del relleno sedimentario Holoceno de la Laguna de Gallocanta, Cordillera Ibérica, NE de España. Geotemas 1 (2), 221–226.

Stearn, C., Carrol, R., 1989. Paleontology: The Record of Life. Wiley & Sons, New York. Stohlgren, T.J., Guenther, D.A., Evangelista, P.H., Alley, N., 2005. Patterns of plant species richness, rarity, endemism, and uniqueness in an arid landscape. Ecological Applications 15 (2), 715–725. Stromberg, J.C., Patten, D.T., Richter, D.T., 1991. Flood flows and dynamics of Sonoran riparian forests. Rivers 2, 221–235. Swirydezuk, K., Wilkinson, B.H., Smith, G.R., 1979. The Pliocene Glenns Ferry oolite: lake-margin carbonate deposition in the Southwestern Snake River Plain. Journal of Sedimentary Petrology 49, 995–1004. van der Made, J., Morales, J., Sen, S., Aslan, C., 2002. The first camel from the Upper Miocene of Turkey and the dispersal of the camels into the Old World. Systematic Palaeontology 1, 117–122. Velasco de León, M.P., 1999. Estudio paleoecológico de una comunidad del Terciario en el estado de Puebla. Doctoral Thesis. Facultad de Ciencias, UNAM. México. Velasco de León, P., Cevallos-Ferriz, S.R.S., 2000. Leaves of Cercocarpus mixteca n. (Rosaceae) from Oligocene sediments, near Tepexi de Rodríguez, Puebla. Review of Palaeobotany and Palynology 111, 285–294. Wolfe, J.A., 1994. Tertiary climatic changes at middle latitudes of western North America. Palaeogeography, Palaeoclimatology, Palaeoecology 108, 195–205.