Content uploaded by Jesus Rivas
Author content
All content in this area was uploaded by Jesus Rivas on Nov 14, 2020
Content may be subject to copyright.
Received: 26-April-2020 Accepted: 31-October-2020 Pulbished: 13-November-2020
S Y N T H E S I S A N D M E T A -A N A LY S E S
Climate changes and speciation pulses in a nearly
ooded continent: tackling the riddle of South
America’s high diversity
Jesús A. Rivas 1∗
1Department of Biology, New Mexico
Highlands University. Las Vegas, New Mexico.
USA.
Corresponding address
Department of Biology, P.O. Box 9000, New
Mexico Highlands University. Las Vegas, New
Mexico 87701. USA.
Email: rivas@nmhu.edu
Funding
This research did not receive any specic grant
from funding agencies in the public,
commercial, or not-for-prot sectors
Academic Editor:
Andrés Ernesto Ortíz-Rodríguez
ABSTRACT
Understanding the origin of South America’s diversity is of great
importance especially considering our current extinction crisis
in the face of climate change. While great debate exists about
the sources of Amazonian diversity, there is no consensus about
an overarching process that explains the Amazon’s uniquely high
diversity. In this contribution, I present the following model con-
sidering the impact of tectonics on the hydrology of the continent.
When the Andes rose, it dammed the paleo-Amazon, which ran
west at the time. This produced generalized ooding with a mo-
saic of forest in the more elevated areas surrounded by ooded
habitats. Because of the at relief of the Amazon basin, small
changes in water level produced forest expansion and contraction,
resulting in speciation pulses. Using data from the literature on
species distribution, as well as the age of new lineages from mo-
lecular studies, I show that the space and timing of speciation pro-
cess in the South America was consistent with the predictions of
this model. This model also posits that there were no marine in-
cursions on the continent. Rather, marine conditions developed
in situ, might be a better explanation for the marine conditions
found in paleo-history of the region. This model provides a the-
oretical framework for evolutionary processes in South America
that explains its uniquely high diversity.
KEYWORDS
Amazon, biodiversity, macroevolution, paleo-ecology, speciation.
INTRODUCTION
As the World heads deeper into the most recent extinction cri-
sis, in the face of climate change, understanding the sources
of biodiversity becomes all the more important. The global
trend of higher species in lower latitudes has been well es-
tablished for many years (Humboldt, 1850; Pianka, 1977; An-
tonelli et al., 2018). A lot of the debate has focused on whether
tropical areas are cradles of diversity that have more species be-
cause they have higher speciation rate; or museums, that have
How to cite this article: Rivas JA. 2020. Climate changes and speciation pulses in a nearly ooded continent: tackling the riddle of South Amer ica’s high
diversity. Ecotrópicos. 32: e0014
Ecotrópicos 2020. 32 saber.ula.ve/ecotropicos Sociedad Venezolana de Ecología 1
2RI VAS
higher diversity because they have a lower extinction rates, or a
combination of both (Fjeldså, 1994; Voelker et al., 2013; Weir,
2006). Within the tropics the Amazon basin stands out as the
most diverse terrestrial ecosystem in the world; with greater di-
versity of organisms across taxa than any other tropical region
(Haer, 2008; Rull, 2011, 2015). The causes of this biodiver-
sity have been the matter of persisting debate, but to the present
there is no consensus on its origin.
For years the Refugia Hypothesis was viewed as the sole
explanation for tropical high diversity. It postulates that du-
ring Pleistocene glaciations rainforest contracted to isolated
patches surrounded by savannah (Haer, 1969; Prance, 1982).
When warmer periods returned, the forest expanded again, re-
connecting all the refuges. Populations separated during the fo-
rest contraction events had accumulated enough mutations that
they were no longer able to interbreed, resulting in new species.
These vicariant events associated with the contraction and ex-
pansion of the forest produced speciation pulses that resulted in
high diversity (Duellman, 1982; Haer, 1982). However, the
Refugia Hypothesis has been discredited lately. Recent pollen
studies show the permanence of pollen transported by insects
or other animals, not wind, and suggesting that the whole area
has been covered by forest since the Miocene (Bush & Oliveira,
2006; Colinvaux & De Oliveira, 2001; Colinvaux et al., 2000).
Furthermore, molecular studies show that a lot of the specia-
tion processes that are responsible for the high diversity occu-
rred during the Miocene before the Pleistocene glaciations took
place (Moritz et al., 2000; Rull, 2008, 2011), and, recent ev-
idence based on the distribution of forest specialists shows
that forest patches expanded, not contracted, during the Last
Glacial Maximum (Leite et al., 2016). Combined, these lines
of evidence debunked the Refugia Hypothesis.
Current notions about speciation processes in the Amazon
are mostly taxa specic and invoke several processes (Haer,
2008) such as neutrality (De Aguiar et al., 2009; Hubbell, 2001;
Latimer et al., 2001), predation (Leigh Jr et al., 2004), hetero-
geneity of sediments (Fine et al., 2005), and the river eect
(Hayes & Sewlal, 2004). The River eect has been advoca-
ted as a strong source of diversity by creating and separating
land masses which allow allopatric speciation between die-
rent habitats (Haer, 1992). Data shows that the river eect
helps to create and maintain diversity in some areas but the evi-
dence is ambiguous in others (Jackson & Austin, 2010; Voelker
et al., 2013; Naka & Brumeld, 2018). Changes of riverbeds
during recent times might confound phylogenetic trends and
the barriers that caused them, as some diversity trends might
be responses to previous barriers. In the Guyana shield where
rivers are hard set, the river hypothesis holds better than in the
Amazon Basin where the riverbeds are loosely dened (Naka
& Brumeld, 2018) (Table 1).
While the processes described above are likely important
in regional specic speciation processes, there is no consen-
sus about an overarching process as the main driver for the
Amazon’s high diversity (Bush, 1994; Haer & Prance, 2001;
Haer, 2008; Hayes & Sewlal, 2004; Rull, 2011, 2015). More
importantly, there is still a big piece of the puzzle missing.
All these processes apply equally for all tropical regions. Yet,
South America stands alone in terms of their species diver-
sity compared to other tropical landmasses. A recent study
on bird diversication (Claramunt & Cracraft, 2015) provides
data on lineages and endemic species present in dierent bio-
geographic regions. Figure 1, made with data from the supple-
mental material of this study, shows how most tropical biomes
have a comparable number of lineages (families), but when it
comes to endemic taxa, South America is in a league of its
own. The higher number of endemic families and endemic
species found in South America suggests that South America
have had higher levels of speciation for a relatively long time.
We would nd a similar picture if we were dealing with other
taxa such as freshwater shes (Albert et al., 2011), amphibians
(Santos et al., 2009; IUCN, 2017), bats (Lim, 2007; Teeling
et al., 2005) (Armstrong et al., 2014). Thus, an explanation for
the high diversity of South America that is unique for the conti-
nent is still lacking. In the following section I propose a hypo-
thetical model aimed to explain South America’s unique high
diversity by taking into consideration the eect of tectonics and
hydrography in the biogeography of the continent, which has
been largely ignored.
The Big Dam (and Big Swamp) Hypothesis.- When South
America separated from Gondwana (approx. 120–112 Mya),
it was mostly drained by a large river very much where the
Amazon River is currently located, but it ran west into the Pa-
cic Ocean, the paleo-Amazon. As South America collided
with the Nazca plate, subduction of the latter initiated the rise
SOUTH AMERICA’S DIVERSITY 3
T A B L E 1 Summary of dierent hypothesis explaining the tropics high diversity.
Hypothesis Rationale Unique to
South America?
Reference
Refugia
Forest expanded and contracted during the
Pleistocene in response to climate
uctuations producing speciation pulses
No Haer (1969)
River Rivers act as vicariant barrier producing
allopatric and peripatric speciation No
Hayes & Sewlal (2004); Ribas
et al. (2012); Crouch et al.
(2019)
Predation
Predators produce Paine eect preventing
prey species to outcompete other, thus
keeping high diversity
No Terborgh (1992); Leigh Jr et al.
(2004)
Neutrality
All individuals are equivalent to each other
so species are not competing or excluding
each other
No Hubbell (2001); Latimer et al.
(2001); De Aguiar et al. (2009)
Interspecic
competition
Competition leads to specialization over
evolutionary time which allows for higher
species packing
No Terborgh & Weske (1975);
Terborgh et al. (1990)
Intermediate
disturbance
Regular intermediate disturbances reset
competition among species preventing one
from excluding the others
No
Molino & Sabatier (2001);
Roxburgh et al. (2004); Crouch
et al. (2019)
Productivity More resources allows for more individuals
which increase the changes of more species No Schoener (1971); Janzen
(1976); Sayer et al. (2010)
Big Dam
The raise of the Andes damned the
paleo-Amazon ooding the continent.
Small climate uctuations produced
contraction and expansion of habitats
leading to multitude of vicariant events
going back until the early Miocene other
Yes This contribution
of the Andes, blocking the westward ow of the Paleo-Amazon
and closing o its connection with the Pacic Ocean (La-
trubesse et al., 2010; Lundberg et al., 1998). Continued sub-
sidence of the Nazca plate produced ridges of exural fore-
bulges, extending east from the Andes that would have resulted
in higher ridges separated by lower depressions between them
(Bicudo et al., 2019). The original paleo-Amazon drained
most of the area currently drained by the Amazon and Orinoco
Rivers. Since this was a very warm period, it is expected that
the continent was at least as moist as it is now; so, this river
must have been a true colossus. It follows that raising a 7,000
km long dam to this river would result in a nearly total conti-
nental ooding. Up to the present models about South Ame-
rica paleo-ecology seem to ignore the likely ooding that this
spectacular volume of water would have done on the continent.
The foundations of the Big Dam Hypothesis (hereafter BDH)
consider existing models but include the eect that the water
of the dammed river would have had on vicariant events both
4RI VAS
F I G U R E 1 (A) Number of endemic linages (families/subfamilies) present in the Tropical biomes across the world. (B)
Number of endemic species present in dierent tropical regions. Data were obtained from Claramunt & Cracraft (2015) supple-
mental information by extracting those lineages that only occur in one of these biogeographic areas. The number of endemic
taxa is a consequence of the intrinsic diversication rate of each biogeographic unit. South America number of endemic species
is one order of magnitude higher than their paleo-tropical counter parts. The fact that the number of endemic lineages is also
higher than the others suggests that the high diversication rate of South America have been in place for a relatively long time.
The data for this gure is provided in the supplementary material.
in land and water.
The damming of the paleo-Amazon, that is estimated to
have started some 90 million years ago, would have been a very
gradual process (Hoorn et al., 1995; Lundberg et al., 1998; La-
trubesse et al., 2010). It would have begun with the mouth be-
coming progressively shallower, which would have resulted in
the river backing up and ooding its ood plains initially. As
the Andes continued to rise, it would have increasingly limited
the drainage of the river, and its waters would have ooded
most of the lowlands of the western part of the continent un-
til the water level was high enough to spill into the ocean
in whichever direction was lower. Data shows that at least
part of it spilled to the north via the Llanos/Magdalena basin
and drained into the Caribbean via what is currently northern
Venezuela and northeastern Colombia. Some of it may have
also drained over the lowlands between the Southern Andes
and the Brazilian Shield via the Madre de Dios and current
Pantanal (Mora et al., 2010; Wesselingh & Hoorn, 2011).
The ow of this river was perhaps similar to the current
Pilcomayo in Bañado La Estrella, northern Argentina; where
the river runs slowly in a very broad, shallow path without a
clear channel (Díaz de Gamero, 1996). This big river, that
has been called paleo-Amazonas-Orinoco system (Lundberg
et al., 1998) for its northern direction, would have produced
abundant marshes on its sides, scattering the landscape with
lacustrine systems surrounded by dierent kinds of rainforest
depending on its elevation. Higher elevations, as in the ridges
of the exural beluges, piedmonts, and river banks, would have
SOUTH AMERICA’S DIVERSITY 5
had Terra Firme forest and other areas with lower elevations
would have had ooded forest of lower c anopy height such as
Várzea or Igapós, even if the oristic c ompositions of these
types of forest had been quite dierent f rom their contempo-
rary equivalents (Pires & Prance, 1985; Daly & Mitchell, 2000;
Cordeiro-Bicudo et al., 2019). These are dierent k inds of
“rainforest” with dierences i n level o f ooding, wa ter qual-
ity, canopy height, and understory composition. Areas that
received nutrient rich water from the Andes became “white
water” systems including Várzea forest; while waters from the
deeply weathered and highly leached soils of the Guyana and
Brazilian shields would have been far more oligotrophic (Von-
hof et al., 2003; Wesselingh & Hoorn, 2011). These waters
would have produced “clear water” bodies if they were running,
but in lacustrine systems or in ooded Igapó forest, the decom-
position of leaves released tannins that turned them into acidic
“black water” systems (Prance, 1979; Daly & Mitchell, 2000).
Blocking the big river would have also produced other areas
of permanently ooded marshes or swamps where tree growth
was impeded due to the ooding r egime (Prance, 1 979). In
fact, there are evidence of swamps, grasses, and extensive tree-
less areas in the Amazon' paleo-history (Latrubesse et al., 2010;
Hoorn et al., 2017). Perhaps this system was similar to what
we find currently in the Llanos or Pantanal (Mittermeier et al.,
2002; Rivas et al., 2002) but in a much large spatial scale.
In the extremely at relief of the basin (ca. 8.3 x10−3 per-
cent), a small vertical dierence in the water level represents a
very large horizontal dierence in the ooding area. So, a rela-
tively small increase in precipitation wouldproduce substantial
vicariant events connecting ooded habitats and isolating Ter ra
Firme. An equivalent drop in precipitation would reconnect
Terra rme, and isolated aquatic habitats, producing speciation
pulses in both systems. These would have been pulses of ood-
ing equivalent to the current yearly ooding pulses of the Ama-
zon (Erwin & Adis, 1982) but these would have been in a larger
temporal and spatial scale. Due to their larger spatial scale it
would have resulted in genetic separation of populations; and
because of its larger temporal scale, it would have resulted in
enough time for populations to start speciation process. Basi-
cally, the BDH ips the Refugia Hypothesis on its head. The
refuges were not ecological islands of forest surrounded by sa-
vannah, but islands of Terra rme surrounded by a mosaic of
other more ooded habitats.
The conditions described above likely remained for seve-
ral million years. There may have been at least semi-ooded
conditions dating back to the Late Cretaceous as the outlet of
the Paleo-Amazon became shallower and could not drain all
its water. Approximately 24 Mya, the continuous uplifting of
the Andes created a mega-wetland that developed in Western
Amazonas, the Pebas system. This mega wetland is believed
to have been ooded because of cratonic deformation. As the
Andes rose the Cratonic Basin deformed producing elevations
below sea level to the east that would have allowed water from
the Caribe to progress into the continent (Hoorn et al., 2010b).
However, just at the beginning of the Pebas system is the time
when the connection with of the great river and the Pacic
Ocean would have been severed. So, the ooding of the Pebas
system can also be explained with the water from the mighty
river. This explanation is more parsimonious than the ocean
transgression because (1) it does not require such a drastic
bending of the South American Craton to allow the ocean to in-
vade thousands of kilometers into the continent; (2) the current
explanation assumes that the continent was dry (not ooded).
Otherwise the ocean could not have made any progress into the
continent because of the dammed river would have prevented
the ocean to come even if the craton was deformed below sea
level. Likely, it would have resulted in some salinity in the
mouth of the estuary such as they are found currently in Mara-
caibo Lake in Norther eastern Venezuela, but not marine condi-
tions (Vonhof et al., 2003). So the notion that the ocean came
into a dry continent because of cratonic deformation assumes
the, unlikely, disappearance of a spectacularly large volume of
water. In fact, studies from isotopes from mollusk shells shows
that the water of this system was essentially fresh water (Von-
hof et al., 2003).
Flooding regime in the area would have become more in-
tense as the Andes reached higher elevation. All this time
South America continued receiving moisture through Atlantic
trade-winds but the air masses that left the continent, after rais-
ing pass the Andes, were dry. This resulted in South Ame-
rica sequestering large amounts of global moisture (Mörner
et al., 2016). As the Andes rose, the amount of sediments in
the Pebas system increased, lling it up. The mostly-ooded
wetlands would have slowly transitioned toward a less-ooded
ecosystem as sediments from the Andes lled up the Pebas sys-
tem, creating the Acre system 10 Mya. This eventually lled
6RI VAS
up too; leading to the current drainage direction of the Amazon,
which began around 7 Mya (Hoorn et al., 2010a). Even after
the Amazon started owing east, there likely remained many
extensions of virtually 0 slope that still produced forest con-
traction and expansion with small dierences in precipitation
or regional climatic uctuations (Strecker et al., 2007; Mora
et al., 2010; Vonhof & Kaandorp, 2010). So, this system of a
semi-ooded continent would have experienced forest contrac-
tion and expansion with relatively small uctuation in precipi-
tation, starting probably early in the Cenozoic and lasting until
relatively recently.
Unlike the Refugia Hypothesis, the described model here
does not require a full-blown glaciation to produce speciation
pulses. Rather, much smaller changes like, say, Milankovitch
cycles, would have suced. In fact, local and regional changes
in precipitation on the continent could result in local expansion
and contraction of forest without the need of a world-wide cli-
mate change (Sepulchre et al., 2010). There is no need of an
arid ecosystem or savannah-like habitat to develop to produce
vicariant events, so it is compatible with the paleolimnological
evidence of presence of forest throughout the continent. Be-
cause the initial damming of the river may have started as early
as the Late Cretaceous, the BDH can explain diversication
processes that occurred much farther back than the Pleistocene.
It is also explains why Terra Firme actually expanded during
glacial/dry periods (Leite et al., 2016).
Furthermore, species pulses were not limited to dry land.
Movement of aquatic animals among these ooded systems
was probably prevented due to the large dierence in water
quality between the dierent ooded habitats (Ferreira et al.,
2010). Even small variations in precipitation patterns within
the basin would have produced a predominance of white water
systems at times when the Andes provided most of the water
(Sepulchre et al., 2010). At other times, the older formations
could have provided black water pulses, disrupting and frag-
menting white water ecosystems and producing species pulses
among aquatic organisms.
There have been several reports of marine transgressions
from the Caribbean into the western Amazon, yet there is con-
siderable debate in how and where these marine incursions
took place (Díaz de Gamero, 1996; Wesselingh & Salo, 2006;
Hovikoski et al., 2010; Jaramillo et al., 2017). It has been assu-
med that the path was via northwestern Venezuela in what is
currently Zulia and Falcon states. However, Díaz de Gamero
(1996), conclusively debunks this hypothesis based on the na-
ture of the sediments and the volume of the ancient river. The
BDH posits that these ocean transgressions never occurred. A
very large river draining into the Caribbean would have pre-
vented seawater from getting into the continent at all (Díaz de
Gamero, 1996). Whatever world-wide meteorological con-
ditions increased sea levels substantially would have also in-
creased proportionally the amount of water on the continent
making the volume of the paleo-Amazon-Orinoco all the larger.
Whether for ocean level’s rise, or for cratonic deformation, the
river’s fresh water would have accumulated, matching the sea
level and keeping the ocean at bay. This is not that dierent
than what happens today in the deltas of big rivers, such as the
Orinoco or Amazon, with the tide cycle (Archer, 2005). When
the ocean tide rises, it dams the river and makes it go up, but
the marine water does not make any substantial progress into
the continent.
Although marine incursions are a big part of South Ame-
rica Paleogeography, there has never been any evidence of such
marine incursions. Rather, marine incursions have been assu-
med due to the presence of marine conditions such as the pre-
sence of foraminifera and mollusk shells, pollen from man-
groves, and a variety of shes from marine or estuarine linea-
ges (Hoorn, 2006; Roddaz et al., 2006; Wesselingh & Salo,
2006; Wesselingh et al., 2006; Boonstra et al., 2015). Yet,
these are only evidence of “marine conditions”; these marine
conditions could have developed in situ.
Clearly the only way to explain marine conditions in a dry
continent is by assuming an unlikely marine transgression of
thousands of kilometers into the continent. However, if the con-
tinent was ooded marine conditions can appear very readily.
Dry periods or regional weather uctuations could result in
freshwater bodies turning haline. They did not have to be dry
enough to produce arid vegetation on the continent, just dry
enough that the big river had such low water that the wetlands
became fragmented and some parts of the estuary stopped ow-
ing, and voiding sediments into the Caribbean (Vonhof et al.,
2003). The uvial system would transition, at least in some
places, to lacustrine at rst, and eventually its water would have
become haline enough to allow the development of marine or-
ganisms from estuarine stocks. It would have been some sort
of mosaic of conditions where larger water bodies, or those
SOUTH AMERICA’S DIVERSITY 7
with good input of water from higher elevations, remained re-
latively fresh, but shallower extensions of the estuary, or lower
elevation that did not drain into the other water bodies, would
have developed fully marine conditions. There wouldhave also
been all the intermediate stages scattered across the landscape.
Areas that received waters from the nutrient rich Andes would
have become haline faster than those receiving water from
the oligotrophic Guyana and Brazilian shields. In fact, esta-
blishment of haline conditions of estuaries during dry periods
occurs today in tropical estuaries (Valle-Levinson & Bosley,
2003; Carrillo et al., 2009), except that this was a much larger
spatial and temporal scale.
In this contribution, I make predictions based on the BDH
as a framework for the evolution of South American diversity
and test them using information from the literature on the geolo-
gical events of the continent and the appearance of new species
in dierent crown lineages in space and time. I revise the evi-
dence for marine incursions and provide explanations based on
the BDH that explain the data as well, or better, than the exist-
ing explanations. The BDH does not attempt to replace other
hypothesis that explain high diversity, but rather to provide a
historical framework where the dierent hypotheses work in
synergy to explain the high diversity. While the evidence pre-
sented here is largely correlational, the goal of this study is to
contribute with a new paleo-ecological scenario that has not
yet being considered and that has the potential to explain a lot
of the questions that remain unanswered.
METHODS
Testing the Big Dam Hypothesis.- The BDH posits that South
America was subjected to variable levels of ooding through-
out the Cenozoic. Changes in ooding regime would have re-
sulted in expansion and contraction of both forest islands and
aquatic ecosystems resulting in vicariant events under both
conditions. Based on this scenario, I make the following
testable predictions: (1) endemism centers of terrestrial taxa
would be found at higher elevations; (2) speciation processes
across taxa did not occur at random through the history of
South America. Rather, several clades would have speciation
bouts at the same time while there would be periods were no, or
few, new lineages emerged. Because the BDH suggests that the
ooding of the continent would have prevented marine incur-
sions and that the Marine Derived Lineages (MDLs) found in
South America descend from estuarine stock that were trapped
following disconnection from the ocean, (3) no strictly marine
or steno-haline fossils should occur in South America. In the
following sections I attempt to provide evidence supporting the
BDH model and the posed predictions using data from the li-
terature.
Speciation in Space.- Using the current distribution of birds
(mostly Terra Firme specialists) and current precipitation
patterns, Haer (1969) speculated about the location of the
Pleistocene forest refuges that kept forest while the rest of
the continent became semiarid. However, we now know that
South America did not undergo the dry periods that the Refu-
gia Hypothesis required (Bush & Oliveira, 2006; Colinvaux
& De Oliveira, 2001). Under the BDH, refuges were Terra
Firme islands that did not ood in periods when the water
level rose. So, the BDH predicts that these refuges identied
by Haer were associated with higher elevations compared to
their surroundings.
Speciation in Time.- To assess the origin of species over time,
I used the divergence dates of new lineages from existing mo-
lecular phylogenies. Since I only consider extant lineages this
analysis does not take into consideration changes in extinction
rates. Rather it focuses on times when the number of extant li-
neages increased vs times when it did not (Morlon, 2014). The
number of lineages could increase because there were bursts of
new lineages with constant extinction rates, because there were
bursts of extinctions with constant speciation rates, or a combi-
nation of both. Since these are relatively new lineages and that
the continent has not experienced catastrophic events in the pe-
riod of study, I will assume that the increase of new linages is
the consequence of increased speciation rates under a constant
background of extinction rates.
The BDH predicts that speciation rates increased when the
Andes became a tall mountain range and trapped a lot of the
continental moisture producing speciation pulses with small
climate changes. It also posits that the speciation processes
occurred in cycles across the continent producing vicariant
events that would have aected a variety of taxa. So, new linea-
ges would appear in synchrony in dierent groups. Rull (2008)
gathered samples of crown lineages that were unbiased both
8RI VAS
taxonomically as well as geographically within South Ame-
rica. I graphed the cumulative number of lineages present over
the last 26 My looking for periods where there were increases
in the number of lineages and others where no new lineages
appeared. Furthermore, I used Rull’s dataset to calculate the
expected number of new lineages appearing every 500 Ky. I
divided the number of new lineages in every time period by
the number of exiting linages at the time. This gave me an av-
erage rate of lineages production per 500 Ky over the last 26
My. I used this fraction to calculate the expected number of
new species per time period under the assumption that new li-
neages appear at random, depending only on the number of ex-
isting lineages. I multiplied the average rate of appearance of
new lineages found in the continent, by the number of lineages
present, and compared it with the actual number of new linea-
ges of that time period using a Kolmogorov-Smirnov test. The
BDH predicts that the number of new species in every interval
will be dierent than that predicted by the number of existing
lineages, sometimes higher sometimes lower.
To further test how the conditions under the BDH model
might have aected the appearance of new lineages, I used mo-
lecular studies from a variety of taxa to explore the synchro-
nicity of speciation events among Sigmodontine mice (Parada
et al., 2013), army ants (Brady, 2003), ovenbirds (Furnariidae)
(Irestedt et al., 2009), Attini ants (Ward et al., 2015), primates
(Perelman et al., 2011), Conopophagidae passerines (Batalha-
Filho et al., 2014), basal snakes (Colston et al., 2013), ar-
mored catsh (Loricaridae)(Silva et al., 2016), and piranhas
(Serrasalmidae) (Thompson et al., 2014). Also, to test the
eect of the inuence of the Andes height on speciation rates,
I calculated a partial correlation between the ages of new li-
neages over time and the elevation of the Andes at that time
(Garzione et al., 2008). If the elevation of the Andes had an
independent eect on speciation, the BDH would predict an in-
crease in the appearance of new lineages, as the Andes became
a taller mountain range. Because more lineages will likely pro-
duce more species I used a partial correlation to remove the
eect of number of lineages present. Also, to test the synchro-
nicity of the appearance of new lineages I calculated a Spear-
man rank correlation with the time of appearance of new linea-
ges across these taxa. A high correlation value would result if
several taxa producing new species, or not, in synchrony. Sta-
tistical analysis were done with SPSS (IBM Corp., 2011). The
data used in all these calculations are included as supplemen-
tary material.
Marine Incursions vs. in situ Marine Conditions.- The BDH
posits marine incursions in the continent were unlikely. Rather,
marine conditions developed in situ in some of the internal
lakes and swamps of the Big Dam associated with dry spells
or other changes in meteorological conditions. Internal marine
conditions would have developed during drier periods asso-
ciated with eustatic drops in sea level when the Big Swamp
stopped voiding sediments into the ocean. This could have
occurred associated with cooling periods when the planet’s wa-
ter was sequestered in glaciers. Alternatively, regional climate
conditions could have produced localized droughts in South
America independent of major world-wide conditions, or other-
wise provided conditions for haline environments to develop
in internal waters. True ocean incursions are supposed to be
worldwide events associated with warmer periods when the
icecaps melted providing the water for worldwide rising of
the oceans. I looked at global reconstruction of the world’s
paleo-climate (Zachos et al., 2001, 2008) and eustatic sea le-
vels (Hansen et al., 2013) seeking correlation between tempera-
ture uctuations, sea levels, and marine conditions on the con-
tinent. The prediction during a true marine incursion is that the
temperatures will be higher with higher sea levels worldwide.
The prediction of marine conditions developed in situ is that
they were associated with cooling periods or periods of lower
sea level.
RES U LT S
Recent studies on South America biodiversity sustain that the
distribution of variety of taxa, is strong indication of their
origin (Honorio Coronado et al., 2015; Guedes et al., 2018;
Crouch et al., 2019). The avian refugia identified by Haffer
(1969) were in fact places of higher elevation. The Figure 2
shows these refuges on an elevation map, and it is apparent
that they would have maintained Terra firme forest in times of
flooding. Simulation of different elevations of South America
over the last few million years also supports the notion that
Haffer’s refuges were, in fact, higher elevation areas
(Cordeiro-Bicudo et al., 2019).
The record of new crown lineages appearing on the conti-
nent from Rull (2008) is presented in Figure 3A. It is possible
to see a ladder-like pattern with periods in which the number
SOUTH AMERICA’S DIVERSITY 9
FIGURE 2 Elevation of Haer (1969) proposed bird Pleistocene refuges on an elevation map. Numbers correspond to
Haer’s original numbering of Pleistocene refuges: Imerí Refuge (4), Napo Refuge (5), East Peruvian Refuge (6), Madeira-
Tapajós Refuge (7), Belen Refuge (8), and Guaianan Refuge (9). Map by J. Zebrowski.
of lineages did not change and brief periods when there was a
jump in the number of lineages overtime. This discontinuity of
the cumulative number of lineages suggests that new lineages
appeared in pulses throughout the Miocene. This gure also
shows a sharp increase in the last 6 My coincident with the last
uplift of the Andes (Gregory-Wodzicki, 2000), when a radi-
cal change in the drainage of the continent took place (Hoorn
et al., 2010a). Using the same dataset, Figure 3B shows a his-
togram with new lineages appearing and the predicted num-
ber of lineages expected based on the number of existing linea-
ges. It is clear that there are burst of new lineages alternated by
pauses in the appearance of new lineages, which signicantly
diers from what would be expected by chance (Kolmogorov-
Smirnov, Z = 2.25, P< 0.001).
After removing the eect of the cumulative number of
species with a partial correlation test, the appearance of new
species is positively correlated with the height of the Andes
more than expected by chance (β = 0.24, n = 66, P= 0.05,
2-tailed test), suggesting an independent eect of the Andes’
height on speciation processes. A Spearman Rank correlation
suggests synchrony in the appearance of species across a va-
riety of taxa from the literature (Table 2). A positive correla-
tion between two lineages means that those lineages had pulses
with new species and pauses of speciation at the same time,
supporting the BDH. This pattern seems clearer among ecto-
therm vertebrates and shes, but not so much among ants and
basal snakes. Figure 4 shows the appearance of new species
among those endotherms with high mobility (birds and mon-
keys). Here a saw-like pattern is apparent in the emergence of
new lineages across a variety of taxa since the early Cenozoic
10 RI VAS
F I G U R E 3 (A) Cumulative frequency of crown species development in the Amazon across a unbiased sampling of taxa
from Rull (2008). The gure shows periods of stasis where no new lineages appeared punctuated by jumps corresponding with
speciation pulses. These jumps are consistent with the predictions of the BDH. (B) Histogram of new lineages appearing in the
last 26 My and the expected number of new lineages under the assumption of a uniform speciation rate based on the number of
existing lineages. Data from Rull (2008).
with periods when new lineages are formed is alternated with
periods where few, or no, lineages appeared. The positive co-
rrelations found constitute a statistical support of this pattern
(Table 2). Lastly, Figure 6 shows global temperature from ben-
thic deposits and indicates known glacial periods, the eustatic
sea level, as well as the reported marine conditions in the Ama-
zon basin. Marine conditions were not associated with warm-
ing periods during the Miocene or by raising sea levels, but
rather the opposite. A clear trend of dropping temperatures
prevails before and during marine conditions, which contra-
dicts the notion that they resulted from marine incursions and
supports a scenario in which the marine conditions developed
in situ during dry periods.
DISCUSSION
A model of a mostly ooded continent with a mosaic of con-
tracting and expanding habitats provides the conditions for
abundant vicariant events across taxa and across most of the
continent that may account for South America’s high diversity.
Since this process occurred only in South America, we have
a system unique to South America that enhance the action of
all known speciation processes which explains the muchhigher
diversity found in South America compared with other tropical
SOUTH AMERICA’S DIVERSITY 11
T A B L A 2 Spear man rank correlation of the appearance of new species among varietyof South American lineages through the
Cenozoic. The data from these correlation was obtained from Sigmodontine mice (Parada et al., 2013), Army ants (Brady, 2003),
Oven birds (Furnariidae) (Irestedt et al., 2009), Attini ants (Ward et al., 2015), Primates (Perelman et al., 2011), Conopophagidae
passerines (Batalha-Filho et al., 2014), basal snakes (Colston et al., 2013), armored catsh (Loricaridae) (Silva et al., 2016), and
piranhas (Serrasalmidae) (Thompson et al., 2014). The dataset is provided in the supplementary material.*Denotes signicance
at 0.05 level, two-tails. ** Denotes signicance at 0.01 level, two tails.
(1) (2) (3) (4) (5) (6) (7) (8)
(1) Sigmodontine 1.000
(2) Dorylinae -0.233 1.000
(3) Conophaga 0.734** -0.133 1.000
(4) Furnaridae 0.599** -0.257* 0.474** 1.000
(5) Attini -0.054 0.260* -0.110 0.161 1.000
(6) Boidae 0.039 0.100 0.088 0.021 -0.028 1.000
(7) Platyrrhines 0.728** -0.213 0.710** 0.715** 0.078 -0.045 1.000
(8) Loricaridae 0.678** -0.114 0.542** 0.694** 0.098 -0.033 0.740** 1.000
(9) Serrasalmidae 0.422** 0.058 0.528** 0.425** 0.135 0.008 0.385** 0.480**
continents (Fig. 1).
The fact that the appearance of new lineages is correlated
with the height of the Andes, after removing the eect of the
lineages present at the time, suggests that its high elevation is
associated with the increase of diversity. The emergence of
the Andes resulted in orogenic precipitation that may have in-
creased ooding conditions (see below) and its eect on gene-
rating species pulses. Since the trade wins continued to bring
moisture in to the continent but the high elevation of the Andes
prevented this moisture from leaving the continent, the results
would have been an increased in ooding as the Andes became
taller; turning the continent into a sink of global water (Mörner
et al., 2016). Approximately 7 Mya, the Amazon acquired its
current direction (Lundberg et al., 1998; Hoorn et al., 2010b).
These would have produced substantial ooding and rearrang-
ing of new habitats, which would have increased the number of
vicariant events in both terrestrial and aquatic systems. Consis-
tent with the BDH, at this time the continent underwent a veri-
table explosion of new lineages across the continent spanning
thousands of taxa and lasting until recent times (Fig. 3).
While the rise of the Andes would have also created a di-
versity of habitats along its altitudinal cline that would con-
tribute to the increase of diversity (Ribas et al., 2007), this
eect would have been additive to the increase of vicariant
events associated with the increase in ooding conditions. The
inuence of these two eects cannot be disentangled with the
evidence presented here, but a mountain range’s emergence
producing new habitats associated with altitudinal clines has
occurred in many other regions. Yet, only in the Andes did
the rise of a mountain range produce the increase in diversity
found in South America, suggesting that it is not only about the
creation of new habitats or its environmental heterogeneity.
The appearance of new species seems to have a syn-
chronous pattern —certainly in the modest number of taxa con-
sidered in this study (Figs. 3 and 4). The positive correlation
found among the appearance of new taxa from dierent linea-
ges is consistent with the idea that there were continent-wide
(or at least region-wide) vicariant events that aected a variety
of taxa. Contraction and expansion of habitats with small cli-
matic uctuations would explain the pattern found. It is in-
triguing that there is a positive correlation between of pulses
speciation among endothermic with high mobility and aquatic
taxa of vertebrates, but it does not seem to be present among
ants or basal snakes. A possible explanation for this is that both
12 RI VAS
F I G U R E 4 Number of new taxa appearance in the last 40 my in South America. Marine incursions alleged to have occurred
(dark bars) (Lundberg et al., 1998; Hovikoski et al., 2010; Bloom & Lovejoy, 2011), presence of Antarctic ice sheet (grey bars),
benthic oceanic temperature (grey line) obtained from (Zachos et al., 2001, 2008), and eustatic sea levels divided by 10 for
scaling (dark line) (Hansen et al., 2013). Notice pulsating nature of the new taxa giving the graph a saw-like appearance.
birds and mammals have larger home ranges than snakes and
ants Clearly birds can easily disperse over relatively extensive
ooded areas. Recent studies with capuchin monkeys also
show the they can disperse over relatively important rivers
(Lima et al., 2017). So, vicariant events for some of these
taxa with less mobility may not have produced isolation among
those taxa with higher mobility. This explains the lack of cor-
relation among endotherms with high mobility and terrestrial
ectotherms. A limitation of this approach is that the precision
of the molecular clocks used to determine the appearance of
new lineages is often in the millions of years and some of
these vicariant events might have lasted much less than that.
Yet, despite this limitation my results suggest a very strong
trend. Clearly a more comprehensive survey of speciation pro-
cesses across a variety of taxa, controlling for mobility, home
range, and habitat would be needed to better understand this
phenomenon.
Figure 4 shows no support for the notion of true marine
incursions on the continent. The reported marine incursions
did not occur during warming periods or during periods of sea
level rise. Rather, there is a cooling trend throughout the last
few million years that instead supports the notion of a drying
world weather leading to marine conditions developing in situ.
The question remains how marine conditions were corrected
if the planet continued to cool after the marine conditions
changed. There are two non-mutually exclusive explanations
for this. One is that the changes in local climate brought more
precipitation into the Guyanan and Brazilian shields. Their
oligotrophic water would have diluted haline conditions deve-
loped in the basin. Alternatively, the continued rise of the An-
des increased retention of water within the continent by devel-
oping orogenic precipitation (Mörner et al., 2016). Losing less
water to the Pacic Ocean via the trade winds would have di-
luted the marine conditions that had developed. With the con-
tinent now voiding its waters, and sediments, into the Atlan-
tic Ocean, there is an outlet for the minerals responsible for
the haline conditions at the continental level. Yet, salinas and
salt water lagoons exist today in Mato Grosso do Sul in Brazil,
which may well be relicts of these internal seas or perhaps ana-
logous systems to them (Furquim et al., 2010). The occurrence
of saltwater conditions thousands of miles from the ocean pro-
vide denite prove that haline conditions may exist without a
marine incursion and without drastic weather patterns.
Unfortunately, we do not have a good reconstruction of lo-
SOUTH AMERICA’S DIVERSITY 13
cal and regional climatic patterns in South America’s Miocene
to see if the synchronic speciation among highly mobile verte-
brates responds to local ooding events. However, the BDH
allows us to hypothesize when they occurred. Figure 4 has
the record of new species of birds and monkeys and allows for
a concrete prediction. Speciation pulses of terrestrial species
of high mobility should be associated with high water marks
that isolated forest patches far enough to produce vicariant
events among birds and monkeys (peaks of appearances of new
species), while times of low appearance of new lineages were
drier period when forest expanded (Fig. 4). In fact, due to
how much water levels would have changed with a small dif-
ference in precipitation, I would expect the peaks of new lin-
eage pulses to be far shorter and more frequent than what is
implied in Fig. 3. As we develop better paleo-climatic models,
better reconstruction of the paleo-climate of South America,
and more nely timed phylogenies, we will be able to better
explore these questions.
One of the strong evidences of marine transgressions
comes from studies of shes from marine clades that are
present in the Amazon basin (Lovejoy & Albert, 2006). Of
particular interest are Stingrays (Potamotrygonidae), since they
are ubiquitous in the Amazon/Orinoco basin (de Carvalho &
McEachran, 2003). The freshwater lineages show a split about
20 million years ago from their marine counterparts, shortly af-
ter the formation of the Pebas system. The authors argue that
the vicariant event that split the lineages was immigration of
a population of Stingrays during a marine incursion (Lovejoy,
1998). However, it is well known that Elasmobranchs venture
freely into freshwater (de Carvalho & McEachran, 2003). An
alternative explanation is that the vicariant event that split the
lineages was a drought that prevented the Pebas system, at least
partially, from draining into the Caribbean and thus interrupted
gene ow between the continental and marine populations. In
fact, if there was a marine transgression it is not clear why linea-
ges would have split, if it all were ocean water. However, the
isolation of an estuary from the ocean would have produced
a real vicariant event. Figure 4 shows a drop in sea level at
this time, which is not consistent with marine incursions, but
rather a drier period. Furthermore, among the hypothesized
drier periods in Figure 4 based on species pulse of primates
and birds, there is precisely one that starts about 20 Mya when
the “ocean transgression” was supposed to have started. Note
that the prediction of this drier period is based on the speciation
patterns of birds and monkeys, so these are independent lines
of evidence for the presence of a dry period at this time, and
they support the idea that the split of stingrays was due to dryer
weather, rather than marine incursions. Salinity developed in
situ does not necessarily explain better the presence of Marine
Derive Lineages (MDLs) in the continent, but it explains it at
least as well as the current interpretation and thus deserves fur-
ther consideration. In addition, an estuary that connected to the
sea and became disconnected in times of drought explains both
the presence of Caribbean sh in the Amazon basin as well as
Amazonian shes in Miocene deposits of the Caribbean coast
in current Venezuela (Lundberg & Aguilera, 2003).
Further evidence that MDLs present in South America des-
cend from estuarine organisms that got trapped on the conti-
nent during continental droughts or sea level fall is the fact that
the main MDLs found in South America are organisms that
occur regularly in estuaries such as mangrove forest (Kathire-
san & Bingham, 2001), foraminifera (Camacho et al., 2015),
seashells (Lima et al.), elasmobranchs as well as a variety of
euryhaline shes (Hubert & Renno, 2006; Lovejoy & Albert,
2006; Bloom & Lovejoy, 2011). If true ocean incursions in
South America lasted as long as the marine incursions are sup-
posed to have lasted, surely there would be fossils, or descen-
dants, of strictly marine organisms present on the continent.
However, all taxa found in South America are euryhaline or-
ganisms that regularly inhabit brackish and even fresh waters
(Hubert & Renno, 2006; Lovejoy & Albert, 2006; Bloom &
Lovejoy, 2011).
Marine incursions are a corner stone of sh biogeography
in the South America. Saltwater bodies would act as barri-
ers for freshwater shes producing vicariant events leading to
their speciation (Hubert & Renno, 2006). However, these ex-
planations still hold under the BDH. The only dierence is the
source of the marine conditions. Hubert & Renno (2006) pro-
posed freshwater refugia during these marine incursions that,
not unlike Haer’s, were located toward the foothills of the An-
des and the Guyanan and Brazilian shields where the marine
waters could not reach. Under the BDH, the refuges remained
freshwater because they received constant inux of fresh water
from the higher watershed and drained their waters (and sedi-
ments) to the lower elevations, while the depressions in the
low lands of the basin, that were not draining, became haline
14 RI VAS
after accumulating enough sediments. In fact, the scenario of
a system of salinity developing in situ better explains isolated
marine conditions during the Acre system (in the current Acre
state, Brazil) (Lundberg et al., 1998) thousands of kilometers
away from any sea entrance, or other marine conditions.
Furthermore, the notion that salt water developed in situ,
explains the reports of “dilute marine incursions” (Wesselingh
& Salo, 2006) as well as variable salinity reported in the Pe-
bas system (Vonhof et al., 1998; Wesselingh & Salo, 2006;
Voelker et al., 2013; Boonstra et al., 2015; Jaramillo et al.,
2017). A drought that did not last very long would have re-
sulted in brackish water, not a fully marine system. The varia-
tion in salinity in the Pebas system found in foraminifera de-
posits (Boonstra et al., 2015) cannot be explained by an ocean
transgression which would have brought in the same level of
salinity across the area. However, variable contribution of
dierent watersheds, and regional climatic dierences in the
basin, would have accounted for the variation in salinity. In-
put from the nutrient-rich Andes would have become haline
faster than those from the oligotrophic Guyanan and Brazil-
ian shields. This scenario also explains the high diversity of
mollusk in the Pebas system (Wesselingh et al., 2006; Wesse-
lingh & Salo, 2006). Because calcium carbonate from mollusk
shells has very high solubility in the acidic pH of black water,
these waters would have been inhospitable for them. Expan-
sion and contraction of black water bodies via variation in the
relative contribution of each watershed would have produced
speciation pulses contributing to the development of this high
diversity.
Models using palynological signatures of the amazon se-
diments indicates that there was not a continuum rainforest in
South America and they suggest that patches were separated by
wedges of dry forest (Pennington et al., 2000). However, stu-
dies of Net Primary Productivity (NPP) suggest the presence
of highly productive ecosystems throughout which is incom-
patible with dry forest. Debate continues regarding the pre-
sence of large percentage of grasses (Poacea) in the palynolo-
gical record that is often attributed to savannah. However, the
pollen of aquatic grasses cannot be distinguished from dry land
grasses (Mayle et al., 2004). A scenario of large extensions
of marshes and ooded aquatic habitat would account for the
abundance of grasses without dry periods and with high NPP.
This scenario explains all the existing, and seemingly contra-
dictory, evidence.
Taken together, the study suggests that the Andes con-
tributed to South America’s diversity via several mechanisms
beyond what has been argued in terms of increasing habitat
heterogeneity with its altitude, mainly by creating conditions
for potentiated speciation processes. Vicariant events would
be produced with the emergence of the Andes by (1) blocking
the big river, ooding the continent producing dry land and
aquatic systems, (2) retaining moisture from the trade winds
and changing the climate that produced expansion and contrac-
tion of forest and aquatic ecosystems and, as the Andes range
grew (3) large volumes of earthen crust that formerly had been
under the ocean accumulated and the removal of such large vo-
lumes of crust would have contributed to eustatic drops of sea
level (Miller et al., 2005), where the drop in sea level isolated
estuaries and river from the ocean and prevented freshwater
ecosystems from voiding sediments, leading to haline condi-
tions which in turn lead to more aquatic vicariant events.
While other authors have noted the correlation of the rise
of the Andes and the increase of South American diversity,
the mechanisms identied have been limited to biogeographic
barrier, elevation gradients, and associated with nutrient and
soil patchiness (Fine et al., 2005; Ribas et al., 2007, 2009;
Quintero et al., 2013). The impact of these explanations,
though, is limited to few taxa and to the extreme western Ama-
zon. The BDH provides a broader geographical eect that di-
rectly, and causally, explains South America’s high diversity
across taxa. The BDH also explains the west to east diversity
and moisture gradients (Cheng et al., 2013; Crouch et al., 2019;
Oberdor et al., 2019), and west to east dispersal of new linea-
ges (Bonaccorso et al., 2006). Thus, rather than replace exist-
ing explanation for South America’s high diversity, the BDH
provides a framework for the synergy of many of these expla-
nation that is unique to South America.
The BDH also explains some of the questions that still
elude explanation when we compare South America with other
tropical continents. For instance, because it is unique to South
America, it explains the greater Amazonian diversity than in
either Africa or Asia. It also explains why South America is
the only tropical continent without terrestrial primates. Exten-
sions of patches of ooded forest would have been good habi-
tat for tree dwelling primates but not ground dwelling ones. It
explains the higher diversity of fresh water shes in the Ama-
SOUTH AMERICA’S DIVERSITY 15
zon and Orinoco basin (Albert et al., 2011). A large ooded
extension of a river that spills into the Caribbean very slowly
is the perfect scenario for the great diversity of crocodilians
(Langston, 1965; Salas-Gismondi et al., 2015) and turtles (Ca-
dena & Jaramillo, 2015) found in the Llanos/Magdalena basin.
It also provides a scenario for the diversication of Boidae, pro-
ducing strictly arboreal clades such as Corallus spp. as well as
strictly aquatic ones such as Titanoboa spp. and Eunectes spp.
(Noonan & Chippindale, 2006; Head et al., 2009).
Furthermore, having large extensions of low-canopy
ooded forest the likes of Varzea, Igapó, and isolated patches
of Terra Firme explains the paucity of gliders. Gliding evolved
repeatedly and in many taxa in the paleotropics, but there are
surprisingly few of them in South America (Emmons & Gen-
try, 1983); despite its higher diversity across the board. If the
tall trees were limited to a few scattered Terra Firme patches,
gliding would not be very adaptive. Rather, powered ight
would be far more useful. Being able to y between isolated
forest patches containing fruit trees, explains the magnicent
explosive radiation of frugivorous microchiropetra we see in
South America that has no equivalent anywhere else in the
paleo-tropics (Teeling et al., 2005; Lim, 2007).
CONCLUSIONS
Up to the present we have paid very little attention to the pos-
sible hydrographic and biogeographic repercussions of stop-
ping the main drainage of a continent as moist and as at as
South America. While the Big Dam Hypothesis is specula-
tive in nature, and its evidence is largely correlational, it does
provide an internally consistent model for the synergistic ac-
tion of multitude of speciation processes that explains all the
current evidence as well as, or better than, the accepted mod-
els. Most importantly, it takes into consideration the hydro-
logical consequences of damming one of the largest rivers of
the Tertiary, which other models have failed to consider. Be-
cause it applies to a broad variety of taxa across the continent, it
has the potential to provide an overarching process explaining
the South America’s high diversity and thus deserves further
scrutiny. Hopefully future studies will support or falsify the
fundamentals of this hypothesis.
ACKNOWLEDGMENTS
I thank J. Zebrowski for GIS and cartographic help, P. Roa-
Morales for seeding my mind with questions, K. Swing for
giving me a closer look at Tiputini Diversity Station, and J.
Ojasti for encouraging me to shoot for the moon. Thanks to the
comments of two anonymous reviewers I was able to greatly
improve this ms and I sincerely thank them for their thorough-
ness. I also thank J. Hansen for providing climatological data,
and E. Brown for editorial comments on the manuscript and P.
Andreadis for so stubbornly challenging my ideas.
CONFLICTS OF INTEREST
The author declare no conicts of interest.
SUPPLEMENTARY MATERIAL
Appendix S1. Original data and statistical analyses used to de-
velop this study. Information could be directly downloaded
at: http://erevistas.saber.ula.ve/index.php/ecotropicos/article/
view/E0014/21921927720
ABS T R AC T
Cambios climáticos y pulsos de espe-
ciación en un continente semi-inundado:
atacando el misterio de la alta diversidad
de Sur América. Entender el origen de la diversidad
biológica de Sur América es especialmente importante hoy, en
vista a nuestras crisis de extinción mundial de cara al cambio
climático. Aunque ha habido mucho debate sobre el origen
de la alta diversidad de Sur América, no hay consenso sobre
un proceso general que afecta todo el continente y variedad de
taxa. En esta contribución, presento un modelo teórico con-
siderando el impacto del tectonismo e hidrología en la historia
del continente. Cuando los Andes se levantaron, represó el rio
Paleo-Amazonas, que corría hacia el oeste. Esto produjo una
inundación generalizada en todo el continente en donde los
bosques estaban en las áreas más elevadas rodeados de hábi-
tats inundados. Debido al relieve plano de la hoya Amazónica,
pequeños cambios en el nivel de agua hubieran resultado en
expansión y contracción de bosques, produciendo pulsos de
16 RI VAS
especiación. En este estudio analizo datos de la literatura en
distribución de especies, y de la edad de nuevos linajes us-
ando estudios moleculares. Muestro que los procesos de espe-
ciación en tiempo y espacio corresponden con las predicción
es del modelo en el continente. Este modelo también postula
que no hubo incursiones marinas en Sur America y que las
condiciones marinas que se han encontrado en la historia del
continente se hubieran desarrollado in situ.
Palabras clave: Amazonas, biodiversidad, macroevolu-
ción, paleo-ecología, especiación.
REFERENCIAS
Albert JS, Petry P & Reis RE.2011. Major biogeographic and
phylogenetic patterns. Historical biogeography of Neotropical
freshwater shes 1: 21–57.
Antonelli A, Zizka A, Carvalho FA, Scharn R, Bacon CD, Sil-
vestro D & Condamine FL.2018. Amazonia is the primary
source of Neotropical biodiversity. Proceedings of the Na-
tional Academy of Sciences of the United States of America
115: 6034–6039.
Archer AW.2005. Review of Amazonian depositional systems.
In: Blum M, Marriot S & Leclair S (Eds.) Fluvial sedimen-
tology VII., Blackwell Publishing Ltd, Kingston-upon-Thames,
UK, pp. 17–39.
Armstrong KE, Stone GN, Nicholls JA, Valderrama Escallón E,
Anderberg AA, Smedmark J, Gautier L, Naciri Y, Milne R
& Richardson JE.2014. Patterns of diversication amongst
tropical regions compared: A case study in Sapotaceae. Fron-
tiers in Genetics 5: 116–129.
Batalha-Filho H, Pessoa RO, Fabre PH, Fjeldså J, Irestedt
M, Ericson PG, Silveira LF & Miyaki CY.2014. Phy-
logeny and historical biogeography of gnateaters (Passeri-
formes, Conopophagidae) in the South America forests. Mole-
cular Phylogenetics and Evolution 79: 422–432.
Bicudo TC, Sacek V, de Almeida RP, Bates JM & Ribas CC.
2019. Andean tectonics and Mantle Dynamics as a Pervasive
Inuence on Amazonian ecosystem. Scientic reports 9: 1–11.
Bloom DD & Lovejoy NR.2011. The Biogeography of Ma-
rine Incursions in South America. In: Albert JS & Reis
RE (Eds.) Historical Biogeography of Neotropical Freshwa-
ter Fishes, University of California Press, Berkley, USA, pp.
137–144.
Bonaccorso E, Koch I & Peterson AT.2006. Pleistocene frag-
mentation of Amazon species’ ranges. Diversity and Distribu-
tions 12: 157–164.
Boonstra M, Ramos M, Lammertsma E, Antoine PO & Hoorn
C.2015. Marine connections of Amazonia: Evidence
from foraminifera and dinoagellate cysts (early to middle
Miocene, Colombia/Peru). Palaeogeography, Palaeoclimato-
logy, Palaeoecology 417: 176–194.
Brady SG.2003. Evolution of the army ant syndrome: the origin
and long-term evolutionary stasis of a complex of behavioral
and reproductive adaptations. Proceedings of the National
Academy of Sciences 100: 6575–6579.
Bush MB.1994. Amazonian Speciation: a necessarily complex
model. Journal of Biogeography 21: 5–17.
Bush MB & Oliveira PE.2006. The rise and fall of the Refu-
gial Hypothesis of Amazonian speciation: a paleoecological
perspective. Biota Neotropica 6: bn00106012006.
Cadena E & Jaramillo C.2015. Early to middle Miocene tur-
tles from the northernmost tip of South America: giant testu-
dinids, chelids, and podocnemidids from the Castilletes Forma-
tion, Colombia. Ameghiniana 52: 188–203.
Camacho S, Moura D, Connor S, Scott D & Boski T.2015. Eco-
logical zonation of benthic foraminifera in the lower Guadiana
Estuary (southeastern Portugal). Marine Micropaleontology
114: 1–18.
Carrillo L, Palacios-Hernández E, Yescas M & Ramírez-
Manguilar AM.2009. Spatial and seasonal patterns of sali-
nity in a large and shallow tropical estuary of the Western
Caribbean. Estuaries and Coasts 32: 906–916.
Cheng H, Sinha A, Cruz FW, Wang X, Edwards RL, D’Horta
FM, Ribas CC, Vuille M, Stott LD & Auler AS.2013. Cli-
mate change patterns in Amazonia and biodiversity. Nature
Communications 4: 1411.
Claramunt S & Cracraft J.2015. A new time tree reveals Earth
history’s imprint on the evolution of modern birds. Science
advances 1: e1501005.
Colinvaux P & De Oliveira P.2001. Amazon plant diversity and
climate through the Cenozoic. Palaeogeography, Palaeoclima-
tology, Palaeoecology 166: 51–63.
Colinvaux PA, Oliveira PE & Bush MB.2000. Amazonian and
neotropical plant communities on glacial time-scales: The fai-
lure of the aridity and refuge hypotheses. Quaternary Science
Reviews 19: 141–169.
Colston TJ, Grazziotin FG, Shepard DB, Vitt LJ, Colli GR,
Henderson RW, Hedges SB, Bonatto S, Zaher H, Noonan
BP et al. 2013. Molecular systematics and historical biogeo-
graphy of tree boas (Corallus spp.). Molecular Phylogenetics
and Evolution 66: 953–959.
SOUTH AMERICA’S DIVERSITY 17
Cordeiro-Bicudo T, Sacek V, de Almeida RP, Bates JM &
Ribas CC.2019. Andean tectonics and Mantle Dynamics as
a Pervasive Inuence on Amazonian ecosystem. Scientic Re-
ports 9: 1–11.
Crouch NM, Capurucho JM, Hackett SJ & Bates JM.2019.
Evaluating the contribution of dispersal to community struc-
ture in Neotropical passerine birds. Ecography 42: 390–399.
Daly DC & Mitchell JD.2000. Lowland vegetation of tropical
South America–an overview. In: Lentz D (Ed.) Imperfect bal-
ance: landscape transformations in the pre-Columbian Ameri-
cas, Columbia University Press, New York, USA, pp. 391–454.
De Aguiar MA, Baranger M, Baptestini EM, Kaufman L &
Bar-Yam Y.2009. Global patterns of speciation and diversity.
Nature 460: 384–387.
de Carvalho MR & McEachran JD.2003. Family Car-
charhinidae (requiem sharks) . In: Reis RE, Kullander SO
& Ferraris CJ (Eds.) Check List of the Freshwater Fishes of
South and Central America, Edipucrs, Porto Alegre, Brazil, pp.
13–16.
Duellman WE.1982. Quaternar y climatic-ecological uctuations
in the lowland tropics: frogs and forests. In: Prance GT (Ed.)
Biological Diversication in the Tropics, Columbia University
Press, New York, USA, pp. 389–402.
Díaz de Gamero MLD.1996. The changing course of the Orinoco
River during the Neogene: a review. Palaeogeography, Palaeo-
climatology, Palaeoecology 123: 385–402.
Emmons LH & Gentry AH.1983. Tropical forest structure
and the distribution of gliding and prehensile-tailed vertebrates.
The American Naturalist 121: 513–524.
Erwin T & Adis J.1982. Amazonian inundation forest: their role
as short term refuges and generators of species richness and
taxon pulses. In: Prance GT (Ed.) Biological Diversication
in the Tropics, Columbia University Press, New York, USA, pp.
358–371.
Ferreira LV, de Almeida SS, Parolin P et al. 2010. Amazonian
white-and blackwater oodplain forests in Brazil: large die-
rences on a small scale. Ecotropica 16: 31–41.
Fine PA, Daly DC & Cameron KM.2005. The contribution of
edaphic heterogeneiyt to the evolution and diversity of bursera-
cear trees in the western Amazon. Evolution 59: 1464–1478.
Fjeldså J.1994. Geographical patterns for relict and young species
of birds in Africa and South America and implications for con-
servation priorities. Biodiversity and Conservation 3: 207–
226.
Furquim SAC, Graham RC, Neto JPQ & Vidal-Torrado P.
2010. Soil mineral genesis and distribution in a saline lake
landscape of the Pantanal Wetland, Brazil. Geoderma 154:
518–528.
Garzione CN, Hoke GD, Libarkin JC, Withers S, MacFadden
B, Eiler J, Ghosh P & Mulch A.2008. Rise of the Andes.
Science 320: 1304–1307.
Gregory-Wodzicki KM.2000. Uplift history of the Central and
Northern Andes: a review. Geological Society of America
Bulletin 112: 1091–1105.
Guedes TB, Sawaya RJ, Zizka A, Laan S, FaurbyS, Pyron RA,
Bernils RS, Jansen M, Passos P, Prudente AL et al. 2018.
Patterns, biases and prospects in the distribution and diversity
of Neotropical snakes. Global Ecology and Biogeography 27:
14–21.
Haer J.1969. Speciation in Amazonian forest birds. Science
131: 131–137.
Haer J.1982. General aspects of the refuge theory. In: Prance
GT (Ed.) Biological diversication in the tropics, Columbia
University Press, New York, USA, pp. 6–24.
Haer J.1992. On the “river eect” in some forest birds of south-
ern Amazonia. Boletin del Museum Para Emilio Goeldi serie
Zoologica 8: 217–245.
Haer J.2008. Hypotheses to explain the origin of species in
Amazonia. Brazilian Journal of Biology 68: 917–947.
Haer J & Prance GT.2001. Climatic forcing of evolution in
Amazonia during the Cenozoic: On the refuge theory of biotic
dierentiation. Amazoniana 16: 579–607.
Hansen J, Sato M, Russell G & Kharecha P.2013. Climate
sensitivity, sea level and atmospheric carbon dioxide. Philo-
sophical Transactions of the Royal Society A: Mathematical,
Physical and Engineering Sciences 371: 20120294.
Hayes F & Sewlal JA.2004. The Amazon River as a dispersal
barrier to passerine birds : eects of river width , habitat and
taxonomy. Journal of Biogeography 31: 1809–1818.
Head JJ, Bloch JI, Hastings AK, Bourque JR, Cadena EA,
Herrera FA, Polly PD & Jaramillo CA.2009. Giant boid
snake from the Palaeocene neotropics reveals hotter past equa-
torial temperatures. Nature 457: 715–717.
Honorio Coronado EN, Dexter KG, Pennington RT, Chave J,
Lewis SL, Alexiades MN, Alvarez E, Alves de Oliveira A,
Amaral IL, Araujo-Murakami A et al. 2015. Phylogenetic
diversity of Amazonian tree communities. Diversity and Dis-
tributions 21: 1295–1307.
18 RI VAS
Hoorn C.2006. Mangrove forests and marine incursions in Neo-
gene Amazonia (lower Apaporis River, Colombia). Palaios 21:
197–209.
Hoorn C, Bogotá-A GR, Romero-Baez M, Lammertsma EI,
Flantua SG, Dantas EL, Dino R, do Carmo DA &
Chemale Jr F.2017. The Amazon at sea: Onset and stages of
the Amazon River from a marine record, with special reference
to Neogene plant turnover in the drainage basin. Global and
Planetary Change 153: 51–65.
Hoorn C, Guerrero J, Sarmiento GA & Lorente MA.1995. An-
dean tectonics as a cause for changing drainage patterns in
Miocene northern South America. Geology 23: 237–240.
Hoorn C, Wesselingh F, Ter Steege H, Bermudez M, Mora A,
Sevink J, Sanmartín I, Sanchez-Meseguer A, Anderson C,
Figueiredo J et al. 2010a. Amazonia through time: Andean
uplift, climate change, landscape evolution, and biodiversity.
Science 330: 927–931.
Hoorn C, Wesselingh FP, Hovikoski J, Guerrero J et al. 2010b.
The development of the amazonian mega-wetland (Miocene;
Brazil, Colombia, Peru, Bolivia). In: Hoorn C & Wesselingh
F(Eds.) Amazonia, landscape and species evolution: a look
into the past, Hoboken: Blackwell-Wiley, West Sussex, UK,
pp. 123–142.
Hovikoski J, Wesselingh FP, Räsänen M, Gingras M & Vonhof
HB.2010. Marine inuence in Amazonia: evidence from the
geological record. In: Hoorn C & Wesselingh F (Eds.) Ama-
zonia, landscape and species evolution: a look into the past,
Hoboken: Blackwell-Wiley, West Sussex, UK, pp. 143–161.
Hubbell SP.2001.The unied neutral theory of biodiversity and
biogeography (MPB-32). Princeton University Press, New Jer-
sey, USA.
Hubert N & Renno JF.2006. Historical biogeography of South
American freshwater shes. Journal of Biogeography 33:
1414–1436.
Humboldt AV.1850.Views of Nature or the contemplation on the
sublime phenomena of creation . Harrison and Sons, London.
IBM Corp.2011. IBM SPSS Statistics for Windows.
Irestedt M, Fjeldså J, Dalén L & Ericson PG.2009. Conver-
gent evolution, habitat shifts and variable diversication rates
in the ovenbird-woodcreeper family (Furnariidae). BMC evo-
lutionary biology 9: 268.
IUCN.2017. The IUCN red list of threatened species. Disponible
en: https://www.iucnredlist.org/. (Consultado el 22 de Mayo
de 2020).
Jackson ND & Austin CC.2010. The combined eects of rivers
and refugia generate extreme cryptic fragmentation within the
common ground skink (Scincella lateralis). Evolution: Inter-
national Journal of Organic Evolution 64: 409–428.
Janzen D.1976. Why are there so many species of insects? In:
Whit D (Ed.) Proceedings of the XV International Congress of
Entomology, The Entomological Society of America, Washing-
ton D.C, USA, pp. 84–94.
Jaramillo C, Romero I, D’Apolito C, Bayona G, Duarte E,
Louwye S, Escobar J, Luque J, Carrillo-Briceño JD, Za-
pata V et al. 2017. Miocene ooding events of western Ama-
zonia. Science Advances 3: e1601693.
Kathiresan K & Bingham BL.2001. Biology of mangroves and
mangrove ecosystems. Advances in Marine Biology 40: 84–
254.
Langston JW.1965. Fossil crocodilians from Colombia and the
Cenozoic history of the Crocodilia in South America. Uni-
versity of California Publications in Geological Sciences 52:
1–157.
Latimer AW, Silander JA & Cowling RM.2001. Neutral eco-
logical theory reveals isolation and rapid speciation in a biodi-
versity hotspot. Science 309: 1722–1725.
Latrubesse EM, Cozzuol M, da Silva-Caminha SA, Rigsby CA,
Absy ML & Jaramillo C.2010. The Late Miocene paleogeo-
graphy of the Amazon Basin and the evolution of the Amazon
River system. Earth-Science Reviews 99: 99–124.
Leigh Jr EG, Davidar P, Dick CW, Terborgh J, Puyravaud JP,
ter Steege H & Wright SJ.2004. Why do tropical forest have
so many species of trees? Biotropica 36: 447–473.
Leite YL, Costa LP, Loss AC, Rocha RG, Batalha-Filho H, Bas-
tos AC, Quaresma VS, Fagundes V, Paresque R, Passamani
Met al. 2016. Neotropical forest expansion during the last
glacial period challenges refuge hypothesis. Proceedings of
the National Academy of Sciences 113: 1008–1013.
Lim BK.2007. Divergence times and origin of neotropical sheath-
tailed bats (Tribe Diclidurini) in South America. Moleculuar
Phylogenetics and Evolution 45: 777–791.
Lima MG, Buckner JC, Silva-Júnior JdSe, Aleixo A, Martins
AB, Boubli JP, Link A, Farias IP, da Silva MN, Röhe F et al.
2017. Capuchin monkey biogeography: understanding Sapa-
jus Pleistocene range expansion and the current sympatry be-
tween Cebus and Sapajus.Journal of Biogeography 44: 810–
820.
SOUTH AMERICA’S DIVERSITY 19
Lima SFB, Lucena RA, Santos GM, Souza JW, Christoersen
ML, Guimarães CR & Oliveira GS. Inventory of mollusks
from the estuary of the Paraíba River in northeastern Brazil.
Biota Neotropica 17: e20160239.
Lovejoy NR.1998. Marine incursion into South Amer ica. Nature
396: 421–422.
Lovejoy NR & Albert JS.2006. Miocene marine incursions
and marine/freshwater transitions: Evidence from Neotropical
shes. Journal of South American Earth Sciences 21: 5–13.
Lundberg JG & Aguilera O.2003. The Late Miocene Phracto-
cephalus catsh (Siluriformes: Pimelodidae) from Urumaco,
Venezuela: additional specimens and reinterpretation as a dis-
tinct species. Neotropical Ichtyology 1: 97–109.
Lundberg JG, Marshall LG, Guerrero J, Horton B, Malabarba
MCSL & Wesselingh F.1998. The stage for Neotropical sh
diversication: A history of tropical South American rivers.
In: Malabarba L, Reis RE, Vari RP, Lucena ZMS & Lu-
cena CAS (Eds.) Phylogeny and Classication of Neotropical
Fishes, Edipucrs, Porto Alegre, pp. 13–48.
Mayle FE, Beerling DJ, Gosling WD & Bush MB.2004. Res-
ponses of Amazonian ecosystems to climatic and atmospheric
carbon dioxide changes since the last glacial maximum. Philo-
sophical Transactions of the Royal Society B: Biological Scien-
ces 359: 499–514.
Miller KG, Kominz MA, Browning JV, Wright JD, Mountain
GS, Katz ME, Sugarman PJ, Cramer BS, Christie-Blick N
& Pekar SF.2005. The Phanerozoic record of global sea-level
change. Science 310: 1293–1298.
Mittermeier RA, Mittermeier CG, Pilgrim J, Fonseca G &
Konstant WR.2002. The pantanal - Wilderness: Earth’s last
wild places. In: Mittermeier RA, Mittermeier CG, Pilgrim
J, Fonseca G & Konstant WR (Eds.) Wilderness: Earth’s
last wild places (No. 333.782 W673w), México, MX: CEMEX,
Mexico D.F., Mexico, pp. 246–263.
Molino JF & Sabatier D.2001. Tree diversity in tropical rain
forests: a validation of the intermediate disturbance hypothe-
sis. Science 294: 1702–1704.
Mora A, Baby P, Roddaz M, Parra M, Brusset S, Hermoza W
& Espurt N.2010. Tectonic history of the Andes and sub-
Andean zones: implications for the development of the Ama-
zon drainage basin. In: Hoorn C & Wesselingh F (Eds.) Ama-
zonia, landscape and species evolution: a look into the past,
Hoboken: Blackwell-Wiley, West Sussex, UK, pp. 38–60.
Moritz C, Patton JL, Schneider CJ & Smith TB.2000. Diversi-
cation of rainforest faunas: an integrated molecular approach.
Annual Review of Ecology and Systematics 31: 533–563.
Morlon H.2014. Phylogenetic approaches for studying diversi-
cation. Ecology Letters 17: 508–525.
Mörner NA et al. 2016. Origin of the Amazonian rainforest. In-
ternational Journal of Geosciences 7: 470–478.
Naka LN & Brumeld RT.2018. The dual role of Amazonian
rivers in the generation and maintenance of avian diversity.
Science advances 4: eaar8575.
Noonan BP & Chippindale PT.2006. Dispersal and vicariance:
the complex evolutionary history of boid snakes. Molecular
Phylogenetics and Evolution 40: 347–358.
Oberdor T, Dias MS, Jézéquel C, Albert JS, Arantes CC,
Bigorne R, Carvajal-Valleros FM, De Wever A, Frederico
RG, Hidalgo M et al. 2019. Unexpected sh diversity gradi-
ents in the Amazon basin. Science Advances 5: eaav8681.
Parada A, Pardiñas UF, Salazar-Bravo J, D’Elía G & Palma
RE.2013. Dating an impressive Neotropical radiation: mole-
cular time estimates for the Sigmodontinae (Rodentia) provide
insights into its historical biogeography. Molecular phyloge-
netics and Evolution 66: 960–968.
Pennington T, Prado DE & Pendry CA.2000. Neotropical sea-
sonally dry forests and Quaternary vegetationchanges. Journal
of Biogeography 27: 261–273.
Perelman P, Johnson WE, Roos C, Seuánez HN, Horvath JE,
Moreira MA, Kessing B, Pontius J, Roelke M, Rumpler Y
et al. 2011. A molecular phylogeny of living primates. PLoS
Genet 7: e1001342.
Pianka ER.1977. Latitudinal gradients in species diversity: a re-
view of concepts. The American Naturalist 100: 33–46.
Pires J & Prance G.1985. The vegetation types of the Brazilian
Amazon. In: Prance GT & Lovejoy TE (Eds.) Key Environ-
ments: Amazonia, Pergamon Press, Oxford, UK, pp. 109–145.
Prance GT.1979. Notes on the vegetation of Amazonia III. The
terminology of Amazonian forest types subject to inundation.
Brittonia 31: 26–38.
Prance GT.1982.Biological Diversication in the Tropics.
Columbia University Press, New York, USA.
Quintero E, Ribas CC & Cracraft J.2013. The Andean Hapa-
lopsittaca parrots (Psittacidae, Aves): An example of montane-
tropical lowland vicariance. Zoologica Scripta 42: 28–43.
Ribas CC, Aleixo A, Nogueira AC, Miyaki CY & Cracraft J.
2012. A palaeobiogeographic model for biotic diversication
within Amazonia over the pastthree million years. Proceedings
of the Royal Society B: Biological Sciences 279: 681–689.
20 RI VAS
Ribas CC, Miyaki CY & Cracraft J.2009. Phylogenetic rela-
tionships, diversication and biogeography in Neotropical Bro-
togeris parakeets. Journal of Biogeography 36: 1712–1729.
Ribas CC, Moyle RG, Miyaki CY & Cracraft J.2007. The as-
sembly of montane biotas: linking Andean tectonics and cli-
matic oscillations to independent regimes of diversication in
Pionus parrots. Proceedings of the Royal Society B: Biological
Sciences 274: 2399–2408.
Rivas RA, Mittermeier CG, Pilgrim J, Fonseca G & Konstant
WR.2002. The llanos. In: Mittermeier RA, Mittermeier
CG, Pilgrim J, Fonseca G & Konstant WR (Eds.) Wilder-
ness: Earth’s last wild places (No. 333.782 W673w), México,
MX: CEMEX, Mexico D.F., Mexico, pp. 265–273.
Roddaz M, Brusset S, Baby P & Hérail G.2006. Miocene
tidal-inuenced sedimentation to continental Pliocene sedi-
mentation in the forebulge–backbulge depozones of the Beni–
Mamore foreland Basin (northern Bolivia). Journal of South
American Earth Sciences 20: 351–368.
Roxburgh SH, Shea K & Wilson JB.2004. The intermediate
disturbance hypothesis: patch dynamics and mechanisms of
species coexistence. Ecology 85: 359–371.
Rull V.2008. Speciation timing and neotropical biodiversity: The
Tertiary-Quaternary debate in the light of molecular phyloge-
netic evidence. Molecular Ecology 17: 2722–2729.
Rull V.2011. Neotropical biodiversity: timing and potential
drivers. Trends in Ecology & Evolution 26: 508–513.
Rull V.2015. Pleistocene speciation is not refuge speciation. Jour-
nal of Biogeography 42: 602–604.
Salas-Gismondi R, Flynn JJ, Baby P, Tejada-Lara JV, Wesse-
lingh FP & Antoine PO.2015. A Miocene hyperdiverse
crocodylian community reveals peculiar trophic dynamics in
proto-Amazonian mega-wetlands. Proceedings of the Royal
Society B: Biological Sciences 282: 20142490.
Santos JC, Coloma LA, Summers K, Caldwell JP, Ree R &
Cannatella DC.2009. Amazonian amphibian diversity is pri-
marily derived from Late Miocene Andean lineages. PLoS Bi-
ology 7: e1000056.
Sayer EJ, Sutclie LM, Ross RI & Tanner EV.2010. Arthro-
pod abundance and diversity in a lowland tropical forest oor
in Panama: the role of habitat space vs. nutrient concentrations.
Biotropica 42: 194–200.
Schoener TW.1971. Large-billed insectivorous birds: a precipi-
tous diversity gradient. The Condor 73: 154–161.
Sepulchre P, Sloan LC & Fluteau F.2010. Modelling the res-
ponse of Amazonian climate to the uplift of the Andean moun-
tain range. In: Hoorn C & Wesselingh F (Eds.) Amazonia,
landscape and species evolution: a look into the past, Hobo-
ken: Blackwell-Wiley, West Sussex, UK, pp. 211–222.
Silva GS, Roxo FF, Lujan NK, Tagliacollo VA, Zawadzki CH
& Oliveira C.2016. Transcontinental dispersal, ecological
opportunity and origins of an adaptive radiation in the Neotro-
pical catsh genus Hypostomus (Siluriformes: Loricariidae).
Molecular Ecology 25: 1511–1529.
Strecker M, Alonso R, Bookhagen B, Carrapa B, Hilley G, So-
bel E & Trauth M.2007. Tectonics and climate of the south-
ern central Andes. Annual Review of Earth and Planetary
Sciences 35: 747–787.
Teeling EC, Springer MS, Madsen O, Bates P, O’Brien SJ &
Murphy WJ.2005. A molecular phylogeny for bats illumi-
nates biogeography and the fossil record. Science 307: 580–
584.
Terborgh J.1992. Maintenance of diversity in tropical forests.
Biotropica 24: 283–292.
Terborgh J, Robinson SK, Parker III TA, Munn CA & Pier-
pont N.1990. Structure and organization of an Amazonian
forest bird community. Ecological Monographs 60: 213–238.
Terborgh J & Weske JS.1975. The role of competition in the
distribution of Andean birds. Ecology 56: 562–576.
Thompson AW, Betancur-R R, López-Fernández H & Ortí G.
2014. A time-calibrated, multi-locus phylogeny of piranhas
and pacus (Characiformes: Serrasalmidae) and a comparison
of species tree methods. Molecular Phylogenetics and Evolu-
tion 81: 242–257.
Valle-Levinson A & Bosley KT.2003. Reversing circulation
patterns in a tropical estuary. Journal of Geophysical Research:
Oceans 108: 3331.
Voelker G, Marks BD, Kahindo C, A’genonga U, Bapeamoni
F, Due LE, Huntley JW, Mulotwa E, Rosenbaum SA &
Light JE.2013. River barriers and cryptic biodiversity in an
evolutionary museum. Ecology and Evolution 3: 536–545.
Vonhof H, Wesselingh F, Kaandorp R, Davies G, Van Hinte
J, Guerrero J, Rasanen M, Romero-Pittman L & Ranzi A.
2003. Paleogeography of Miocene Western Amazonia: Iso-
topic composition of molluscan shells constrains the inuence
of marine incursions. Geological Society of America Bulletin
115: 983–993.
SOUTH AMERICA’S DIVERSITY 21
Vonhof HB & Kaandorp RJ.2010. Climate variation in Ama-
zonia during the Neogene and the Quaternary. In: Hoorn C
& Wesselingh F (Eds.) Amazonia, landscape and species evo-
lution: a look into the past, Hoboken: Blackwell-Wiley, West
Sussex, UK, pp. 201–210.
Vonhof HB, Wesselingh FP & Ganssen GM.1998. Reconstruc-
tion of the Miocene Western Amazonian aquatic system using
molluscan isotopic signatures. Palaeogeography, Palaeoclima-
tology, Palaeoecology 141: 85–93.
Ward PS, Brady SG, Fisher BL & Schultz TR.2015. The evolu-
tion of myrmicine ants: phylogeny and biogeography of a hy-
perdiverse ant clade (Hymenoptera: Formicidae). Systematic
Entomology 40: 61–81.
Weir JT.2006. DivergentTiming and Patterns of Species Accumu-
lation in Lowland and Highland Neotropical Birds. Evolution
60: 842–855.
Wesselingh F, Anderson L & Kadolsky D.2006. Molluscs
from the Miocene Pebas Formation of Peruvian and Colom-
bian Amazonia. Scripta Geologica 133: 19–290.
Wesselingh F & Salo J.2006. A Miocene perspective on the evolu-
tion of the Amazonian biota. Scripta Geologica 133: 439–458.
Wesselingh FP & Hoorn C.2011. Geological development of
Amazon and Orinoco basins. In: Albert JS & Reis RE (Eds.)
Historical Biogeography of Neotropical Freshwater Fishes,
University of California Press, Berkely, Los Angeles, USA, pp.
59–67.
Zachos J, Pagani M, Sloan L, Thomas E & Billups K.2001.
Trends, rhythms, and aberrations in global climate 65 Ma to
present. Science 292: 686–693.
Zachos JC, Dickens GR & Zeebe RE.2008. An early Ceno-
zoic perspective on greenhouse warming and carbon-cycle dy-
namics. Nature 451: 279–283.
