SEARCH

SEARCH BY CITATION

Keywords:

  • Paleocene;
  • South America;
  • biochronology;
  • Tiupampan;
  • Peligran;
  • Puercan;
  • mammals

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Concluding Remarks
  7. Acknowledgments
  8. References
  9. Supporting Information

Abstract:  The oldest Cenozoic mammalian assemblages in South America have been recovered from levels of the Hansen Member of the Salamanca Formation, Punta Peligro locality in Argentina, and from the Santa Lucía Formation in Tiupampa, Bolivia. These faunas led to the recognition of the Peligran and Tiupampan South American Land Mammal Ages (SALMAs), each alternatively regarded as the oldest Paleocene SALMA. Due to the lack of radioisotopic dates for mammals bearing levels at these localities, no agreement has been reached yet about their relative ages. In this paper, the role of mammal faunas in age inference is discussed. Analysis of the SALMAs shows that the presence of non-therian mammals in the Peligran is of little consequence to the biochronological evaluation, reflecting instead a relict Mesozoic distribution. In contrast, therian mammals are particularly important in that (1) they were Lauraisan immigrants and (2) they support direct comparisons between the Tiupampa and Punta Peligro faunas. Parsimony and cluster analysis were used to quantitatively test hypotheses concerning the relative age of the Peligran and Tiupampan SALMAs. Our results support the hypothesis that the Tiupampan SALMA (early Danian) is older than the Peligran SALMA (early Selandian). This alignment results in an interpretation of the evolutionary history of South American land mammals that is more straightforward than the alternative.

At the time of the Cretaceous–Tertiary transition, South American mammals underwent major changes in taxonomic composition (Pascual and Ortiz-Jaureguizar 1990, 1992; Pascual et al. 1996). Several gondwanan lineages became completely extinct, while others experienced cladogenetic events and survived into the Paleogene. This faunal turnover was in part caused by the immigration of therian mammals from North America, which radiated throughout the whole continent, reached Antarctica (Reguero et al. 2002) and, in the case of marsupials, Australia (Simpson 1978; Woodburne and Case 1996). Whereas this general pattern is understood, many of its details are still poorly known. Palaeontological evidence is still scarce, and there is an important gap in the South American mammalian fossil record between the latest Cretaceous and the earliest Paleocene.

The earliest known Paleocene land-mammal faunas in South America come from the localities of Punta Peligro (Patagonia, southern Argentina), and Tiupampa (south central Bolivia; Text-fig. 1); both have been regarded as early Paleocene in age (Bonaparte et al. 1993; Muizon 1998). A third locality, the Grenier Farm site near Paso del Sapo in western Patagonia (Text-fig. 1), includes the oldest Cenozoic mammal-bearing levels up to now known for this continent. However, as only a single mammalian specimen is known from the Lefipán Formation, the marsupial Cocatherium lefipanum (Goin et al. 2006a), this taxon is not included in the present analysis. The lack of radioisotopic dates for Punta Peligro and Tiupampa localities led to several attempts to calibrate them through their biological content and magnetostratigraphy; however, results were controversially interpreted (Marshall et al. 1981; Bonaparte et al. 1993; Somoza et al. 1995; Marshall et al. 1997; Muizon 1998; Muizon and Cifelli 2000). In short, no agreement has been reached yet about the numerical or relative ages of these levels and faunas.

image

Figure TEXT-FIG. 1..  South American map with localities mentioned in the text.

Download figure to PowerPoint

The Punta Peligro fauna comes from the basal levels of the ‘Banco Negro Inferior’ (BNI), in the Hansen Member of the Salamanca Formation (Text-fig. 2) (Andreis et al. 1975). The BNI levels are not restricted to Punta Peligro and are widely distributed in southeastern Chubut and northeastern Santa Cruz provinces (Bond et al. 1995). Despite the broad distribution of these sediments, the richest mammalian record still derives from the Punta Peligro locality. The Punta Peligro fauna is composed of leptodactylid frogs, chelid turtles, alligatorid crocodyles (Bonaparte et al. 1993), and a mixture of non-tribosphenic Gondwanan mammals, as well as mammals derived from the Laurasian boreosphenid stock represented by placentals and marsupials (Pascual et al. 1992; Bonaparte et al. 1993; Bond et al. 1995; Bonaparte and Morales 1997; Gelfo and Pascual 2001).

image

Figure TEXT-FIG. 2..  Geochronology and magnetostratigraphy from Luterbacher et al. (2004). Biochronological units follow Woodburne (2004) for NALMAs. The litostratigraphic units are from: 1Paso del Sapo, Chubut, Argentina; 2Itaboraí, Río de Janeiro, Brazil; 3Tiupampa and Pajcha Pata, Bolivia; 4Santo Anastacio, Brazil; 5Río Negro, Argentina.

Download figure to PowerPoint

The relative age of the BNI was inferred from several 40K–40Ar dates taken from rocks below the BNI, as well as from paleomagnetic data. Marshall et al. (1981) dated two samples at the base of the Salamanca Formation at the upper course of the Río Chico, just east of Lago Colhué-Huapí as 64 ± 0.8 and 62.8 ± 0.8 Ma respectively. A less reliable date was obtained from a sample of vitric tuff from the upper Hansen Member of the Salamanca Formation at Cañadón Hondo as 61 ± 5 Ma (Marshall et al. 1981), subsequently corrected to 62.5 ± 5 Ma (Marshall 1982). Even though this last date was mentioned as coming from the upper Hansen Member of the Salamanca Formation, is not belonging to the BNI. Marshall et al. (1981, 1997) considered the BNI to be at the base of the overlying Rio Chico Formation, not at the top of Salamanca Formation as stated by several authors (e.g. Andreis et al. 1975; McCartney 1933; Simpson 1935a, b; Bonaparte et al. 1993). So, the BNI should be younger than the dated tuff which occurs stratigraphically about 10 m below. A hundred kilometres at West from Punta Peligro, Iglesias et al. (2007) sampled a tuff horizon located approximately 40 m above the BNI, for 40Ar–39Ar analysis, but the resulting isochron age of 57.80 ± 6.00 Ma has little interpretative value.

The magnetostratigraphic calibration of the BNI has shown contradictory results. Marshall et al. (1981) stated that in the area of Cerro Redondo (Text-fig. 1), the BNI and the overlying 90 m of sediments, as well as the BNI at the Punta Peligro Locality correspond to Chron 26r, which could be assigned to an age between 58.7 ± 0.2 and 61.70 ± 0.2 Ma. (Luterbacher et al. 2004). In localities farther north of Punta Peligro, at the El Gauchito and Las Violetas farms, the BNI was correlated with Chron 27n (Somoza et al. 1995) which belongs to the last span of the Danian up to 61.70 ± 0.2 Ma. (Luterbacher et al. 2004). According to these paleomagnetic data, the BNI does not seem to be regionally synchronous (Somoza et al. 1995); this could be interpreted as reflecting the retreat of the Salamancan sea, in which the Salamanca Formation was deposited. However, Bonaparte et al. (1993) reinterpreted the Marshall et al. (1981) analysis and stated that the BNI could represent the span of Chron 27; the authors regarded the middle of C27r as a better estimation. They also argued that the diachronism shown by the various BNI levels were probably an artefact due to methodological errors.

The Tiupampa locality is located about 95 km southeast of the city of Cochabamba, Mizque Province, Department of Cochabamba, Bolivia (Text-fig. 1). The faunal assemblage was recovered from the Santa Lucía Formation (Text-fig. 2) (sensu Gayet et al. 1991; Muizon 1991; Muizon and Brito 1993), and is composed of a diverse array of therian mammals. The age of the Tiupampa mammal-bearing levels was initially interpreted as Late Cretaceous (Marshall et al. 1983; Muizon et al. 1983, 1984; Marshall and Muizon 1988), subsequently as early Paleocene (Ortiz-Jaureguizar and Pascual 1989; Marshall 1989; Gayet et al. 1991; Muizon 1991; Bonaparte et al. 1993; Muizon and Brito 1993), earliest late Paleocene (Marshall et al. 1997), and finally as Puercan equivalent, that is early Paleocene (Muizon 1998 and Muizon and Cifelli 2000). Although all the Santa Lucía Formation deposits were considered as belonging to Chron 26r and referred to the base of the Selandian (Marshall et al. 1997; Sempere et al. 1997); this interpretation was rejected by Muizon (1998).

In contrast to the Punta Peligro fauna, where Mesozoic Gondwanan mammal lineages have been found, the mammal fauna of the Santa Lucía Formation is exclusively represented by Laurasian therian immigrants. A few mammal teeth probably belonging to dryolestoids were collected at the nearby locality of Pajcha Pata, also in Bolivia (Text-fig. 1), in the upper part of the lower member of the El Molino Formation (Text-fig. 2) (Gayet et al. 2001). This unit underlies the Santa Lucia Formation (Text-fig. 2), which is separated from it by a slight unconformity, and was regarded as Late Cretaceous according to the magnetostratigraphy and sequence stratigraphic interpretation (Sempere et al. 1997).

As knowledge regarding the Punta Peligro and Tiupampa faunas increased, several biochronological inferences were attempted. Pascual and Ortiz-Jaureguizar (1990) regarded both faunas as contemporaries, Danian in age, and as such were referred to the same (Tiupampan) South American Land Mammal Age (SALMA). Later on, the Peligran SALMA was defined by Bonaparte et al. (1993) and recognized as different from, and younger than, the Tiupampan. Marshall et al. (1997) equated the contact of the marine succession of the Salamanca Formation and the BNI as the regression event identified by Haq et al. (1987) between 60.5 and 60 Ma. They argued that the base of the BNI at Punta Peligro could be correlated with the base of the Santa Lucía Formation in Bolivia (Text-fig. 2), and that both represent the Danian–Selandian boundary at Cron 26r. Because the Tiupampan fauna belongs to the middle part of the Santa Lucía Formation, the authors concluded that the Peligran fauna should be older than the Tiupampan (Marshall et al. 1997). Their conclusions were indirectly supported by a multivariate faunal analysis in which the Peligran turned out to be more closely related to the Late Cretaceous Alamitan fauna (Text-fig. 2) of the Los Alamitos locality (Rio Negro Province, Argentina, Text-fig. 1) than to any younger Paleocene fauna (Ortiz-Jaureguizar et al. 1999).

In this study, we challenge these previous hypotheses (Marshall et al. 1997; Ortiz-Jaureguizar et al. 1999; Pascual and Ortiz-Jaureguizar 2007), providing a new faunal analysis of the Peligran and Tiupampan SALMAs, using different methods of comparison. We also critically review previous geological arguments and discuss the value of mammalian Gondwanan lineages, as well as of the Laurasian therian immigrants, in South American biochronological inferences.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Concluding Remarks
  7. Acknowledgments
  8. References
  9. Supporting Information

To evaluate the faunal relationships between the Peligran and Tiupampan and other LMAs, two different analytical methods were performed: a cluster and a parsimony analysis. A parsimony-based method, known as ‘evolutionary lineages’ (Martinez 1995) and ‘cladistic biochronological analysis’ (Makovicky 2007), previously had been used to search for the relative chronological position of localities and mammal ages respectively. However, the method followed here is based on the Parsimony Analysis of Endemicity (PAE; Rosen 1988) and Cladistic Analysis of Distribution and Endemism (CADE; Porzecanski and Cracraft 2005). Ortiz-Jaureguizar and Posadas (1999) used PAE to analyze the compositional changes in South American mammal faunas since the Late Cretaceous. In contrast to the biogeographic application of the parsimony approach, which attempts to either elucidate the history of areas or localities (Rosen 1988), or to detect areas of endemism (Morrone 1994); in this paper, we use it to make inferences about the relative age of biochronological units.

A data matrix of 263 taxa and 10 faunas (LMAs) was built for the quantitative analyses (see Table S1). Our taxonomic list differs from previous ones particularly in the South American taxa (e.g. Pascual et al. 1996; Pascual and Ortiz-Jaureguizar 2007). Data were obtained partially from the literature (Bond et al. 1995; Bonaparte 2002; Chornogubsky 2003; Woodburne 2004) and personal observations. Mammal assemblages include the well-known, oldest Paleocene North American Land Mammal Age (NALMA), the Puercan, divided in three subages, respectively named Puercan 1, 2 and 3 from older to younger. SALMAs include the Late Cretaceous Alamitan, and the Paleogene Tiupampan, Peligran, Itaboraian, Riochican and Casamayoran (Text-fig. 2). The Casamayoran SALMA was divided into Vacan and Barrancan subages following Cifelli (1985). The Paso del Sapo fauna from western Patagonia (Text-fig. 1; Goin et al. 2001; Bond et al. 2002), which seems to fill part of the gap between the Riochican and the Vacan (Text-fig. 2), was not included in the analysis because it is currently under study (Tejedor et al. in prep.). Pseudo-extinctions were not coded in the matrix. For the CADE, the raw temporal distribution using a single taxonomic level (e.g. genus) is insufficient to resolve all the relationships of a data matrix. Consequently, the hierarchical information implied by the accepted classification schemes or phylogenetic relationships is used as a framework to code distribution, increasing the historical signal of area relationships (Porzecanski and Cracraft, 2005 and literature cited therein). In our analyses, the areas used in the CADE were replaced by biochronologic units (LMAs) and taxa were considered up to three hierarchical categories: genera, family, and order, or as a cladistically equivalent node of a suprafamilial taxonomic rank. Three independent data sets were analyzed from the main matrix. The first set includes only genera as characters (182 taxon-characters). The second set considers genera and families (240 taxon-characters); the last set uses the whole evidence (263 taxon-characters). The use of these hierarchical levels is justified to account for the absence of common genera between some LMAs and because the higher taxonomic categories would allow a more inclusive grouping of the biochronologic units.

The cluster analysis was performed with NTSYS 2.0 software (Rohlf 1977) using the Jaccard coefficient, and the unweighted pair-group method using arithmetic averages. This methodology only takes into account the presence of taxa between faunas to establish similarity, and does not consider shared absences as a similarity criterion (which may distort the relationships between the faunas; Cheetham and Hazel 1969). The Cophenetic Correlation Coefficient (Sneath and Sokal 1973) was computed for the resulting phenograms. Parsimony analysis was performed with TNT software version 1.1 (Goloboff et al. 2003) using the implicit enumeration option. In this analysis, another taxon was added to the matrix as an out-group, to root it with a hypothetical ancestral LMA where no taxa were present. Because of this, taxon absence or an extinction event would be considered as primitive from a cladistic point of view. The absolute and relative Bremer support of the nodes was calculated by searching up to 50 suboptimal trees.

To check our inferences we implemented a temporal ordination technique using the spectral-ordering algorithm (Fortelius et al. 2006) using the MATLAB 7.0 (The Mathworks, Natick, MA, USA).

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Concluding Remarks
  7. Acknowledgments
  8. References
  9. Supporting Information

The cluster analysis performed using the ‘genera’ subset of data (Text-fig. 3), shows three unrelated groups. One branch belongs to the Alamitan, which does not share genera with any other faunal assemblage. The second branch is formed by the three Puercan subages plus the Tiupampan SALMA, while the remaining one is arranged in the following way: ((((Vacan + Barrancan) Riochican) Itaboraian) Peligran).

image

Figure TEXT-FIG. 3..  Cluster analysis using the genera data set. Cophenetic Correlation Coefficient: 0.998.

Download figure to PowerPoint

When the ‘genera’ + ‘family’ data sets were considered, the same relationship between the faunal assemblages was found, but now the three independent groups obtained in the previous analysis appear as sequentially organized (Text-fig. 4). Two main groups are distinguished, the first formed only by the Alamitan SALMA, and the second clustering together the two other groups of the ‘genera’ analysis, and keeping the same internal relationships. That is, the Tiupampan SALMA is more closely related to the Puercan NALMA than to any other South American fauna. The remaining South American Paleogene mammalian faunas join together, with the Peligran SALMA being the most external.

image

Figure TEXT-FIG. 4..  Cluster analysis using the genera + family data set. Cophenetic Correlation Coefficient: 0.996.

Download figure to PowerPoint

The cluster analysis using the total-evidence matrix (Text-fig. 5) shows a minor change in the similarity of the faunas. Again, two basic groups are shown, the Alamitan on the one hand, and the rest of the assemblages on the other. However, the relationships of the Tiupampan fauna show a significant change. In contrast with the other two cluster analyses, the Tiupampan is split from the Puercan cluster, and is more closely related to all other South American faunas. The strong relationship between Tiupampan and Puercan LMAs in the first two analyses (Text-figs 3–4) is due to the small number of shared genera (Peradectes) and families (Peradectidae and Mioclaenidae). But in the last analysis (Text-fig. 5), the Tiupampan appears as the most external Paleogene SALMA, not directly related to the Puercan NALMA as in the previous analyses. This change in relationships when the complete matrix was employed happens because the Tiupampan and the rest of the Paleogene SALMAs share many higher taxonomic groups (see Table S1). In fact, the Tiupampan has an important number of endemic genera (i.e. Allqokirus, Khasia, Tiulordia, Szalinia, Jaskahadelphys, Pucadelphys, Mizquedelphys, Incadelphys, Andinodelphys, Mayulestes, Roberthoffstetteria, Molinodus, Tiuclaenus, Pucanodus, Silmoclaenus and Andinodus), which belong to ordinal groups well represented in the other SALMAs (i.e. Polydolopimorphia, Sparassodonta, ‘Condylarthra’ and Notoungulata).

image

Figure TEXT-FIG. 5..  Cluster analysis using the complete data set. Cophenetic Correlation Coefficient: 0.993.

Download figure to PowerPoint

Considering the results shown in Text-figures 3 and 4, a close relationship between the Tiupampan SALMA and the Puercan NALMA can be inferred. The Peligran never associates with the Late Cretaceous Alamitan SALMA or with the early Paleocene Puercan NALMA subages. In short, in our analyses, the Tiupampan SALMA appears to be more similar to the earliest Paleocene Puercan than any other LMAs, while the Peligran is always more closely related to the remaining Paleogene SALMAs.

The parsimony analysis (Text-fig. 6A–C) shows slightly different results, even when the same conclusion can be obtained. If only the ‘genera’ data subset is considered, the single most parsimonious tree (L: 196 and Text-fig. 6A) shows, in contrast to the cluster analysis, a polytomy in node 1 made out of four branches belonging to the Alamitan, the Peligran, and two resolved clades. The first of these clades (node 2 of Text-fig. 6A) show the Tiupampan as a sister LMA of the Puercan subdivisions. The other clade (node 3 of Text-fig. 6A) shows the Itaboraian as the sister LMA of the Riochican, and the latter, in turn, as the sister-group of the Casamayoran subages (Vacan and Barrancan). The parsimony analyses of the ‘genera + family’ data subsets (Text-fig. 6B) and the one employing the complete data set (Text-fig. 6C) show no unresolved polytomies. In both cases, only one most parsimonious tree was obtained, both with the same topology. In these analyses, the polytomy present in node 1 of Text-figure 6A is now resolved, with the Alamitan being the more external and isolated LMA and relocating the Peligran as the sister-group of node 3. Two groups are clearly defined in these analyses: one comprises the Puercan subages plus the Tiupampan SALMA as its sister-group (node 2 of Text-fig. 6B–C). The other group shows the Peligran as the sister-group of the clade which joins together the Itaboraian, Riochican, Vacan and Barrancan (node 4 of Text-fig. 6B–C). Even though there are no topological differences between the parsimony analysis of the ‘genera’ + ‘family’ data subset and the complete data matrix, the trees differ in length (L: 267 and 301 respectively), as well as in their absolute and relative Bremer support values (Text-fig. 6B–C).

image

Figure TEXT-FIG. 6..  Most parsimonious trees using: A, The ‘genera’ subset of data (L: 196). B, The ‘genera + family’ subset of data (L: 267). C, The complete data matrix (L: 301). The Bremer values are marked as ‘Absolute/Relative’ values.

Download figure to PowerPoint

In the above-mentioned parsimony analyses, the different taxa present in the nodes of all the LMAs behave as synapomorphic characters. All nodes are defined by a great number of ‘synapomorphies,’ except the most basal one which shows the Alamitan as the sister-group of the rest of the LMAs. This basal node does not have any ‘derived’ characters or common taxa with other nodes. The common presence of one taxon-character (Multituberculata) shared between Alamitan and the Puercan LMAs is considered as ‘homoplasic’. The presence of Sudamericidae and Dryolestida in the Alamitan and the Peligran is also interpreted a posteriori as an ‘homoplasic’ character. In the node 2, and despite the common taxa between the Tiupampan and the Puercan (which, in principle, allow the inference of great similarity), only two ‘synapomorphies’ are present in node 2 that associate these LMAs (the taxon-character Peradectidae and Peradectes). In fact, most of the taxa the Tiupampan has in common with the Puercan are exclusively shared with the younger subages of the latter (i.e. Puercan 2 and 3).

The node 5, which groups the Puercan subdivision, is supported by 18 and 22 common and exclusive taxa in the ‘genera + family’ subset and in the complete data matrix trees, respectively. Bremer values for this node are also very high: for the ‘genera + family’ tree there are 17 and 82 absolute and relative Bremer support, respectively, while there are 21 and 77 for the complete data matrix tree. The highest Bremer values occur at the node which joins the Puercan 2 and Puercan 3, with 28 (absolute) and 84 (relative) in the first tree and 30 (absolute) and 78 (relative) for the other.

Only three SALMAs were characterized by ‘synapomorphies’ related to secondary absences of taxa, representing an extinction event or a sampling deficit. But these SALMAs are not only supported by these absences but also by a substantial number of presences. In fact, of the 11 ‘synapomorphies’ of the Riochican only three are absences (Bonapartheriidae, Didelphimorphia and ‘Didelphidae’) and, due to their presence in younger and older SALMAs, probably reflect a sampling bias. In the Vacan there is only one absence (Amphidolops) out of 10 ‘synapomorphies’, while there are two for the Barrancan (Sparnotheriodontidae and Victorlemoinea) of 26.

Summing up, cluster and parsimony analyses show that the Tiupampan SALMA is more closely related to the oldest Paleocene mammal ages here included (i.e. the Puercan NALMA), than to the Peligran SALMA. This conclusion agrees with previous inferences (Bonaparte et al. 1993; Bonaparte and Morales 1997; Muizon 1998; Muizon and Cifelli 2000) about the relative age of both LMAs. We conclude that the best hypothesis that explains our current knowledge of South American Paleocene faunas-LMAs is to regard the Tiupampan SALMA as older than the Peligran SALMA.

Aware that these analyses do not guarantee by themselves the temporal arrangement of the faunas-LMAs, we test our results applying the spectral-ordering algorithm to our matrix, following the requirements of the method (Fortelius et al. 2006). In contrast to our analysis the results were suggestive of an ordering where the Riochican SALMA was younger than the Vacan and Barrancan, whereas on the basis of geological information, the outcrop order is in fact and without doubt just the opposite (see for example Legarreta and Uliana 1994). One possible explanation for these anomalous results is the Riochican sample bias problem mentioned above, which is also testified by the scarcity of the smallest ungulates. But, the main difficulty with the spectral-ordering analysis applied to our data, are the few isotopic dates available to temporally tie our units. So, due to the limits imposed by our data set, we did not considered the results from the spectral ordering analysis.

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Concluding Remarks
  7. Acknowledgments
  8. References
  9. Supporting Information

The analyses here performed are not entirely compatible with other multivariate analyses of early Tertiary South American faunas. Pascual and Ortiz-Jaureguizar (1991), using Land Mammals Ages as operational units, and families as characters, reviewed the pattern of mammal faunal changes in South America throughout the Late Cretaceous-Paleocene span, and compared them with the changes in North America. It should be noted that, when this analysis was performed, the Peligran SALMA was not formally recognized so that the Punta Peligro fauna was regarded as part of the Tiupampan SALMA. Later, in another cluster analysis of South American Tertiary faunas, Ortiz-Jaureguizar et al. (1999) considered that the Alamitan was related to the Peligran. It should be noted that, in contrast to the present cluster analysis, they included among the operational units the three Paleocene faunal zones defined by Simpson (1935b) and fully described by Bond et al. (1995), as well as the local fauna of Laguna Umayo in Perú. According to the stratigraphy of the San Jorge Gulf (Chubut Province), Legarreta and Uliana (1994) considered that: (1) the youngest of Simpson’s zones, the Carodnia faunal zone that overlies the Salamanca Formation, is represented by the Peñas Coloradas Formation; (2) the ‘Kibenikhoria zone’ in the Las Flores Formation; and (3) the ‘Ernestokokenia zone’ in the Koluel Kaikhe Formation (Text-fig. 2). Of these, the Kibenikhoria and Ernestokokenia faunal zones are referable to the Itaboraian and Riochican SALMAs respectively (Bond et al. 1995) and are included in those SALMAs in the present analyses. In contrast, we did not include the still poorly known ‘Carodnia zone’ in our analysis (see below).

The second mammal fauna considered by Ortiz-Jaureguizar et al. (1999), Laguna Umayo from Peru (Text-fig. 1), was variously referred as Late Cretaceous or Paleocene in age. But this locality was correlated to an interval within Chron 24r, so the strata and the fossils they bear would thus have been deposited during the latest Paleocene–earliest Eocene interval (late Thanetian–early Ypresian, i.e. between 55.9 and 53.4 Ma; Sigéet al. 2004). Pascual and Ortiz-Jaureguizar (2007) considered Laguna Umayo mammal assemblage as probably equivalent to the Tiupampan SALMA, but it can not be unequivocally referred to any particular mammal age. In fact, discounting some indeterminate therians, the only certain taxa described are Peradectes austrinum Sigé, 1971; based on an isolated upper molar, and the probable notoungulate Perutherium altiplanense Grambast et al., 1967 (Sigéet al. 2004). Even when the Tiupampan mammal assemblage includes an isolated molar referred as Peradectes sp. (Marshall and Muizon 1988; Muizon 1991) a Tiupampan age for Laguna Umayo could not be maintained subsequent to the conclusions of Sigéet al. (2004).

The analysis performed by Ortiz-Jaureguizar et al. (1999) sharply disagrees with our results in that they concluded that the Peligran SALMA should be considered as older than the Tiupampan SALMA, as also was inferred by Marshall et al. (1997). In the Ortiz-Jaureguizar et al. (1999) analysis, the low similarity between the Alamitan, Peligran, and Tiupampan faunas, and the ‘Carodnia zone’, was related to the high degree of endemic taxa. In fact, the affinity between the Peligran and the Alamitan SALMAs is, in their analysis, due only to the common presence of the family Sudamericidae. In the present work another common taxon shared between the Peligran and the Alamitan (Dryolestida) was added to the complete data set; however, no close relationships between them could be disclosed by our analysis. The presence of shared ordinal taxa, belonging to Mesozoic Gondwanan lineages, is not enough to support a relationship in the cluster or the parsimony analysis.

As stated above, we did not include the ‘Carodnia zone’ in our analysis. Even though a complete analysis of this zone exceeds the scope of the present paper, a few observations about it are needed to clarify our decision. The ‘Carodnia zone’ was based on the Xenungulate Carodnia feruglioi Simpson 1935a; found at the locality of Bajo de la Palangana (Simpson 1935b) (Text-fig. 1). The lower mammal level at Cerro Redondo (Text-fig. 1) was also regarded by Simpson (1935b) as belonging to the ‘Carodnia zone’. However, it is interesting to note that no common taxa were shared between these localities. Regarding the lower mammal level at Cerro Redondo, Simpson (1935b, p. 9) stated that it ‘…is somewhat but not significantly, earlier than the horizon of Carodnia in the Palangana basin.Marshall et al. (1997) considered the ‘Carodnia zone’ at Bajo de la Palangana and Cerro Redondo as related to the Tiupampan SALMA. Ortiz-Jaureguizar et al. (2007) also stated that the ‘Carodnia zone’ is more closely related to Tiupampan SALMA than to any other South American unit. But at least two arguments led us to reject this statement, and to not include the ‘Carodnia zone’ in our Tiupampan faunal list (see Table S1). First, we note the absence of shared common specific, generic or even familial taxa between the Tiupampan SALMA and ‘Carodnia zone’. Actually, the situation seems to be just the opposite: out of the five taxa described for the ‘Carodnia zone’, Amphidolops is also present in the Itaboraian, Riochican and Casamayoran (particularly Barrancan), and Carodnia is also recorded in the Itaboraian. Second, the Tiupampan SALMA shows a strong presence of North American mammal lineages (e.g. mioclaenid condylarths and alcidedorbignyid pantodonts), while in the ‘Carodnia zone’ only typical South American lineages are present. In short, we consider that the ‘Carodnia zone’ is not related to the Tiupampan SALMA and, in contrast, it could be either related to the Itaboraian SALMA or, more probably (as suggested by Bond et al. 1995), it represents another ‘mammal age’. The genus Carodnia is present in the Itaboraian SALMA with the larger C. vieirai Paula Couto 1952, only known from the Brazilian locality (Text-fig. 2). If C. feruglioi is more primitive than the Itaboraian C. vieirai, and recognizing that the Peñas Coloradas Formation stratigraphically underlies the Las Flores Formation (assigned to the Itaboraian SALMA; Bond et al. 1995), the ‘Carodnia zone’ could represent a new SALMA which, in any case, should be younger than the Peligran and older than the Itaboraian SALMAs (Gelfo et al. 2008) (Text-fig. 2).

A final comment should be provided regarding the age of the Itaboraian SALMA, which is here considered younger than previously stated (e.g. Marshall 1985; Bonaparte et al. 1993; Flynn and Swisher 1995; Pascual and Ortiz-Jaureguizar 2007). Stratigraphic levels bearing Itaboraian faunas have never been dated by isotopic or magnetostratigraphic methods in either the Las Flores (Patagonia, Argentina) or the Itaborai (Brazil) formations. In consequence, the chronology of the Itaboraian SALMA has been indirectly inferred (Marshall 1985). In Patagonia, the referral of the Casamayoran SALMA to the early Eocene (Simpson 1948) set the late Paleocene as the youngest possible boundary for the Itaboraian age. Later and because the Salamanca Formation was interpreted as representing the Danian marine transgression in Patagonia (Legarreta and Uliana 1994), a late Paleocene age was also inferred for the deposition of Itaboraian levels of the stratigraphically higher Las Flores formation. Subsequently Kay et al. (1999) dated tuffs belonging to the Sarmiento Formation at Gran Barranca and concluded that the Barrancan Subage of the Casamayoran SALMA was of late Eocene (Bartonian) age. Considering the close affinities between the Vacan and the Barrancan subages, it is reasonable to conclude that the Casamayoran is of late, or middle–late Eocene age. This leaves the possibility that the Itaboraian SALMA may be not Paleocene but actually early Eocene in age. In addition, based on the events of hydrothermal cementation in the Continental Rift of southeastern Brazil, Sant’Anna and Riccomini (2001) conclude that the main phase of these events at the Itaboraí Basin must have occurred around 50 Ma. An ankaramitic dike that partially cuts the carbonaceous levels of Itaboraí was dated by 40K–40Ar methods in 52.6 ± 2.4 Ma (Riccomini and Rodrigues-Francisco 1992). Because of these data, and because there is no evidence supporting a Paleocene age for the Itaboraian SALMA, we provisionally regard its age as early Eocene (Text-fig. 2).

The role of Mesozoic non-therian mammals as biochronological tools

The results of our study suggest that the presence of mammal lineages in the Peligran SALMA, otherwise typical of Mesozoic faunas, reflects a relict situation which has no biochronological significance. Mesozoic non-therian mammals were widely distributed in Gondwanan landmasses. This is attested to by the presence of several sudamericid gondwanatherians in the Late Cretaceous of Madagascar (i.e. Lavanify; Krause et al. 1997); India (i.e. Dakshinia and an unnamed taxon) (Anantharaman and Das Sarma 1997; Wilson et al. 2007) and, probably, another one from the Creataceous Red Sandstone Group sediments in Tanzania, Africa (Krause et al. 2003). No Cretaceous mammalian record is known from Antarctica. However, a dentary of a sudamericid from the Eocene of Antarctica suggests their previous presence (Goin et al. 2006b). In addition, an indeterminate isolated tooth from the ?late Eocene of Yahuarango Formation at Santa Rosa, Perú (Text-fig. 1) was favourably compared with Ferugliotherium (Goin et al. 2004). The same wider Mesozoic distribution can be inferred for dryolestoids if, aside from the Patagonian records, the unnamed mammals of the El Molino Formation, Late Cretaceous at Pajcha Pata, Bolivia (Text-fig. 1) (Gayet et al. 2001), and an isolated jaw from the Adamantina Formation in Santo Anastasio, Brazil (Text-fig. 1) (Bertini et al. 1993) referred as Mammalia incertae sedis, could be confirmed as belonging to the Dryolestida (Candeiro et al. 2006). Thus, the presence of Gondwanatheria and Dryolestida in South American Paleocene localities probably reflects a wider and more ancient Gondwanan distribution, rather than being positive evidence to link the Paleocene Peligran SALMA with the Late Cretaceous Alamitan unit. In short, stronger links between the Tiupampan and the Puercan faunas shown by our analysis suggests that the presence of Mesozoic lineages present in the Peligran fauna are ‘homoplasic’; i.e. they do not argue in favour of an older age of this fauna with respect to the Tiupampan, but instead attest to the survival of some lineages beyond the Cretaceous-Paleocene boundary. The Peligran fauna records a mixture of therians, originally derived from Laurasian stems, as well as ornithorhynchids (probably representing Australian immigrants), dryolestoids, and gondwanatheres (Table 1). But these last Mesozoic lineages are not represented by the same taxa present in the Late Cretaceous of Gondwana, but by genera with new and more derived characters. In fact non-therian mammals from the Peligran show advanced features with respect to their dryolestid and gondwanathere counterparts from the Alamitan SALMA. Peligrotherium differs from its relative Mesungulatum, in its larger size, more lophodont molars, and in the hypertrophy of the mesial molariforms (Gelfo and Pascual 2001). The gondwanatherian Sudamerica also shows derived features with respect to its Alamitan relative, such as gliriform incisors, and in being one of the geologically earliest occurrences of hypsodont cheek teeth with thick cementum (Pascual et al. 1999). The cladogenetic event that led to these specializations could be related to the progressive ecological replacement of the gondwanan Mesozoic mammals by the Laurasian therians. In fact, the change in mammalian composition, during the Cretaceous–Paleocene span was very different in North and South America. While in North America, the Mesozoic mammal families survived into the Paleocene (Cifelli et al. 2004; Lofgren et al. 2004), in South America gondwanan mammals were replaced by Laurasian immigrants. Only a couple of gondwanan mammal lineages survived into the Paleocene (Table 1). At first glance, the presence of both mammal groups in Punta Peligro suggests that the arrival of Laurasian mammals may have been responsible for the gondwanan mammals’ extinction, possibly via ecological replacement. But present data are insufficient for further analysis.

Table 1.    Comparisson of taxa present in the latest known Cretaceous and earliest known Paleocene faunas from South America. Thumbnail image of

One factor that could explain the absence of derived dryolestoids and gondwanatherians in other Paleocene localities (e.g. Tiupampa), and their presence in Punta Peligro, stems from the regional differences in southern South America that could be related to the extinction or the survival of these groups. Relevant to this discussion are the recent biogeographic attempts to understand South America not as a single unit but as a complex continent including at least two major components: on one side, the Neotropical Region (including most of South America except the Andean Range and Patagonia) shows closer biogeographic affinities with most of Africa and Southeastern Asia (the Holotropical Kingdom); on the other, the Andean Region (including Patagonia) has affinities with the Austral Kingdom, the latter also including Antarctica and Australia (Morrone 2001). Empirical evidence at hand suggests that several major mammalian lineages are (or were) endemic to the Austral Kingdom: Sudamericid gondwanatherians, monotremes and, among therians, the marsupials Microbiotheria and the extremely abundant Polydolopinae Polydolopimorphians.

However, the absence of these taxa at Tiupampa could be also explained by taphonomical reasons. The Tiupampa fauna comes from a very specific ecological environment. The whole outcrop of Tiupampa (ca. 10 × 3 km) traduces a tropical alluvial plain with meandering channels. The precise site where 95 per cent of the mammals and all the skeletons were found (no more than 100 m2) probably corresponds to an oxbow suddenly invaded by a flood of the main river, which would have brought almost instantaneously a major sand mass. This would explain the abundance of articulated skeletons (mammals, frogs, snakes and new born crocodiles) of animals living there and buried during this catastrophic event. The great abundance of frogs (especially), snakes and crocodiles eggs and nests are also suggestive of some pond isolated from the main river (like an oxbow or something similar). Therefore, the environment of the Tiupampa mammal fauna clearly suggests a specific biotope and consequently a strongly biased mammalian faunal list for the Tiupampan SALMA, because 95 per cent of the Tiupampan mammals known proceed from this site.

So, considering our parsimony and cluster analysis, we suggest that the ‘persistence’ of sudamericids and some dryolestoids at Punta Peligro are indicative of biogeographic, rather than biochronological, affinities. In turn, the absence of these taxa at Tiupampa could be related either to biogeography or to taphonomy. Whatever the interpretation, it is noteworthy that the presence or absence of these groups in both localities is a neutral feature in the biochronological inference presented here. In any case, the absence of taxa cannot be regard as a positive argument.

The role of Paleocene therian mammals as biochronological tools

The Cretaceous therian record in North America, and the younger Paleocene record in South America, supports the widespread inference that therian mammals dispersed from North to South America (Case et al. 2005 and literature therein). Even though the South American Paleocene mammal record is incompletely known, one should expect that the oldest record of therians would stem from Laurasian immigrants. In the same way, and even considering the possible difference in the evolutionary rates of some characters, older therians are expect to show a wide set of primitive characters with respect to their descendents. So, in contrast to the non-therian mammals previously discussed, therians, which were recorded in Punta Peligro and Tiupampa localities, should serve as positive evidence for estimating the relative ages of the SALMAs involved. Setting aside the isolated specimen of the ?polydolopimorph Cocatherium lefipanum from the Lefipán Formation in Paso del Sapo (Text-fig. 1), Chubut province (Goin et al. 2006a), the oldest therian assemblage recovered in Patagonia comes from the BNI of the Salamanca Formation at Punta Peligro (Bonaparte et al. 1993; Bond et al. 1995) which was regarded as the earliest Paleocene South American land mammal fauna (Marshall et al. 1997; Pascual and Ortiz-Jaureguizar 2007). But the Peligran therian assemblage (e.g. the didelphid Derorhynchus, the notoptern Requisia, and the ‘condylarths’Escribania and Raulvaccia) show distinct derived features regarding their Laurasian stem lineages. The eutheria consist of endemic South American families (e.g. Notonychopidae and Didolodontidae) instead of any of the older and more primitive, North American families (Table 1). Similarly, the Peligran marsupials include a moderately sized borhyaenoid, a derorhynchid, derived polydolopimorphians, and a ‘didelphid’ referable to Didelphopsis sp. (Table 1). In short, the Peligran therians are far advanced relative to what should be expected for the first immigrants from Laurasia.

In contrast to the Peligran theria, the Tiupampan mammals are represented only by Laurasian eutherian immigrants, pantodonts and mioclaenid ‘condylarths’, plus basal ‘didelphoid’ metatherians. All of these appear to be the morphologically most primitive South American therian mammals in their respective groups and comparable to those of North American Puercan faunas. Among metatherians, the basal polydolopimorphian Roberthoffstetteria nationalgeographica makes its single appearance in South America among the Tiupampa fauna, and is absent from any other Paleogene association (e.g. Punta Peligro). Interestingly, Roberthoffstetteria is most closely related, not to any other South American taxon, but to Ectocentrocristus foxi Rigby and Wolberg, 1987; from the Campanian of North America (Case et al. 2005). In turn, Tiupampan ‘didelphids’ seem to be more derived than the Late Cretaceous North American peradectoid stock, but less than Itaboraian opossums (Case et al. 2005).

The most diverse Tiupampan eutherians are represented by the ‘condylarths’Molinodus, Tiuclaenus, Pucanodus, Andinodus, and Simoclaenus which were referred to the Mioclaenidae; a family also recorded in the Puercan NALMA (see Muizon and Cifelli 2000; and literature cited therein). The presence of other North American archaic ungulates was also mentioned for the Tiupampan SALMA (Pascual and Ortiz-Jaureguizar 2007) and it should be clarify. In fact the maxilla fragment identified as a Periptychidae cf. Mimatuta (Muizon and Marshall 1991) was later referred to Pucanodus (Muizon and Cifelli 2000); and Andinodus, originally considered as a doubtful didolodontid or phenacodontid (Muizon and Marshall 1987), was later included in the Mioclaenidae Kollpaniinae, along with the rest of the Tiupampan ‘condylarths’ (Muizon and Cifelli 2000). The Peligran ‘condylarths’Raulvaccia and Escribania were first considered as Mioclaenidae (Bonaparte et al. 1993; Muizon and Cifelli 2000), but the analysis of a larger sample, particularly of upper dentition remains, favours the inclusion of these taxa within the Didolodontidae. This interpretation was reinforced by subsequent phylogenetic analysis (Gelfo 2004, 2007) and by the presence of derived features in Raulvaccia and Escribania, such as the presence of a well developed hypocone, cusp which was still not developed in the more primitive molars of the Tiupampan mioclaenid Kollpaniinae (Gelfo 1999, 2004, 2006, 2007).

The pantodont Alcidedorbignya inopinata is the only representative of the order known so far in South America (Muizon and Marshall 1992). The teeth of Alcidedorbignya differ from those of Asiatic pantodonts Harpyodidae and Bemalambdidae, in having a paracone and metacone separated basally, while they are connate in the latter two groups. This character state has been regarded a synapomorphy of the other pantodonts, which is present in all North American pantodonts (Muizon and Marshall 1992). Alcidedorbignya is 60 per cent smaller than the smallest and oldest North American pantodont, Pantolambda bathmodon. It is dentally less derived than P. bathmodon in lacking a mesostyle (formed by a marked labial inflexion of the centrocrista), a structure present in all the North American pantodonts. A complete skeleton of Alcidedorbignya with complete and undistorted skull (Text-fig. 7) has been discovered recently at Tiupampa (Muizon in prep.). This remarkable specimen clearly indicates that Alcidedorbignya more resembles P. bathmodon than any other North American pantodont. However, it is distinctly more generalized than Pantolambda in all its cranial and postcranial morphology and could represent an almost perfect morphological ancestor for this taxon.

image

Figure TEXT-FIG. 7..  Almost complete skeleton (MHNC 8372) of the pantodont Alcidedorbignya inopinata (Muizon in prep.). The specimen is from the site ‘the Quarry’ at Tiupampa, where have been found most of the other specimens including the holotype.

Download figure to PowerPoint

The skull of Alcidedorbignya shows a set of primitive characters with respect to Pantolambda, such as the presence of a lower narial opening, posteriorly wider nasals bones, and angular process smaller and hook-like. In contrast to Pantolambda and Eocene pantodonts, Alcidedorbignya shows a primitive postcranial skeleton with no trace of the graviportal trend. This could be inferred from the astragalus morphology which has a strongly condyloid head with a well defined neck; and from gracile limb bones. In fact, the general morphology of the limbs of Alcidedorbignya is more similar to that of the mioclaenid condylarths of Tiupampa than to those of the other pantodonts (Muizon et al. 1998).Therefore, the primitive morphology of Alcidedorbignya predates that of Pantolambda; thus an Alcidedorbignya-like ancestor of Pantolambda should be older than Pantolambda bathmodon. This latter species is from the medial Torrejonian NALMA (=To2), approximately between 63 and 62 Ma, and which is correlated approximately with the Chron 28n, a time span, which also includes To1 (Text-fig. 2). Because the beds of the Santa Lucia Formation at Tiupampa were referred to a single reversed chron (26r according to Marshall et al. 1997 but see below) it is probable that the beds containing Alcidedorbignya (in South America) are correlated to the Chron 28r, which approximately corresponds to Pu3. According to the age of this zone the Tiupampa fauna could be as old as 64–64.5 Ma (Woodburne 2006; Westerhold et al. 2008). The relationships of Alcidedorbignya with Pantolambda bathmodon therefore reinforce the hypothesis of a late Puercan equivalent age for the Tiupampan SALMA.

Summarizing, the Tiupampan mammals record is represented by Laurasian eutherian immigrants at family and ordinal taxonomic rank. The analyses here performed reinforce previous observations about the close link between the Tiupampan and the faunas from the early Paleocene of North America (van Valen 1988; Muizon and Cifelli 2000). The Tiupampan therian mammals are more primitive than those of the Peligran, which supports their older age.

The hypothesis of an older age for the Peligran, tested by its mammal content

In several phylogenetic analyses (e.g.: Muizon and Cifelli 2000; Gelfo 2004, 2006, 2007) therians from Tiupampa are placed in a basal position with respect to the Peligran ones. Their morphology fits well with that of the expected more primitive South American groups. Furthermore, they are comparable to earlier Paleocene North American groups, rather than to any of the Cenozoic South American taxa (Text-fig. 3–6). So, if the Tiupampan SALMA is considered as younger than Peligran, following Marshall et al. (1997), the mammals should be considered as an anachronistic and isolated assemblage. None of the characteristic South American therians, which were already present in the Peligran SALMA (e.g. the Litopterna, Didolodontidae, Bonapartheriidae, or Derorhynchidae), were found at Tiupampa. In fact, the only Tiupampan eutherian related to a native South American ‘ungulate’ is an isolated partial left M1? referred to the Notoungulata: Henricosbornidae or Oldfieldthomasiidae (Muizon et al. 1984). But this record should be carefully analysed, since no notoungulate has been found in the Peligran, and the fossil record of these ungulates seems to be older in lower latitudes than in higher ones of South America.

A complementary hypothesis to explain the high diversity and primitiveness of Tiupampan mammals relative to Peligran ones lies in the distinctly different latitudinal position of each locality, which are 18°S and 45°S respectively (Text-fig. 1). Actually, there is a latitudinal diversity gradient in which the average number of species per unit area increases with its proximity to the Equator (Pianka 2000). A wide variety of hypotheses have been proposed to explain the latitudinal variation in species diversity in both hemispheres. Arguments relate it to the relative surface area of biomes, topographic heterogeneity, precipitation levels, and several measures of energy availability in local environments, such as mean and maximum temperature, net primary productivity, actual and potential evapotranspiration, and solar radiation received per unit area (Ruggiero and Kitzberger 2004 and literature cited therein). In this context, the different taxonomic composition of Tiupampan and Peligran SALMAs was partially attributed to different ecological situations, which are related to the latitudinal differences of the type-localities (Pascual and Ortiz Jaureguizar, 2007). But at least two aspects attenuate the biogeographic argumentation favouring diachronism. In first place extreme differences in paleoclimatic parameters should be discarded. The presence of alligatorids and chelids in the Salamanca Formation attests to a warm, probably subtropical climate, which seems to be not so different from the conditions inferred for Santa Lucía Formation. The crocodiles’ presence allowed an inference of a mean annual temperature (MAT) equal to or higher than 14.2°C (Markwick, 1998). This is coincident with a MAT estimation of 14.1 ± 2.6°C based from the leaves analysis of a megaflora, which underlies the BNI (Iglesias et al. 2007).

The second issue deals with the taphonomic conditions of both localities. The high faunistic diversity at Tiupampa seems to be better explained by taphonomic conditions rather than biogeographic conditions. The mammals were deposited in channels of meandering rivers on a flat alluvial plain. The presence of almost complete skeletons, articulated bones and extremely smallsized animals suggests very little no or post mortem transportation (Marshall and Sigogneau-Russell 1995). In contrast, Peligran mammals do not seem to be found in situ (Bonaparte et al. 1993), they are isolated, fragmentary and mostly restricted to medium- to moderately large-sized animals. Summing up, the primitiveness of the Tiupampan therians seems to be better explained by chronologic arguments rather than by any others.

The hypothesis of an older age for the Peligran, tested by geological correlations

Mammalian biochronology suggests an earlier age for the Tiupampan than for the Peligran fauna, which is in contrast to Marshall et al. (1997) interpretation about the relative age of both localities. One of the Marshall et al. (1997) assumptions was that the contact between the marine succession of the Salamanca Formation and the BNI was equated with the Danian–Selandian boundary following Cande and Kent (1992, 1995) and Haq et al. (1987). Also, the regression marked by the El Molino and Santa Lucía formation contact was correlated with the supposed Danian–Selandian boundary (Sempere et al. 1997; Marshall et al. 1997). So, the base of the BNI at Punta Peligro, the base of the Santa Lucía Formation in Tiupampa and a third outcrop, the Mealla Formation at SSW of Tres Cruces in Jujuy Province, Argentina (Text-figs 1, 2) were correlated with the base of the Selandian (Marshall et al. 1997). These authors argued that, because the Tiupampan fauna belongs to the middle part of the Santa Lucía Formation, and that the Peligran fauna occurs at the base of the BNI, the latter should be older than the former.

The mammals found at the Mealla Formation were regarded as belonging to the Peligran SALMA, due to their stratigraphic position (Marshall et al. 1997). Despite the difficulties of correlating outcrops separated by long distances (e.g. Punta Peligro and Tiupampa are set apart more than 3000 km from each other), there are no data on sedimentation rates for the Santa Lucía, Salamanca or Mealla formations. So, it is not possible to state that the relative position of the fossils is sufficient to establish a relative chronological inference. Moreover, assigning the Mealla Formation to the Peligran SALMA cannot be based on its mammal content. The two henricosborniid notoungulates found there belong to the genus Simpsonotus, which has not been recorded elsewhere (Pascual et al. 1978). However, due to the first undoubted record of the Henricosborniidae and their diversity in Patagonia, the fauna of the Mealla Formation was considered as probably related to the Riochican SALMA (Pascual et al. 1981). Because until now no new taxa, paleomagnetic inferences or isotopic dates are known for Mealla Formation, we follow the Pascual et al. (1981) interpretation.

Having discarded the Mealla fauna as correlative with the Peligran, another important aspect against the Marshall et al. (1997) arguments is the identification of the Danian–Selandian boundary at Tiupampa. This was based on palaeomagnetic inferences and on the interpretation of the increase in grain size of the sediments as a global regression event. Following palaeomagnetic data, the Danian–Selandian boundary is considered now to correlate with that between C27n and C26r; that is, 61.70 ± 0.2 Ma (Luterbacher et al. 2004). However, Marshall et al. (1997) considered the boundary as occurring somewhere within the span of the C26r, following Cande and Kent (1992, 1995). So, if all the Santa Lucía and the upper part of the underlying El Molino Formation, of reversed polarity, was interpreted as Chron 26r (Marshall et al. 1997; Sempere et al. 1997), then the Danian–Selandian boundary should be recorded within the El Molino Formation and not at the contact between these formations. The lower part of the El Molino Formation was considered to be Maastrichtian due to its dinosaurs and fishes (Gayet et al. 1991). If the reversed magnetic polarity comprising Santa Lucía and the upper El Molino formations was referred to C26r (Marshall et al. 1997; Sempere et al. 1997), the Danian–Selandian boundary should be, at least, at the middle part of the El Molino Formation. However, there is no reason to assume that these reversed strata are in fact correlated with C26r. Sempere et al. (1997, pp. 718R–719L)) explain that: ‘A simple one-to-one correlation…to the geomagnetic polarity time scale is not evident…The correlation…is our preferred interpretation because it is corroborated by the paleontological data’. But palaeontological data, particularly mammalian biochronology, suggests an earlier age for the Tiupampan SALMA and a close link to medial or late Puercan age of North America. The mioclaenid Molinodus suarezi recorded in the Tiupampa fauna may represent an early dispersal event from North to South America (Muizon and Cifelli 2000, p. 145), perhaps of Puercan age. As summarized by Gayet et al. (1991) most metatherian and eutherian Tiupampan mammals are suggestive of an early Paleocene age. Furthermore, as mentioned above, the primitiveness of the Tiupampa pantodont, Alcidedorbignya, is an indication of an early Paleocene age (Puercan equivalent) for the Tiupampa fauna, as stated by Muizon (1998). This scenario implies that the reversed magnetic polarity with which the Tiupampa Fauna is associated could pertain to C28r rather than C26r, and that the regression interpreted from the presence of coarse-grained material in the Tiupampa sediments is open to interpretations other than a global cause.

Sempere et al. (1997, p. 719L) indicate that the non-marine sediments of the basal part of the Santa Lucía Formation reflect a eustatically controlled regression, without tectonic influence. On the other hand, Sempere et al. (1997, p. 715L) describe the distribution of the red-brown lacustrine mudstones of the lower Santa Lucía Formation as having been controlled by remaining subsidence of the basin and by the structural framework in which they occur. This suggests a local, rather than regional, cause for these deposits and diminishes the interpretation of a eustatic origin. Notwithstanding the proposed correlation of the Santa Lucía sediments with Chron C26r, there appears to be no compelling reason to equate the base of the Santa Lucía Formation with the Danian–Selandian boundary.

Sempere et al. (1997, p. 715L) also state that the middle Santa Lucía Formation begins with a ‘somewhat coarser facies’, but discuss this in the context of ‘highly subsident’ areas, and also indicate that the palaeostructural corridor in which they were deposited was reactivated at the time the top of the Middle Santa Lucía sandstone beds were deposited. The sandstone interval, which contains the Tiupampa Fauna, is about 50 m thick, according to Sempere et al. (1997; Text-fig. 5). Again, the evidence of tectonic activity associated with the Middle Santa Lucía Formation (also Sempere et al. 1997, p. 719L) diminishes the interpretation of its regional correlation at 59 Ma. In summary, the 64–64.5 Ma age for the Tiupampan SALMA based on its fossil mammals appears plausible.

The palaeomagnetic data at the BNI in Punta Peligro are less reliable. The BNI was considered as pertaining to C26r (Marshall et al. 1981, 1997) and also including most of C28n and the span of C27 (Bonaparte et al.1993). Despite that, a maximal age could be inferred from foraminifera and the macrofloras found in the Salamanca Formation below the BNI. Bertels (1975a, b) identified the lower section of the Salamanca Formation at Cerro Hansen as Danian, due to the presence of Globanomalina compressa which belongs to the P1c zone (Berggren et al. 1995; Luterbacher et al. 2004). The P1c zone is below the BNI and comprises the span between the last part of C28n up to the base of C27n (Luterbacher et al. 2004). These values refuted the inferred palaeomagnetic values of Bonaparte et al. (1993) and imply a younger age for the BNI. Marshall et al. (1981, 1997) argued for a younger age correlating the base of the BNI with the Danian–Selandian boundary (see comments above). But Iglesias et al. (2007) identified this boundary below the BNI in the fluvial sediments of the Salamanca Formation, where a rich macroflora was found. In consequence, the BNI should be younger than the Selandian base.

Summarizing, overall data are inconclusive as to the age of Santa Lucía Formation and the BNI at the Salamanca Formation and do not support a long-distance correlation between them. Our biochronological interpretation disagrees with those of Marshall et al. (1997) about the relative age of these faunas. In contrast, they match well with the relative age between them considered by Bonaparte et al. (1993), Flynn and Swisher (1995) and Muizon (1998).

Concluding Remarks

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Concluding Remarks
  7. Acknowledgments
  8. References
  9. Supporting Information

The cluster and the parsimony analyses performed here show similar results, even though only in the parsimony example is the Tiupampa SALMA clearly more closely related to the North American Puercan NALMA than to any of the South American faunas considered by us. This conclusion agrees with previous observations (Muizon 1991;Muizon and Brito 1993; Muizon 1998; Muizon and Cifelli 2000), which argued in favour of the similarity of the Tiupampan therian mammals to those of the early Paleocene Puercan NALMA. This does not necessarily mean that the Tiupampan is as old as the Puercan, but instead it testifies to the existence of a dispersal event from North to South America, recording the oldest known presence of foreign lineages (e.g. pantodonts, mioclaenids, didelphimorphians, and polydolopimorphians) in South America.

Having discarded the possibility that the Peligran and the Tiupampa faunas were contemporary (Bonaparte et al. 1993), two mutually exclusive hypotheses about their relative age were debated. As was argued in the discussion above, both hypotheses generate different kinds of predictions and assumptions. In the absence of precise radioisotopic dates for the relevant localities, these hypotheses could only be tested by indirect evidence. The geological correlation between the distant localities of Punta Peligro and Tiupampa, previously argued to support an older age for the Peligran SALMA (Marshall et al. 1997), is questionable as it does not fit with a detailed faunal analysis. Moreover, this last hypothesis needs a high number of ad hoc explanations to support it including an explanation of the mammal diversity, their biogeographic pattern and the stage of evolution of the taxa involved. In contrast, we find that the hypothesis which considers a relative older age for the Tiupampan SALMA, is a simpler model that gives more insights regarding the South American mammalian evolution during the (still poorly known) Paleocene.

In summary, our own analyses agree better with an older age for the Tiupampan fauna, which is probably early Danian in age, and correlated with C28r. In contrast, the Peligran represents a younger post-Danian SALMA, most probably early Selandian in age.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Concluding Remarks
  7. Acknowledgments
  8. References
  9. Supporting Information

Acknowledgements.  This research was supported by a CONICET Post-doctoral fellowship to JNG. FJG would like to thank CONICET (PIP 5621) and the Alexander von Humboldt Foundation for their support. We thanks Mikael Fortelius and Heikki Mannila for advised with the spectral ordering technique, and David P. Polly and two anonymous referees for the suggestions made to en early version of this paper. We would like to thank Cecilia Morgan for her valuable help.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Concluding Remarks
  7. Acknowledgments
  8. References
  9. Supporting Information
  • ANANTHARAMAN, S. and DAS SARMA, C. 1997. Palaeontological studies on the search of micromammals in the infra and intertrappean sediments of Karnataka. Records of the Geological Survey of India, 130, 239240.
  • ANDREIS, R., MAZZONI, M. and SPALLETTI, L. 1975. Estudio estratigráfico y paleoambiental de las sedimentitas terciarias entre Pico Salamanca y Bahía Bustamante, Provincia de Chubut. Revista de la Asociación Geológica Argentina, 30 (1), 85103.
  • BERGGREN, W. A., KENT, D. V., SWISHER, C. C. and AUBRY, M. 1995. A Revised Cenozoic Geochronology and Chronostratigraphy. 129212. In BERGGREN, W. A., KENT, D. V., SWISHER, C. C., AUBRY, M. and HARDENBOL, J. (eds). Geochronology, time scales and global stratigraphic correlation. SEPM Special Publication, 54, 594 pp
  • BERTELS, A. 1975a. Bioestratigrafía del Paleoceno marino en la provincia del Chubut, República Argentina. Actas I Congreso Argentino de Bioestratigrafía y Paleontología, Tucumán, 1974.
  • BERTELS, A. 1975b. Biostratigrafia del Paleogeno en la Republica Argentina. Revista Española de Micropaleontologia, Madrid, 7 (3), 429450.
  • BERTINI, R. J., MARSHALL, L. G., GAYET, M. and BRITO, P. M. 1993. The vertebrate fauna of the Adamantina and Marília formations, Upper Cretaceous of the Paraná Basin, southeast Brazil. Neues Jahrbuch für Geologie und Paläontologie, Abhandlungen, 188, 71101.
  • BONAPARTE, J. F. 2002. New Dryolestida (Theria) from the Late Cretaceous of Los Alamitos, Argentina, and paleogeographical comments. Neues Jahrbuch für Geologie und Paläontologie, Abhandlungen, 244, 339371.
  • BONAPARTE, J. F. and MORALES, J. 1997. Un primitivo Notonychopidae (Litopterna) del Paleoceno Inferior de Punta Peligro, Chubut, Argentina. Estudios Geológicos, 53, 183286.
  • BONAPARTE, J. F., Van VALEN, L. and KRAMARZ, L. 1993. La Fauna Local de Punta Peligro. Paleoceno inferior de la provincia de Chubut, Patagonia, Argentina. Evolutionary Monograph, 14, 161.
  • BOND, M., CARLINI, A. A., GOIN, F. J., LEGARRETA, L., ORTIZ-JAUREGUIZAR, E., PASCUAL, R. and ULIANA, M. A. 1995. Episodes in South American land mammal evolution and sedimentation: testing their apparent concurrence in a Paleocene succession from central Patagonia. Actas VI Congreso Argentino de Paleontología y Bioestratigrafía, Trelew, Argentina, 4758.
  • BOND, M., LÓPEZ, G. M. and REGUERO, M. A. 2002. El valor bioestratigráfico de los Ungulados registrados en el Eoceno de Paso del Sapo, Chubut. Ameghiniana Suplemento, 39 (4), 6R7R.
  • CANDE, S. C. and KENT, D. V. 1992. A New Geomagnetic polarity Time Scale for the Late Cretaceous and Cenozoic. Journal of Geophysical Research, 97 B10, 13.91713.951.
  • CANDE, S. C. and KENT, D. V. 1995. Revised calibration of the geomagnetic polarity timescale for the Late Cretaceous and Cenozoic. Journal of Geophysical Research, 100, B4, 60936095.
  • CANDEIRO, C. R. A., MARTINELLI, A. G., AVILLA, L. S. and RICH, T. H. 2006. Tetrapods from the Upper Cretaceous (Turonian–Maastrichtian) Bauru Group of Brazil. Cretaceous Research, 27, 923946.
  • CASE, J., GOIN, F. J. and WOODBURNE, M. 2005. South American marsupials in the Late Cretaceous of North America and the Origin of Marsupial Cohorts. Journal of Mammalian Evolution, 12 (3–4), 461494.
  • CHEETHAM, A. H. and HAZEL, J. E. 1969. Binary (presence–absence) similarity indices. Journal of Paleontology, 43, 11301136.
  • CHORNOGUBSKY, L. 2003. Revisión preliminar de los mamíferos de la formación Los Alamitos (Campaniano-Maastrichtiano, provincia de Río Negro, Argentina. Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires. Tésis de Grado, 159 pp.
  • CIFELLI, R. L. 1985. Biostratigraphy of the Casamayoran, Early Eocene of Patagonia. American Museum Novitates, 2820, 126.
  • CIFELLI, R. L., EBERLE, J. J., LOFGREN, D. L., LILLEGRAVEN, J. A. and CLEMENS, W. A. 2004. Mammalian Biochronology of the Latest Cretaceous in North America. 2142. In WOODBURNE, M. O. (ed.). Cenozoic mammals of North America. University of California Press, California, xix + 391 pp.
  • FLYNN, J. J. and SWISHER, C. C. III 1995. Cenozoic South American land mammal ages: correlation to global geochronologies. In BERGGREN, W. A., KENT, D. V., AUBRY, M.-P. and HARDENBOL, J. (eds). Geochronology, time-scales and global stratigraphic correlation: a unified framework for an historical geology. Society of Stratigraphic Geology, Special Publication, 54, 317333.
  • FORTELIUS, M., GIONIS, A., JERNVALL, J. and MANNILA, H. 2006. Spectral ordering and biochronology of European fossil mammals. Paleobiology, 32 (2), 206214.
  • GAYET, M., MARSHALL, L. G., MEUNIER, F., CAPPETA, H. and RAGE, J. 2001. Middle masstrichtian vertebrates (fishes, amphibians, dinosaurs and other reptiles, mammals) from Pajcha Pata (Bolivia). Biostratigraphic, paleoecologic and paleobiogeographic implications. Palaeogeography Paaleoclimatology, Palaeoecology, 169, 3968.
  • GAYET, M., MARSHALL, L. G. and SEMPERE, T. 1991. The Mesozoic and Paleocene vertebrates of Bolivia and their stratigraphic context: a review. In SUAREZ-SORUCO, R. (ed.). Fosiles y facies de Bolivia 1 Vertebrados. Revista Técnica de Yacimientos Petrolíferos Fiscales de Bolivia, 12 (3–4), 393433.
  • GELFO, J. N. 1999. New aspects of the Paleocene genus Escribania (Mammalia: Condylarthra). Ameghiniana Suplemento, 36 (4), 12R.
  • GELFO, J. N. 2004. A new South American mioclaenid (Mammalia Ungulatomotpha) from the Tertiary of Patagonia, Argentina. Ameghiniana, 41 (3), 475484.
  • GELFO, J. N. 2006. Los Didolodontidae (Mammalia: Ungulatomporpha) del Terciario Sudamericano. Sistemática, origen y evolución. Unpublished PhD thesis, Facultad de Ciencias Naturales y Museo, Universidad Nacional de La Plata, Argentina, 884, 455 pp.
  • GELFO, J. N. 2007. The ‘Condylarth’Raulvaccia peligrensis (Mammalia: Didolodontidae) from the Paleocene of Patagonia, Argentina. Journal of Vertebrate Paleontology, 27 (3), 651660.
  • GELFO, J. N., LÓPEZ, G. M. and BOND, M. 2008. A new Xenungulata (Mammalia) from the Paleocene of Patagonia, Argentina. Journal of Paleontology, 82 (2), 329335.
  • GELFO, J. N. and PASCUAL, R. 2001. Peligrotherium tropicalis (Mammalia, Dryolestida) from the Early Paleocene of Patagonia, a survival from a Mesozoic Gondwanan radiation. Geodiversitas, 23, 369379.
  • GOIN, F. J., PASCUAL, R., RTEJEDOR, M., GELFO, J. N., WOODBURNE, M., CASE, J., REGUERO, M., BOND, M., CIONE, A., UDRIZAR SAUTHIER, D., BALARINO, L., SCASSO, R., MEDINA, F. A and UBALDÓN, M. C. 2006a. The earliest Tertiary therian mamal from South America. Journal of Vertebrate Paleontology, 26(2), 505510.
  • GOIN, F. J., REGUERO, M. A., PASCUAL, R., VON KOENIGSWALD, W., WOODBURNE, M., CASE, J. A., MARENSSI, S. A., VIEYTES, C. and VIZCAÍNO, S. 2006b. First gondwanatherian mammal from Antarctica. 135144. In FRANCIS, J. E., PIRRIE, D. and CRAME, J. A. (eds). Cretaceous – tertiary high-latitude palaeoenvironments, James Ross Basin, Antarctica. Geological Society, London, Special Publications, 258, 206 pp.
  • GOIN, F. J., TEJEDOR, M. F. and ABELLO, A. 2001. Conclusiones preliminares sobre la asociación de marsupiales paleógenos de Laguna Giordanella (Paso del Sapo, Chubut, Argentina; Eoceno medio?). Ameghiniana Suplemento, 38 (4), 9R10R.
  • GOIN, F. J., VIEYTES, E. C., VUCETICH, M. G., CARLINI, A. A. and BOND, M. 2004. Enigmatic Mammal from the Paleogene of Perú. 145154. In CAMPBELL, K. E (ed.) The Paleogene mammalian fauna Of Santa Rosa, Amazonian Perú. Natural History Museum of Los Angeles, Sciences Series 40, 163 pp.
  • GOLOBOFF, P. A., FARRIS, J. S. and NIXON, K. 2003. Tree Analysis Using New Technology Version 1.0 ã. Available from the authors and from http://www.zmuc.dk/public/phylogeny
  • GRAMBAST, L., MARTINEZ, M., MATTAUER, M. and THALER, L. 1967. Perutherium altiplanense nov. gen., nov. sp., premier mammifère mésozoïque d’Amérique du Sud. Comptes Rendus de l’Académie des Sciences de Paris, D, 264, 707710.
  • HAQ, B. U., HARDENBOL, J. and VAIL, P. R 1987. Chronology of fluctuating sea levels since the Triassic. Science, 235, 11561167.
  • IGLESIAS, A., WILF, P., JOHNSON, K., ZAMUNER, A., CÚNEO, N. R and MATHEOS, S. D. 2007. A Paleocene lowland macroflora from Patagonia reveals significantly greater richness than North American analogs. Geology, 35 (10), 947950.
  • KAY, R. F., MADDEN, R. H., VUCETICH, M. G., CARLINI, A. A., MAZZONI, M. M., RE, G. H., HEIZLER, M. and SANDEMAN, H. 1999. Revised geochronology of the Casamayoran South American Land Mammal Age: climatic and biotic implications. Proceedings of the National Academy of Sciences of the United States of America, 96, 1323513240.
  • KRAUSE, D., GOTTFRIED, M. D., O’CONNOR, P. M. and ROBERTS, E. M. 2003. A Cretaceous mammal from Tanzania. Acta Palaeontologica Polonica, 48 (3), 321330.
  • KRAUSE, D., PRASAD, G., KOENIGSWALD, W., VON SAHNI, A. and GRINE, F. 1997. Cosmopolitanism among Late Cretaceous mammals. Nature, 390, 504507.
  • LEGARRETA, L. and ULIANA, M. 1994. Asociación de fósiles y hiatos en el Supracretácico-Neógeno de la Patagonia: una perspectiva estratigráfico-secuencial. Ameghiniana, 31 (3), 257281.
  • LOFGREN, D. L., LILLEGRAVEN, J. A. CLEMENS, W. A. GINGERICH, P. D. and WILLIAMSON, T. E. 2004. Paleocene biochronology: the Puercan through Clarkforkian land mammal ages. 43105. In WOODBURNE, M. O. (ed.). Late Cretaceous and Cenozoic Mammals of North America. Biostratigraphy and geochronology. Columbia University Press, New York. xix + 391 pp.
  • LUTERBACHER, H., PALI, J. R., BRINKHUIS, H., GRADSTEIN, F. M., HOOKER, J. J., MONECHI, S., OGG, J. G., POWELL, J., RÖHL, U., SANFILIPPO, A. and SCHMITZ, B. 2004. The Paleogene Period. 384408. In GRADSTEIN, F., OGG, J. and SMITH, A. (eds). A geologic time scale. Cambridge University Press, Cambridge, New York, 610 pp.
  • MAKOVICKY, P. J. 2007. Telling time from fossils: a phylogeny-based approach to chronological ordering of paleobiotas. Cladistics, 23, 122.
  • MARKWICK, P. J. 1998. Fossil crocodilians as indicators of Late Cretaceous and Cenozoic climates: implications for using palaeontological data in reconstructing palaeoclimate. Palaeogeography Palaeoclimatology, Palaeoecology, 137, 205271.
  • MARSHALL, L. G. 1982. Calibration of the Age of Mammals in South America. In BUFFETAUT, E., JANVIER, P., RAGE, J. C. and TASSY, P. (eds). Phylogénie et Paléobiogeógraphie. Geobios, Mémoire Special, 6, 427437.
  • MARSHALL, L. G. 1985. Geochronology and land-mammal biochronology of the transamerican faunal interchange. 4985. In ATEHLI, F. G. D. and WEBB, S. (eds). The great American biotic interchange. Plenum press, New York, xviii + 532 pp.
  • MARSHALL, L. G. 1989. The K-T boundary in South America: on which side is Tiupampa? National Geographic Research, 5, 268270.
  • MARSHALL, L. G., BUTLER, R., DRAKE, R. and CURTIS, G. 1981. Calibration of the Beginning of the age of mammals in Patagonia. Science, 212, 4345.
  • MARSHALL, L. G. and MUIZON, C. de 1988. The dawn of the age of mammals in South America. National Geographic Research, 4, 2355.
  • MARSHALL, L. G., MUIZON, DE C. and SIGÉ, B. 1983. Late Cretaceous mammals (Marsupialia) from Bolivia. Geobios, 16 (6), 739745.
  • MARSHALL, L. G., SEMPÈRE, T. and BUTLER, R. 1997. Chronostratigraphy of the Mammal-Bearing Paleocene of South America. Journal of South American Earth Sciences, 10 (1), 4970.
  • MARSHALL, L. G. and SIGOGNEAU-RUSSELL, D. 1995. Part I: The locality of Tiupampa: age, taphonomy and mammalian fauna. 1120. In MUIZON, DE C. (ed.). Pucadelphys andinus (Marsupialia, Mammalia) from early Paleocene of Bolivia. Mémoires Muséum National d’ histoire Naturelle, 165, 164 pp.
  • MARTINEZ, J. N. 1995. Biochronologie et methods de parcimonie. Bulletin de la Société Géologique de France, 166 (5), 517526.
  • McCARTNEY, G. C. 1933. The bentonites and closely related rocks of Patagonia. American Museum Novitates, 630 (15), 116.
  • MORRONE, J. J. 1994. On the identification of areas of endemism. Systematic Biology, 43, 438441.
  • MORRONE, J. J. 2001. A proposal concerning formal definitions of the Neotropical and Andean regions. Biogeographica, 77 (2), 6582.
  • MUIZON, C. de 1991. La fauna de mamíferos de Tiupampa (Paleoceno inferior, Formación Santa Lucía) Bolivia. Revista Técnica de Yacimientos Petrolíferos Fiscales de Bolivia, 12 (3–4), 575624.
  • MUIZON, C. DE 1998. Mayulestes ferox, a borhyaenoid (Metatheria, mammalia) from the early Paleocene of Bolivia. Phylogenetic and paleobiologic implications. Geodiversitas, 20 (1), 19142.
  • MUIZON, C. DE and BRITO, I. M. 1993. Le bassien calcaire de São José de Itaboraí (Rio de Janeiro, Brésil): ses relations fauniques avec le site de Tiupampa (Cochabamba, Bolivie). Annales de Paléontologie, 79 (3), 233268.
  • MUIZON, C. DE and CIFELLI, R. L. 2000. The ‘condylarths’ (archaic Ungulata, Mammalia) from the early Palaeocene of Tiupampa (Bolivia): implications on the origin of the South American ungulates. Geodiversitas, 22 (1), 47150.
  • MUIZON, C. DE, CIFELLI, R. L. and BERGQVIST, P. L. 1998. Eutherian tarsals from the early paleocene of Bolivia. Journal of Vertebrate Paleontology, 18 (3), 655663.
  • MUIZON, C. DE, GAYET, M., LAVENU, A., MARSHALL, L. G., SIGÉ, B. and VILLARROEL, C. 1983. Late Cretaceous vertebrates including mammals from Tiupampa, southcentral Bolivia. Geobios, 16, 747753.
  • MUIZON, C. DE and MARSHALL, L. G. 1987. Deux nouveaux condylarthres (Mammalia) du Maastrichtien de Tiupampa (Bolivie). Comptes rendus des séances de l’académie des sciences. Série 2, 304 (15), 947950.
  • MUIZON, C. DE and MARSHALL, L. G. 1991. Nouveaux Condylarthres du paléocène inférieur de Tiupampa (Bolivie). Bulletin du Muséum national d’histoire naturelle. Section C, Sciences de la terre, paléontologie, géologie, minéralogie, 13, (3–4), 201227.
  • MUIZON, C. DE and MARSHALL, L. G. 1992. Alcidedorbignya inopinata (Mammalia: Pantodonta) from the early Paleocene of Bolivia: phylogenetic and paleobiogeographic implications. Journal of Paleontology, 66, 509530.
  • MUIZON, C. DE, MARSHALL, L. G. and SIGÉ, B. 1984. The mammal fauna of the El Molino Formation (Late Cretaceous–Maestrichtian) at Tiupampa southcentral Bolivia. Bulletin du Muséum national d’Histoire naturelle, 6 (4), 327351.
  • ORTIZ-JAUREGUIZAR, E., BOND, M., CARLINI, A. A. and GOIN, F. J. 2007. Relaciones de similitud de las faunas de mamíferos continentales de Amétrica del Sur durante el Cretácico Superior–Paleógeno (‘Edades-mamífero’ Alamitense-Deseadense). XXIII Jornadas de la Sociedad Española de Paleontología. Libro de Resúmenes, 1, 171172.
  • ORTIZ-JAUREGUIZAR, E., CLADERA, G. and GIALOMBARDO, A. 1999. Relaciones de similitud entre las faunas del lapso Cretacico Superior-Paleoceno Superior en América del Sur. Temas Geológicos Mineros ITGE, 26, 280283.
  • ORTIZ-JAUREGUIZAR, E. and PASCUAL, R. 1989. South American land-mammal faunas during the Cretaceous–Tertiary transition: evolutionary biogeography. Contribuciones de los simposios sobre el Cretácico de América Latina, Parte A: Eventos y registro sedimentario A231A251.
  • ORTIZ-JAUREGUIZAR, E. and POSADAS, P. 1999. Desde el Cretácico Tardío hasta la actualidad: Los cambios composicionales en la fauna de mamíferos de América del Sur a la luz del PAE (Parsimony Análisis of Endemicity). II Reunión Argentina de Cladística y Biogeografía, Buenos Aires, 22 pp.
  • PASCUAL, R., ARCHER, M., ORTIZ-JAUREGUIZAR, E., PRADO, J. L., GODTHELP, H. and HAND, S. J. 1992. First discovery of monotrenes in South America. Nature, 356, 704706.
  • PASCUAL, R., BOND, M. and VUCETICH, M. G. 1981. El Subgrupo Santa Bárbara (Grupo Salta) y sus vertebrados, cronología, paleoambientes y paleobiogeografía. Actas 8th Congreso Geológico Argentino, 3, 743778.
  • PASCUAL, R., GOIN, F. J., KRAUSE, D. W., ORTIZ-JAUREGUIZAR, E. and CARLINI, A. A. 1999. The first gnathic remains of Sudamerica: implications for gondwanathere relationships. Journal of Vertebrate Paleontology, 12, 373382.
  • PASCUAL, R. and ORTIZ-JAUREGUIZAR, E. 1990. Evolving climates and mammal faunas in Cenozoic South America. Journal of Human Evolution, 19, 2360.
  • PASCUAL, R. and ORTIZ-JAUREGUIZAR, E. 1991. El Ciclo Faunístico Cochabambiano (Paleoceno temprano): su incidencia en la historia biogeográfica de los mamíferos neotropicales. In SUÁREZ SORUCO, R. (ed.). Fósiles y Facies de Bolivia I: vertebrados. Revista Técnica de Yacimientos Petrolíferos Fiscales Bolivianos, 12, 559574.
  • PASCUAL, R. and ORTIZ-JAUREGUIZAR, E. 1992. Evolutionary pattern of land mammal faunas during the Late Cretaceous and Paleocene in South America: a comparison with the North American pattern. Annales Zoologici Fennici, 28 (3–4), 245252.
  • PASCUAL, R. and ORTIZ-JAUREGUIZAR, E. 2007. The Gondwanan and South American episodes: two major and unrelated moments in the history of the South American mammals. Journal of Mammalian Evolution, 14 (2), 75137.
  • PASCUAL, R., ORTIZ-JAUREGUIZAR, E. and PRADO, J. L. 1996. Land-mammals: paradigm for Cenozoic South American geobiotic evolution. In ARRATIA, G. (ed.). Contributions of Southern South America to vertebrate paleontology. Münchner Geowissenschaftliche Abhandlungen, 30 (A), 265319.
  • PASCUAL, R., VUCETICH, M. G. and FERNÁNDEZ, J. 1978. Los primeros mamíferos (Notoungulata, Henricosborniidae) de la Formación Mealla (Grupo Salta,Subgrupo Santa Bárbara). Sus implicancias filogenéticas, taxonómicas y cronológicas. Ameghiniana, 15, (3–4), 366390.
  • PAULA COUTO, C. 1952. Fossil mammals from the beginning of the Cenozoic in Brazil. Condylarthra, Litoptema, Xenungulata and Astrapotheria. Bulletin of American Museum of Natural History, 99, 355394.
  • PIANKA, E. R. 2000. Evolutionary ecology, Sixth edition. Benjamin-Cummings, Addison-Wesley-Longman, San Francisco, CA, 528 pp.
  • PORZECANSKI, A. L. and CRACRAFT, J. 2005. Cladistic analysis of distributions and endemism (CADE): using raw distributions of birds to unravel the biogeography of the South American aridlands. Journal of Biogeography, 32, 261275.
  • REGUERO, M. A., MARENSSI, S. A. and SANTILLANA, S. N. 2002. Antarctic Peninsula and South America (Patagonia) Paleogene terrestrial faunas and environments: biogeographic relationships. Palaeogeography Palaeoclimatology Palaeoecology, 179, 189210.
  • RICCOMINI, C. and RODRIGUES-FRANCISCO, B. H. 1992. Idade potássio-argônio do derrame de ankaramito da Bacia de Itaboraí, Rio de Janeiro, Brasil: implicações tectônicas. Congresso Brasileiro de Geologia, 37, 469470.
  • RIGBY, J. K. and WOLBERG, D. L. 1987. The therian mammalian fauna (Campanian) of Quarry 1, Fossil Forest study area, San Juan Basin, New Mexico. Geological Society of America Special Paper, 209, 5179.
  • ROHLF, F. J. 1977. Computational efficiency of agglomerative clustering algorithms. IBM Watson Research Center, 6831, 36 pp.
  • ROSEN, B. R. 1988. From fossils to earth history: applied historical biogeography. 437481. In MYERS, A. A. and GILLER, P. S. (eds). Analytical biogeography: an integrated approach to the study of animal and plant distributions. Chapman and Hall, London, England. 584 pp.
  • RUGGIERO, A. and KITZBERGER, T. 2004. Environmental correlates of mammal species richness in South America: effects of spatial structure, taxonomy and geographic range. Ecography, 27, 401416.
  • Sant’ANNA, L. G. and RICCOMINI, C. 2001. Cimentação hidrotermal em depósitos paleogênicos do Rift Continental do Sudeste do Brasil: mineralogia e relações tectônicas. Revista Brasileira de Geociências, 31(2), 231240.
  • SEMPERE, T., BUTLER, R. F., RICHARDS, D. R., MARSHALL, L. G., SHARP, W. and SWISHER, C. C. III 1997. Stratigraphy and chronology of Upper Cretaceous–lower Paleogene strata in Bolivia and northwest Argentina. Bulletin of the Geological Society of America, 109, 709727.
  • SIGÉ, B. 1971. Les Didelphoidea de Laguna Umayo (formation Vilquechico, Crétacé supérieur, Pérou), et le peuplement marsupial d’Amérique du Sud. Comptes Rendus de l’Académie des Sciences de Paris, 273, 24792481.
  • SIGÉ, B., SEMPERE, T., BUTLER, R., MARSHALL, L. G. and CROCHET, J. Y. 2004. Age and stratigraphic reassessment of the fossil-bearing Laguna Umayo red mudstone unit, SE Peru, from regional stratigraphy, fossil record, and paleomagnetism. Geobios, 37, 771794.
  • SIMPSON, G. G. 1935a. Descriptions of the oldest know South American mammals from Río Chico formation. American Museum Novitates, 793, 125.
  • SIMPSON, G. G. 1935b. Occurrence and relationship of the Río Chico fauna of Patagonia. American Museum Novitates, 818, 121.
  • SIMPSON, G. G. 1948. The Beginning of the Age of Mammals in South America. Bulletin of the American Museum of Natural History, 91, 1232.
  • SIMPSON, G. G. 1978. Early mammals in South America: fact, controversy, and mystery. Proceedings of the American Philosophical Society, 122, 318328.
  • SNEATH, P. H. A. and SOKAL, R. R. 1973. Numerical taxonomy. Freeman, San Francisco. 573 pp.
  • SOMOZA, R., CLADERA, G. and ARCHANGELSKY, S. 1995. Una nueva Tafoflora Paleocena de Chubut Patagonia. Su Edad y Ambiente de depositación. Actas del VI Congreso Argentino de Paleontología y Bioestratigrafía, Trelew, Argentina, 265269.
  • Van VALEN, L. 1988. Paleocene dinosaurs or Cretaceous ungulates in South America. Evolutionary Monographs, 10, 179.
  • WESTERHOLD, T., ROHL, U., RAFFI, I., FORNACIARI, E., MONECHI, S., REALE, V., BOWLES, J. and EVANS, H. F. 2008. Astronomical calibration of the Paleocene time. Palaeogeography, Palaeoclimatology, Palaeoecology, 257(4), 377403.
  • WILSON, G. P., DAS SARMA, D. C. and ANANTHARAMAN, S. 2007. Late Cretaceous Sudamericid Gondwanatherians from India with paleobiogeographic considerations of Gondwanan mammals. Journal of Vertebrate Paleontology, 27(2), 521531.
  • WOODBURNE, M. O. 2004. Late Cretaceous and Cenozoic mammals of North America: biostratigraphy and geochronology. Columbia University Press, New York. xix + 391 pp.
  • WOODBURNE, M. O. 2006. Mammal ages. Stratigraphy, 3(4), 229261.
  • WOODBURNE, M. O. and CASE, J. A. 1996. Dispersal, vicariance, and the post-Gondwana Late Cretaceous to early Tertiary biogeography from South America to Australia. Journal of Mammalian Evolution, 3, 121161.

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Concluding Remarks
  7. Acknowledgments
  8. References
  9. Supporting Information

Table  S1   Data matrix of taxa and faunas/LMAs.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
PALA_835_sm_suppl.doc314KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.