Extinción de Equidae y Proboscidea en América del Sur. Un test usando datos de isótopos de carbono-The extinction of Equidae and Proboscidea in South America. A test using Carbon isotope data

Carbon isotopes, preserved in 166 samples of fossil teeth and bone, provide key data for understanding the ecology of extinct horses and gomphotheres during the Plio-Pleistocene in South America. To analyze the patterns of dietary partitioning throughout this time we divided the samples into 19 groups, taking into account the genus and the age of the corresponding localities. In this study, the diets of both groups are assessed to test extinction hypotheses. The strong resource partitioning among herbivores assumed under Co-evolutionary disequilibrium hypothesis is supported by isotopic data of horses from latest Pleistocene. Hippidon and Equus had very different diets. In contrast, species of gomphotheres from late Pleistocene in South America seem to have had less specialized diets containing a broad mix of both C3 and C4 plants, which is in line with the dietary assumptions of the mosaic-nutrient hypothesis, but does not support the assumptions of Co-evolutionary disequilibrium hypothesis.


Introduction
Gomphothere and horse species arrived in South America during the Great American Biotic Interchange around 3 Ma (Webb, 1991).During this time, two main corridors developed in South Amer-ica that shaped the biogeography of both groups (Alberdi et al., 2011).The model with the greatest support of dispersal and diversification processes postulated seem to indicate that the small forms of both gomphotheres and horses (Cuvieronius hyodon, Equus andium, Equus insulatus and Hippidion saldiasi) utilized the Andes corridor, whereas the large forms of these groups (Stegomastodon waringi, Stegomastodon platensis, Equus neogeus and Hippidion principale) utilized the Eastern route and some coastal areas.The route of each form seems to reflect an adaptive shift in their ecology (Alberdi & Prado, 1992;Sánchez et al., 2004;Prado et al., 2011).
Despite their wide distribution and abundance for most of the Pleistocene, gomphotheres and horses had disappeared from South America, along with many other large animals by the end of this epoch (Barnosky et al., 2004).Two types of theories have been offered for this extinction.One of these groups of theories attributes the extinction of large mammals to climatic and ecological changes, while the other holds human hunting activities responsible.Martin (1984) proposed that the extinction of large mammals from North America, South America and Australia is related to various, sudden impacts resulting from human activity.This "overkill" hypothesis is supported by the synchronism of extinction with the arrival of large numbers of humans to these continents.For North America, some authors (Haynes, 2002;Fiedel, 2009) argue that large mammals, especially mastodons and mammoths, were exterminated by Clovis hunters at c. 11 000 years BP.Conversely, the survival of megamammals and large mammals till c. 7000 14 C BP in South America indicate that extinctions would have occurred throughout an extended period (c.6000-4000 14 C BP) considering the timing of human dispersal in the southern cone of South America (Steele & Politis, 2009).This situation does not provide any support for the Overkill and Blitzkried models in South America (Gutierrez & Martinez, 2008;Politis & Messineo, 2008;Borrero, 2009;Cione et al., 2009).
Authors who doubt the role of human hunting activities often attribute the extinctions to climatic and ecological changes, particularly to nutritional stress induced by rapid changes in plant communities.Climate may have provoked changes in communities of flora and as a result the diets of herbivores were altered, causing heightened periods of competition.Although gomphotheres and horses may have been able to adapt to any one of these environmental perturbations, the combination of all of them at the same time may have been devastating for species that showed more selective dietary adaptations.The "mosaic-nutrient hypothesis" argues that climate change reduced the growing season and local plant diversity, and also increased plant antiherbivore defenses, all of which reduced the carrying capacity of the environment for herbivores (Guthrie, 1984).A more general hypothesis (the coevolutionary disequilibrium hypothesis) was postulated by Graham & Lundelius (1984) who suggested that the high herbivore diversity of Pleistocene ecosystems was maintained by extensive resource partitioning, analogous to the grazing succession of modern African savannas, and that an extremely rapid glacial-interglacial transition reorganized floras, disrupting this tightly co-evolved system.
Dietary assumptions about Pleistocene herbivores being used in environmental hypotheses it can be tested with stable isotope data (Koch et al., 1994).Isotopic analyses can reveal information about resource use and resource partitioning among species and is also able to determine diet and habitat use (Tütken & Vennemann, 2009;Feranec et al., 2009).In this study, the diets of gomphotheres and horses from South America are assessed using carbon isotope analysis to test both hypotheses for extinction.
Stable carbon and oxygen isotopes are incorporated into the tooth and bone apatite of fossil specimen and are representative, respectively, of the food and water consumed while alive.The carbon isotope ratio is influenced by the type of plant material ingested, which is in turn influenced by the photosynthetic pathway utilized by the plants.During photosynthesis, C 3 plants in terrestrial ecosystems (trees, bushes, shrubs, forbs, and high altitude and high latitude grasses) discriminate more markedly against the heavy 13 C isotope during fixation of CO 2 than C 4 plants (tropical grasses and sedges).Thus C 3 and C 4 plants have distinct δ 13 C values.C 3 plants usually have δ 13 C values of -30 per mil (‰) to -22‰, with an average of approximately -26‰, whereas C 4 plants have δ 13 C values of -14 to -10‰, with an average of about -12‰ (Smith & Epstein, 1971;Vogel et al., 1978;Ehleringer et al., 1986Ehleringer et al., , 1991;;Cerling et al., 1993).Animals incorporate carbon isotopes from food into their teeth and bone with an additional fractionation of ~12 to 14‰ (Cerling & Harris, 1999;Passey et al., 2005).Mammals feeding on C 3 plants characteristically have δ 13 C values between -14 and -8‰, while animals that eat C 4 tropical grasses have δ 13 C values between +2 and -2‰.A mixed-feeder would fall somewhere in between these two extremes (Lee-Thorp & van der Merwe, 1987; Quade et al., 1992).Hence, the relative proportions of C 3 and C 4 vegetation in the diet of an animal can be determined by analyzing the δ 13 C value of its teeth and bones.A number of previous studies have used the carbon and oxygen isotopic abundance of fossils and paleosols from South America to reconstruct the diets of extinct herbivores and the paleoenvironmental parameters of ancient terrestrial communities and ecosystems (Latorre et al., 1997;MacFadden, 2000aMacFadden, , 2005;;MacFadden & Higgins, 2004;MacFadden et al., 1996;Sánchez et al., 2004).
Carbon isotopic data for horses from South America have been presented in several papers (MacFadden et al., 1994;MacFadden & Shockey, 1997;MacFadden, 2000b;Sánchez et al., 2006).MacFadden et al. (1999) interpreted the ancient distributions and latitudinal gradients of C 3 and C 4 grasses in North and South America from the stable isotopes preserved in the teeth of Pleistocene New World horses.In addition, some papers investigated the application of geochemical techniques in conjunction with morphological data to characterize and reconstruct the feeding ecology and niche characterization of individual species (MacFadden & Shockey, 1997;MacFadden, 1998).

Materials and methods
Here, we use stable carbon isotopes preserved in 166 fossil teeth and bone samples for 29 localities where horses have been recorded and 23 localities where gomphotheres have been recorded, published in Sánchez et al. (2004); MacFadden et al. (1994MacFadden et al. ( , 1996) ) and Prado et al. (in press).We have also included results previously published by Bocherens et al. (1996) for the modern elephant Loxodonta africana from the Amboseli Park (Kenya) to compare the dietary partitioning in the South American fossil taxa with that in an extant analogue (Table 1).
To analyze the patterns of dietary partitioning throughout the Plio-Pleistocene we divided the samples into 19 groups, taking into account the genera as well as the age of the corresponding localities.The nineteen groups are listed in tests to evaluate δ 13 C differences between each of the groups (Table 3), accepting the null hypothesis of no differences among means unless p < 0.05.We use SPSS 11.5 software for the statistical analysis.

Analytical results and discussion
The carbon isotopic ratio of gomphothere and horse samples provides significant ecological results.Between horses, the Hippidion samples are  more homogeneous than the Equus (Amerhippus) one (Table 1 and Figure 1).All the species of Hippidion were almost exclusively C 3 feeders but some individuals from Bolivia and Argentina fall at the lower end of the mixed C 3 /C 4 range.For instance, Hippidion principale from the Eastern corridor (at sea level) and Hippidion devillei from the Andes corridor yield similar δ 13 C values suggesting that they ate mainly C 3 plants.The same pattern of dietary partitioning was obtained when comparisons were made between the same taxa at different latitudes (between 22°S and 52°S).From the upper Pliocene (Hippidion devillei from Uquía locality) to the lower Pleistocene (Hippidion principale, from the province of Buenos Aires and Hippidion devillei from the Tarija locality) the dietary partitioning remains similar.The same pattern in dietary partitioning is observed throughout the Middle to Late Pleistocene (Figure 1) showing a predominance of C 3 plants.Also, we did not find differences between Hippidion saldiasi from the Ultima Esperanza in southern Patagonia and the other Hippidion species present at different localities across South America.Equus species have predominantly been grazers, and as such, carbon isotopic values provide evidence of the C 3 and C 4 grasses.The carbon isotope data indicates that Equus (Amerhippus) shows three different patterns of dietary partitioning.Samples of Equus (Amerhippus) neogeus from the province of Buenos Aires indicate a preference for C 3 plants in the diet.The samples from Ecuador and Bolivia [Equus (Amerhippus) andium and Equus (Amerhippus) insulatus] show a preference for a diet of mixed C 3 -C 4 plants, while those from La Carolina (sea level of Ecuador), Bolivia, and Brazil are mostly C 4 feeders.A few outliers (e.g.δ 13 C values of 9,2; 6,1 and 5,4‰ from La Carolina) cannot be easily explained.These extremely high δ 13 C values (above 3‰) cannot be explained by consumption of C 4 vegetation, which should impart an upper limit of about 3‰.These outliers could be the result from one of several possibilities, such as individuals living in costal peninsula areas of Ecuador during the time in which C 4 grasses were abundant and may have produced δ13C values not observed in the modern ecosystem, or the sample presents taphonomic alteration.
For gomphotheres, the δ 13 C values of Cuvieronius samples indicate mixed feeding (Table 1 and Figure 2).Carbon isotopic data from Cuvieronius from Bolivia (MacFadden et al., 1994;MacFadden & Shockey, 1997) suggests an adaptation from mixed feeding to grazing.One notable exception was Cuvieronius from Chile.Specimens of this group indicate that they were exclusively C 3 feeders (Figure 2).Stegomastodon shows two different adaptations.Samples from Buenos Aires Province (except for Quequén Salado samples) and Brazil indicate that these species were mixed-feeders whereas those from La Carolina (Ecuador) were mostly C 4 feeders.Substantial differences in isotopic composition are also observed between these two genera as they evolved from the Middle to the Late Pleistocene (Figure 2).Stegomastodon from the Middle Pleistocene of Buenos Aires fed on a mixed diet, as their isotopic values are more homogeneous.Alternatively, Stegomastodon from the Late Pleistocene exhibit a wider range of dietary adaptations.These include an exclusively C 3 diet (Quequén Salado), a mixed C 3 -C 4 diet (Buenos Aires), and a diet completely composed of C 4 plants (Ecuador).The dietary regimes of Cuvieronius samples from the Middle and Late Pleistocene, on the other hand, show less variation.With the exception of the strictly C 3 feeding Cuvieronius of the Late Pleistocene in Chile, mixed feeding predominated in both the Middle Pleistocene (Bolivia) and the Late Pleistocene (Ecuador).A trend from a mixed C 3 -C 4 diet in the middle Pleistocene to a more strictly C 3 diet in the upper Pleistocene can be more clearly observed in the Buenos Aires remains.

Concluding Remarks
The Late Quaternary Extinction roughly coincided with the most recent glacial-interglacial transi- tion, leading some to conclude that megafauna extinction was due to environmental change.Habitat loss hypotheses argue that as climate changed, areas with adequate conditions to maintain large mammals either disappeared entirely or became too small and fragmented to support viable populations (Koch & Barnosky, 2006).The mosaic-nutrient hypothesis is a special case of habitat loss.It argues that climate change reduced the growing season and local plant diversity, and also increased plant antiherbivore defenses, all of which reduced carrying capacity for herbivores (Guthrie, 1984).On the other hand, the Co-evolutionary disequilibrium is a more general hypothesis.It posits that the high herbivore diversity of Pleistocene ecosystems was maintained by extensive resource partitioning and that an extremely rapid glacial-interglacial transition reorganized floras, disrupting this tightly coevolved system (Graham & Lundelius, 1984).The premise under these entire hypotheses is the claim that the last glacial-interglacial transition was unusually large and unusually rapid relative to earlier glacial-interglacial transitions, too fast for animal adaptation or redistribution in the new climate space.In this study, the diets of gomphotheres and horses from South America are assessed using carbon isotope analysis to test principally the mosaicnutrient and Co-evolutionary disequilibrium.
Based on modern analogues, Pleistocene horses are inferred to be grazers but none of the grazing horses were interpreted as consumers of only C 4 grasses.Our data shows that Equus (Amerhippus) had three different patterns of dietary partitioning.Equus (Amerhippus) neogeus from the province Buenos Aires indicates a preference for C 3 plants.Equus (Amerhippus) andium from Ecuador and Equus (Amerhippus) insulates from Bolivia show a preference in a mixed diet of C 3 -C 4 plants, while Equus (Amerhippus) santaeelenae from La Carolina (sea level of Ecuador) and Brazil are mostly C 4 feeders (Prado et al., 2011).In particularly, the specimen from the latest Pleistocene shows a more focused dietary adaptation that suggests that they were specialized feeders.
Specimens of gomphotheres from the Middle and upper Pleistocene in South America exhibit feeding strategies similar to those of modern elephants, which live in diverse habitats, are opportunists, and therefore are capable of living on nearly any dietary mixture.Same specimen of Cuvieronius from Chile fed on C 3 plants exclusively, whereas other specimens of the genus had a mixed C 3 -C 4 diet.The genus Stegomastodon showed a broader range of dietary adaptations, including predominantly C 3 feeders (in Buenos Aires Province), exclusively C 4 feeders (in La Carolina Peninsula, Ecuador), and mixed C 3 -C 4 feeders (from several localities form Brazil and Argentina).
In sum, the strong resource partitioning among herbivores assumed under Co-evolutionary disequilibrium hypothesis is supported by isotopic data of horses from latest Pleistocene.Hippidon and Equus had very different diets.In contrast, species of gomphotheres from late Pleistocene in South America seem to have had less specialized diets containing a broad mix of both C 3 and C 4 plants, which is in line with the dietary assumptions of the mosaic-nutrient hypothesis, but does not support the assumptions of Co-evolutionary disequilibrium hypothesis.

Fig. 1 .
Fig. 1.-Patterns of dietary partitioning through time for South American horses.