Andean tectonics as a cause for changing drainage patterns in

Andean tectonics as a cause for changing drainage patterns in
Miocene northern South America
Carina Hoorn Hugo de Vries-Laboratorium, Kruislaan 318, 1098 SM Amsterdam, Netherlands
Javier Guerrero
Ingeominas, Diagonal 53 no. 34-53, Bogotá, Colombia
Gustavo A. Sarmiento
Maria A. Lorente Maraven S. A., Apartado 829, Caracas 1010A, Venezuela
New data from Neogene strata in northern South America suggest that Miocene tectonism in the northeastern Andes was responsible for the genesis of the Amazon River and changes in the drainage patterns of other major rivers such as the Magdalena and the
Orinoco. Here we present a new model for the paleogeographic
evolution of northern South America during the Miocene. In the
early Miocene, a large part of the drainage of northwest Amazonia
was directed northward along the paleo–Orinoco river system to a
delta in Lake Maracaibo. Uplift of the Eastern Cordillera in the late
middle Miocene caused the first development of the Amazon River;
however, no connection with the Atlantic was established, and the
Amazon fed the paleo–Orinoco river system, which drained toward
the Caribbean. Substantial Andean uplift in the late Miocene resulted in major changes in paleogeography: the Orinoco changed its
course, the Amazon established a connection to the Atlantic, caus-
ing the drowning of carbonate platforms, and the Amazon-Caribbean connection was closed. Thus the drainage and paleogeography
of northern South America in the Miocene were strongly controlled
by tectonic movements in the northeastern Andes.
The northeastern Andes is composed of the Eastern Cordillera,
which bifurcates at 108N into the Santander massif and the Sierra de
Perijá in the northwest and the Mérida Andes (or Venezuelan
Andes) in the northeast (Fig. 1). These two mountain ranges are
separated by the Táchira saddle, a depression that resulted from the
relative movement of the Eastern Cordillera and the Mérida Andes.
It contains a 7-km-thick Jurassic to Quaternary sedimentary sequence, which absorbed the stress produced by the motion of the
two blocks. The Eastern Cordillera is separated from the Central
Cordillera by the Magdalena Valley, and in the north, the Mérida
Figure 1. Paleogeographic maps for late Oligocene to Holocene of
northern South America showing development of northeastern mountain ranges and effect on drainage patterns in peripheral areas. WC—
Western Cordillera, CC—Central Cordillera, EC—Eastern Cordillera,
GM—Garzón massif, SP—Sierra de Perijá, SM—Santander massif,
MA—Mérida Andes, T—Táchira saddle, M—Lake Maracaibo, MV—
Magdalena Valley, SSM—Sierra de Santa Marta, B—Bogotá basin,
FB—Falcon basin; BAB—Barinas-Apure basin, LLB—Llanos basin,
SB—Solimões basin, AB—Amazonas basin.
Geology; March 1995; v. 23; no. 3; p. 237–240, 1 figure.
Andes is limited by the Falcon basin (Fig. 1). Along the Andean
fold-and-thrust belt, in the west, a deep foreland basin (BarinasApure, Llanos basins) extends toward the Guyana shield and the
intracratonic basins (Amazonas, Solimões basins). These basins are
subdivided and limited by structural highs (arches) in the basement
(Fig. 1).
The Neogene-Quaternary uplift of the northeastern Andes and
development of the sub-Andean fold-and-thrust belt resulted from
plate-tectonic reorganization. The Farallon plate broke up into the
Nazca and Cocos plates at the end of the late Oligocene, and a new
spreading center was created, resulting in increased convergence
rates at the plate margins (Wortel and Cloetingh, 1981; Wortel,
1984). The increased convergence rates triggered tectonic activity in
the Andes, an effect that was also observed during an earlier phase
of rapid convergence in the middle Eocene (Pardo-Casas and Molnar, 1987). A secondary factor that influenced the tectonic history
in the area was the west-to-east motion of the Caribbean plate,
which closed the Panama isthmus between 3.7 and 3.4 Ma (Sykes
et al., 1982; Duque-Caro, 1990).
Uplift of the northeastern Andes occurred in discrete periods
(Steinmann, 1930; Van der Hammen, 1961; Megard, 1984). Studies
of the fill of the intramontane and foreland basins show that there
were at least six phases of uplift and tectonic quiescence between the
Late Cretaceous and the Pleistocene (Van der Hammen, 1961; Van
der Hammen et al., 1973; Zambrano et al., 1971; Gonzalez de Juana
et al., 1980). The main uplift of the northeastern Andes occurred
between the late Oligocene and the Pleistocene, with a climax in the
Pliocene-Pleistocene. This led to erosion of a large volume of sediments resulting in thick Miocene-Pleistocene molasse successions
in the sub-Andean basins. Age control for this major period of tectonism is provided by fission-track, 40Ar/39Ar, paleomagnetic, and
tephra dates combined with biostratigraphic evidence (Kohn et al.,
1984; Shagam et al., 1984; Van der Wiel, 1991; Guerrero, 1993;
Flynn et al., 1994; Andriessen et al., 1994).
The effect of Andean tectonics on paleogeography has until
now been discussed only at a local level. Here we present an integration of our own published data from northwest Amazonia and
the Magdalena Valley with unpublished data from the Mérida
Andes and the Llanos basin provided by Maraven and Ecopetrol.
From these new data and a review of previous local studies in the
region, we have developed a new paleogeographical model for the
Miocene of northern South America.
Major changes in sediment provenance, paleotransport directions, and paleoenvironment indicate that the depositional history
of northwest Amazonia was strongly influenced by the uplift of the
Eastern Cordillera during the Miocene (Hoorn, 1993). In the early
to early middle Miocene, the intracratonic Amazonas and Solimões
basins were dominated by a low-sinuosity, northwest-directed fluvial
system (Fig. 1A). The stable heavy-mineral assemblage (zircon,
tourmaline, rutile, etc.) and the paleotransport directions indicate
that the Guyana shield was the main sediment source. At that time,
the vegetation in the area was dominated by palm swamps and riverine and tropical lowland forest. Episodic marine incursions are
reflected by the presence of mangrove pollen and marine palynomorphs such as microforaminifera and dinoflagellate cysts (Hoorn,
During the middle Miocene, drainage patterns and provenance
changed. In the middle to late Miocene, sediments were deposited
in northwest Amazonia by a fluvio-lacustrine system with an eastward transport direction (Fig. 1B). These sediments are characterized by an unstable heavy-mineral suite (epidote, garnet, chloritoid,
etc.) that is thought to have had its origin in the Cretaceous met238
amorphic rocks of the Ecuadorian Andes (Hoorn, 1993). The environment was dominated by extensive wetlands, and the vegetation
was palm swamps, riverine forest, and relatively abundant ferns,
floating meadows, and aquatic taxa. This fluvio-lacustrine system
was the precursor of the Amazon River, which developed fully as a
transcontinental drainage system in late Miocene time (Fig. 1C)
(Campbell, 1992; Hoorn, 1993). Marine influences are reflected by
the presence of marine palynomorphs, mangrove pollen, and molluscs and ostracods tolerant of brackish conditions. Marine incursions, possibly via a connection toward the Caribbean, were correlated to phases of global sea-level rise during the Burdigalian,
Langhian, and Serravallian (Hoorn, 1993, 1994). The middle to upper Miocene sequence is unconformably overlain by an upper Miocene to Pleistocene fluvial, conglomeratic unit with pebbles of Andean origin (chert, lithic fragments, quartz). This unit is possibly
contemporaneous with the main uplift of the Eastern Cordillera.
In the upper and middle Magdalena Valley (Fig. 1), deposition
of Neogene fluvial sequences was controlled almost entirely by volcanism and tectonism in the Eastern Cordillera and Central Cordillera (Van Houten and Travis, 1968; Wellman, 1970; Howe, 1974;
Van der Wiel, 1991). On the basis of sedimentological, paleomagnetic, 40Ar/39Ar, and fission-track data, it has been suggested that
between 13.5 and 12.9 Ma (early middle Miocene), the area of the
present Magdalena Valley was dominated by a meandering to
braided river system with an east-southeast transport direction
(Guerrero, 1993; Flynn et al., 1994). This system drained the Central
Cordillera and flowed into the large foreland basin that extended
from the Central Cordillera to the Guyana shield (Fig. 1A). Between 12.9 and 11.8 Ma (late middle Miocene), the Eastern Cordillera started developing; at the time, a meandering fluvial system
existed, with flow to the north and northeast in addition to the still
predominant flow direction to the east and southeast. At 11.8 Ma,
current directions shifted completely to the west in a meandering to
anastomosing fluvial system. This shift indicates the existence of a
new sediment source area to the east. Regional stratigraphy and
sedimentological data from the upper and middle Magdalena Valley
confirm the appearance of the Eastern Cordillera as a continuous
range at about 11.8 Ma (Fig. 1B). This new range was high enough
to permanently divide the former foreland basin into the Magdalena
and Llanos basins (Van der Wiel, 1991; Guerrero, 1993; Flynn et al.,
According to Guerrero (1993), after a short period of erosion
and no sedimentation (11.5 to 10.1 Ma), deposition of conglomerates in a braided fluvial system with north and northeast transport
directions began at about 10.1 Ma (early late Miocene; Fig. 1C).
These oldest known relics of the Magdalena River were deposited
at the same time as an episode of volcanism and uplift of the Central
Cordillera and Eastern Cordillera.
The only known Neogene marine influence in the upper and
middle Magdalena Valley occurred during the earliest Miocene, as
indicated by the brackish-water mollusc fauna (Nuttall, 1990). Furthermore, freshwater fossil fishes from the upper Magdalena Valley
indicate that a river connection with northwest Amazonia still existed during the early middle Miocene (Lundberg and Chernoff,
In the Mérida Andes (Fig. 1), rapid exhumation began in late
Oligocene time and accelerated further during the Miocene, as deduced from fission-track analysis (Kohn et al., 1984) and from petrographic, sedimentological, seismic, and biostratigraphic studies of
the northern flank of the Mérida Andes. It has been speculated that
GEOLOGY, March 1995
uplift of the Mérida Andes might have started as early as the late
Eocene. However, evidence indicates that uplift did not take place
until the late Oligocene, when unroofing of Cretaceous and Paleogene strata resulted in syntectonic deposition of chert pebbles, reworked foraminifera and palynomorphs in the northern foredeep of
the Mérida Andes (V. Rull, 1992, unpublished report; R. Pitelli and
L. Echeverria, 1992, unpublished report; R. Higgs and S. Mederos,
1993, unpublished report; Higgs, 1993). In the early and middle
Miocene, input of Andean clastic sediments increased significantly;
however, the upper Miocene conglomerates are the first evidence of
true Andean molasse in the northern foredeep. During the early
Miocene, the depositional environment in the foredeep was characterized by an alternation of coastal and alluvial plain conditions,
whereas from late early Miocene to middle Miocene, coastal
marshes with marine intervals prevailed. Coarse, deltaic deposition
dominated during a period of strong uplift in the late Miocene
(Maraven S. A., 1993).
During early and middle Miocene time, deltaic deposition related to the paleo–Orinoco river system dominated the area of the
Falcon basin (Diaz de Gamero, 1993) (Fig. 1, A and B). At the end
of the middle Miocene, when the main uplift of the northeastern
Andes and the western part of the Caribbean mountains occurred,
the Orinoco shifted toward the east, but a well-defined course only
developed during the Pliocene-Pleistocene in eastern Venezuela
(Fig. 1C).
The role of Andean tectonics on paleogeographic change in
northern South America is further emphasized in the northeastern
Andes and the adjacent Llanos basin.
The middle Miocene–Pleistocene phase of uplift of the Eastern
Cordillera is marked in the intramontane basins by conglomeratic
alluvial-fan and gravity-flow deposits, which indicate massive movements of sediment at the time of the uplift (Helmens, 1990; Van der
Wiel, 1991). Biostratigraphic data (Van der Hammen et al., 1973;
Hooghiemstra, 1984) and chronostratigraphic evidence (Helmens,
1990; Andriessen et al., 1994) obtained from the Neogene-Quaternary sediment sequence in the intramontane Bogotá area (Fig. 1C)
indicate a major phase of tectonic uplift between 5 and 3 Ma. Pollen
analysis of sediments in the Bogotá basin has shown that during its
emergence, the Cordillera passed through different climate belts,
and so the vegetation changed from the initial tropical lowland type
to a cold, mountainous type (Van der Hammen et al., 1973; Hooghiemstra, 1984).
The Santander massif arose between early and middle Miocene
time, almost contemporaneously with the uplift of the Eastern Cordillera (Fig. 1, B and C). On the basis of fission-track data, Shagam
et al. (1984) inferred the uplift of the western part of the massif at
19 –14 Ma and that of the central part at 16 –14 Ma with a late
Miocene to early Pliocene (7– 4 Ma) uplift recorded for the central
and northern part. Along the eastern foredeep, in the Maracaibo
area, mud, sand, and conglomerate were deposited in a late Miocene to early Pliocene(?) low-energy deltaic complex, which was fed
from the massif. During the uplift of the Santander massif, progressively coarser sediments were supplied to the delta (Van Houten
and James, 1984).
Sedimentological data from the Táchira saddle (Fig. 1C) confirm the ages of uplift inferred from fission-track analyses (Kohn et
al., 1984; Shagam et al., 1984) and indicate a strong effect on sedimentation of the uplift of the Mérida Andes. Provenance and current directions of Miocene-Pliocene sequences in the Táchira saddle
indicate that during deposition of these sequences, the Mérida
Andes was the main sediment source and no sediment input was
provided by the Eastern Cordillera (Macellari, 1984). Conglomerate
GEOLOGY, March 1995
and sand derived from the Mérida Andes were deposited in an
anastomosing to meandering fluvial system in the Táchira area.
In the Llanos basin (Fig. 1), lower to middle Miocene deposits
indicate coastal and lagoonal conditions with marine episodes (Instituto Colombiano del Petroleo, 1992, unpublished report). The
Miocene coastal environment in the Llanos basin relates to northwest Amazonia and the foredeep of the Mérida Andes, where
coastal environments with marine incursions were also recorded in
early and middle Miocene time. The base of the upper Miocene–
Pleistocene molasse sequence is a regional unconformity in the
Llanos basin which relates to the uplift of the Eastern Cordillera
(Miller and Etayo-Serna, 1972; Robertson Research, 1986, unpublished report).
The uplift of the northeastern Andes played a key role in the
paleogeographic development of northern South America. Our integration of biostratigraphic, geochronological, and sedimentological data shows that plate-tectonic adjustments and subsequent tectonic events led to a reorganization of drainage patterns in northern
South America during the Miocene. The late middle Miocene uplift
of the Eastern Cordillera in the Colombian Andes coincided with
widespread global tectonism, with a major cooling event during the
establishment of the Antarctic ice cap, and with deep-sea hiatus
NH3 dated as 12.9 to 11.8 Ma by Keller and Barron (1987).
On the basis of data from the sub-Andean basins we postulate
the following paleogeographical model for the Miocene (Fig. 1).
This model is a refinement of that of Harrington (1962).
During the late Oligocene to early middle Miocene the Central
Cordillera was drained by a fluvial system that had an eastern transport direction (Fig. 1A), and the Eastern Cordillera was embryonic.
In northwest Amazonia, fluvial systems drained the Guyana shield
and had northwest transport directions; they probably formed tributaries of the ancient Orinoco river system, which had a northward
course, toward a delta in the present Lake Maracaibo. The Central
Cordillera was drained by fluvial systems that had an eastward direction and also formed tributaries to the Orinoco. The Mérida
Andes was a more prominent mountain range that had developed
already in the late Oligocene and early Miocene and supplied sediment to the northern foredeep and the Barinas-Apure basin. Periods of global high sea level (Burdigalian) caused marine incursions
connecting the present Lake Maracaibo, Llanos region, and northwest Amazonia.
During the late middle Miocene (Fig. 1B), the first effects of the
rise of the Eastern Cordillera were noticeable in northwest Amazonia with the development of the Amazon River. At the time,
however, no connection existed with the Atlantic. The paleo–Amazon River formed a fluvio-lacustrine system with an estuarine character that was still partly connected to the paleo-Orinoco. The area
of the present upper and middle Magdalena Valley, previously connected to the Llanos and Amazonian lowlands, became an isolated
area, and a new west-directed fluvial system developed, originating
in the Garzón massif. The Mérida Andes were uplifted more, and
fluvial systems provided large amounts of clastic sediments to the
northern foredeep, Barinas-Apure basin, and Táchira area. During
periods of high sea level (Langhian and Serravallian), Maracaibo,
the Llanos, and Amazonia were still connected.
Between the late Miocene and the Holocene, the Andes attained their present configuration (Fig. 1C). The late Miocene
marked the start of a paroxysm in the uplift of the northeastern
Andes and represents the most dynamic episode during the entire
Miocene. During this time, the Eastern Cordillera, the Santander
massif, and the Mérida Andes were all uplifted further. In the foredeeps, molasse sediments were deposited by alluvial fans and
braided fluvial systems. The Amazon River evolved as a transcontinental drainage system and covered the carbonate platforms of the
Atlantic shelf. At the same time the Orinoco changed to its present
course and abandoned the area of Lake Maracaibo. In the upper
Magdalena Valley, the Magdalena River started its development as
a braided river, with the Central Cordillera and the newly formed
Eastern Cordillera as source areas. The Amazonian-Caribbean connection was closed by late Miocene tectonic events and a relatively
low global sea level.
Supported by the Dutch Foundation for the Advancement of Tropical
Research (WOTRO, grant W 75-312). We thank Maraven S. A., Ecopetrol,
and Ingeominas for their cooperation and for providing proprietary information; P. L. de Boer and R. Wortel for valuable discussions and for encouraging us to write this paper; and T. L. Gubbels and an anonymous
reviewer for their constructive reviews.
Andriessen, P. A. M., Helmens, K. F., Hooghiemstra, H., Riezebos, P. A.,
and Van der Hammen, T., 1994, Absolute chronology of the PlioceneQuaternary sediment sequence of the Bogotá area, Colombia: Quaternary Science Reviews v. 12, p. 483–503.
Campbell, A. E., 1992, Unconformities in seismic records and outcrops
[Ph.D. thesis]: Amsterdam, Free University, 187 p.
Diaz de Gamero, M. L., 1993, The changing course of the Orinoco River
during the Neogene [abs.]: American Association of Petroleum Geologists Bulletin, v. 77, p. 313.
Duque-Caro, H., 1990, Neogene stratigraphy, paleoceanography and paleobiogeography in northwest South America and the evolution of the
Panama Seaway: Palaeoclimatology, Palaeogeography, Palaeoecology,
v. 77, p. 203–234.
Flynn, J. J., Guerrero, J., and Swisher, C. C., 1994, Geochronology of the
Honda Group (Middle Miocene), Colombia, in Kay, R. F., et al., eds.,
A history of neotropical fauna: Vertebrate paleobiology of the Miocene
of tropical South America: Washington, D. C., Smithsonian Institution
Gonzalez de Juana, C., Iturralde, J. M., and Picard, X., 1980, Geologı́a de
Venezuela y sus cuencas petrolı́feras: Caracas, Ediciones Foninves,
2 volumes, 1031 p.
Guerrero, J., 1993, Magnetostratigraphy of the upper part of the Honda
Group and Neiva Formation. Miocene uplift of the Colombian Andes
[Ph.D. thesis]: Durham, North Carolina, Duke University, 108 p.
Harrington, H. J., 1962, Paleogeographic development of South America:
American Association of Petroleum Geologists Bulletin, v. 46,
p. 1773–1814.
Helmens, K. F., 1990, Neogene-Quaternary geology of the high plain of
Bogotà, Eastern Cordillera, Colombia (stratigraphy, paleoenvironments and landscape evolution): Berlin, J. Cramer, Dissertationes Botanicae, v. 163, 202 p.
Higgs, R., 1993, Sedimentological evidence for early uplift (Oligocene) of
the Venezuelan Andes [abs.]: American Association of Petroleum Geologists Bulletin, v. 77, p. 325.
Hooghiemstra, H., 1984, Vegetational and climatic history of the high plain
of Bogotá, Colombia: A continuous record of the last 3.5 million years:
Berlin, J. Cramer, Dissertationes Botanicae, v. 79, 368 p.
Hoorn, C., 1993, Marine incursions and the influence of Andean tectonics on
the Miocene depositional history of northwestern Amazonia: Results of
a palynostratigraphic study: Palaeoclimatology, Palaeogeography,
Palaeoecology, v. 105, p. 267–309.
Hoorn, C., 1994, Fluvial palaeoenvironments in the Amazonas Basin (Early
Miocene to early Middle Miocene, Colombia): Palaeoclimatology,
Palaeogeography, Palaeoecology, v. 109, p. 1–54.
Howe, M. W., 1974, Nonmarine Neiva Formation (Pliocene?), Upper Magdalena Valley, Colombia: Regional tectonism: Geological Society of
America Bulletin, v. 85, p. 1031–1042.
Keller, G., and Barron, J. A., 1987, Paleodepth distribution of Neogene
deep-sea hiatuses: Paleoceanography, v. 2, p. 697–713.
Kohn, B. P., Shagam, R., Banks, P. O., and Burkley, L. A., 1984, MesozoicPleistocene fission track ages on rocks of the Venezuelan Andes and
their tectonic implications, in Bonini, W. E., et al., eds., The Caribbean–South American plate boundary and regional tectonics: Geological
Society of America Memoir 162, p. 365–384.
Lundberg, J. G., and Chernoff, B., 1992, A Miocene fossil of the Amazonian
fish Arapaima (Teleostei, Arapaimidae) from the Magdalena River region of Colombia—Biogeographic and evolutionary implications: Biotropica, v. 24, p. 2–14.
Macellari, C. E., 1984, Late Tertiary history of the Táchira depression, southwestern Venezuelan Andes, in Bonini, W. E., et al., eds., The Caribbean–South American plate boundary and regional tectonics: Geological Society of America Memoir 162, p. 333–342.
Maraven S. A., 1993, Cretaceous and Paleogene sedimentation in the southwestern Venezuelan Andes: American Association of Petroleum Geologists Field Guide, 73 p.
Megard, F., 1984, The Andean orogenic period and its major structures in
central and northern Peru: Geological Society of London Journal,
v. 141, p. 893–900.
Miller, T. A., and Etayo-Serna, F., 1972, The geology of the Eastern Cordillera between Aguazul–Sogamoso–Villa de Leiva [geological field
trips, Colombia 1959 –1978]: Colombian Society of Petroleum Geologists and Geophysicists, Margal Lta., p. 349–372.
Nuttall, C. P., 1990, A review of the Tertiary non-marine molluscan faunas
of the Pebasian and other inland basins of northwestern South America:
British Museum of Natural History Bulletin(Geology), v. 45, p. 165–371.
Pardo-Casas, F., and Molnar, P., 1987, Relative motion of the Nazca (Farallon) and South American plates since late Cretaceous time: Tectonics,
v. 6, p. 233–248.
Shagam, R., Kohn, B. P., Banks, P. O., Dasch, L. E., Vargas, R., Rodriguez,
G. I., and Pimentel, N., 1984, Tectonic implications of Cretaceous-Pliocene fission-track ages from rocks of the circum–Maracaibo Basin region of western Venezuela and eastern Colombia, in Bonini, W. E., et
al., eds., The Caribbean–South American plate boundary and regional
tectonics: Geological Society of America Memoir 162, p. 385– 412.
Steinmann, G., 1930, Geologı́a del Perú: Heidelberg, Germany, Carl Winters
Universitatsbuchhandlung, 448 p.
Sykes, L., McCann, W. R., and Kafka, A. L., 1982, Motion of the Caribbean
plate during last 7 million years, and implications for earlier Cenozoic
movements: Journal of Geophysical Research, v. 87, p. 10,656 –10,676.
Van der Hammen, T., 1961, Late Cretaceous and Tertiary stratigraphy and
tectogenesis of the Colombian Andes: Geologie en Mijnbouw, v. 40,
p. 181–188.
Van der Hammen, T., Werner, J. H., and Van Dommelen, H., 1973, Palynological record of the upheaval of the Northern Andes: A study of the
Pliocene and Lower Quaternary of the Colombian Eastern Cordillera
and the early evolution of its High-Andean biota: Review of Palaeobotany and Palynology, v. 16, p. 1–122.
Van der Wiel, A. M., 1991, Uplift and volcanism of the SE Colombian Andes
in relation to Neogene sedimentation in the Upper Magdalena Valley
[Ph.D thesis]: Wageningen, Netherlands, Agricultural University, 208 p.
Van Houten, F. B., and James, H. E., 1984, Late Cenozoic Guayabo delta
complex in SW Maracaibo Basin, in Bonini, W. E., et al., eds., The
Caribbean–South American plate boundary and regional tectonics: Geological Society of America Memoir 162, p. 325–332.
Van Houten, F. B., and Travis, R. B., 1968, Cenozoic deposits, Upper Magdalena Valley, Colombia: American Association of Petroleum Geologists Bulletin, v. 52, p. 675–702.
Wellman, S. S., 1970, Stratigraphy and petrology of the nonmarine Honda
Group (Miocene), Upper Magdalena Valley, Colombia: Geological Society of America Bulletin, v. 81, p. 2353–2374.
Wortel, R., 1984, Spatial and temporal variations in the Andean subduction
zone: Geological Society of London Journal, v. 141, p. 783–791.
Wortel, R., and Cloetingh, S., 1981, On the origin of the Cocos-Nazca
spreading center: Geology, v. 9, p. 425– 430.
Zambrano, E., Vasquez, E., Duval, B., Latreille, M., and Coffinieres, B.,
1971, Sı́ntesis paleogeográfica y petrolera del occidente de Venezuela:
Congreso Geologico Venezolano IV, Proceedings, v. I, p. 483–552.
Manuscript received July 18, 1994
Revised manuscript received November 28, 1994
Manuscript accepted December 6, 1994
Printed in U.S.A.
GEOLOGY, March 1995