• Adaptive radiation;
  • Fossils;
  • Herbivory;
  • Life-History Evolution;
  • Morphological Evolution


  1. Top of page
  2. Abstract
  5. Supporting Information

In a recent paper, Eronen et al. (2010; hereafter EEFJ) observe differences in occlusal morphology, tooth crown height, and mesowear pattern between populations of the Miocene tridactyl equid Anchitherium from Spain and Germany, proposing that Spanish Anchitherium underwent adaptive evolution to local or regional arid conditions. However, these authors do not take into account the actual diversity of Iberian representatives of Anchitherium, or the fact that the Spanish fossils cover a wider temporal and geographical range than those from Germany. For these reasons, we suggest that their subsequent statistical work should be reconsidered.

To test their hypotheses, EEFJ “examine Anchitherium teeth collected from Spain and Germany from early Middle Miocene (17–14 m. y.), a time interval when Anchitherium was in its prime and the dispersal of hipparionine horses to the Old World was millions of years away” (Eronen et al. 2010, p. 399). However, this assumption is not justified in relation to the material from Spain, because the Spanish anchitheriines selected by EEFJ come from four localities ranging from basal MN 5 to MN 7/8 (early to late Middle Miocene) and three different continental basins (Madrid, Loranca, and Duero; see Fig. 1A). Consequently,


Figure 1. (A) Chronostratigraphic and geographic position of the Spanish localities with Anchitherium used by EEFJ, and calibrated phylogenetic hypothesis of the Spanish Anchitherium (modified from Sánchez et al. 1998). Black boxes in the phylogenetic tree indicate the chronostratigraphic distribution of each of the species present in the Spanish sites selected by EEFJ. Only these sites are highlighted between brackets next to the specific names. Anchitherium hippoides is a French species that entered the Iberian Peninsula as an immigrant in the late MN 6 (Hernández-Fernádez et al. 2003). Both the French (dotted gray box) and Spanish ranges (white box) of this species are displayed. The Miocene Climatic Optimum (MCO, 17–15 my, Zachos et al. 2001) and the Middle Miocene Climatic Transition (MMCT, 14.2–13.8 my, Shevenell et al. 2004) are represented by gray bands, including their corresponding peaks (dark gray within the MCO, around 16.8–16.2 my following Shevenell and Kennett 2004, and dotted line within the MMCT for the minimum temperatures estimated around 13.9 my by Holbourn et al. 2005). (B) Life reconstruction of Anchitherium cursor. Illustration by Mauricio Antón. Abbreviations: ALH = Alhambra-Túneles (Madrid Province, Madrid Basin); Fr. = France; LRT = La Retama (Cuenca Province, Loranca Basin); MV = Montejo de la Vega (Segovia Province, Duero Basin); PV = Puente de Vallecas (Madrid Province; Madrid Basin); Sp. = Spain.

Download figure to PowerPoint

  • 1
    EEFJ do not consider that the palaeoecological signal provided by the Spanish material could actually be reflecting environmental shifts associated with both the changing climatic conditions throughout the Middle Miocene (Zachos et al. 2001; Shevenell and Kennett 2004; Shevenell et al. 2004; Holbourn et al. 2005) and/or environmental differences between the different Spanish closed basins.
  • 2
    Anchitherium was not in “its prime” as EEFJ state, as this genus underwent two important consecutive radiation events (Fig. 1B) in the Iberian Peninsula (Sánchez et al. 1998), and EEFJ selected four species belonging to both radiations.
  • 3
    The dispersal of hipparionine horses to the Old World was not millions of years away, as the youngest remains used in the analysis come from the MN 7/8, around 1.65 million years before the arrival of Hipparion.

By contrast, the German Anchitherium material analyzed by EEFJ is very homogeneous, as it comes from a single locality, Sandelzhausen (MN 5), and represents a single species, A. aurelianense. Hence the palaeoecological signal provided by the German Anchitheriinae is very local, and corresponds to a very specific part of its temporal distribution. Therefore it seems clear that the characteristics of the Spanish and German samples used in the analysis are not suited to testing the hypotheses proposed.

EEFJ consider the analyzed Spanish anchitheriine teeth as belonging to three taxa (Eronen et al. 2010; Table 1, p. 401), but it is apparent that they did not completely consider the systematic revision of the Spanish Anchitherium carried out by Sánchez et al. (1998) on the same Spanish collections. The latter authors pointed out the existence of two lineages and a number of new species of Anchitherium (Fig. 1B). Although EEFJ recognized A. castellanum in La Retama, they did not use the specific names for the fossils from Puente de Vallecas (A. matritense), Alhambra-Túneles (A. cursor), and Montejo de la Vega (Anchitherium sp.). In the light of these points, the conclusions of EEFJ, although valid in a broad sense (i.e., Spanish vs. Central-European anchitheriines), cannot be supported at a species-specific scale as none of the Spanish specimens belong to A. aurelianense. In Spain, this species has been only found in the MN 3-MN 4, and outside the Madrid Basin (Sánchez et al. 1998). In addition, the sample of EEFJ mixes different evolutionary stages, which actually belong to the two lineages recognized by Sánchez et al. (1998) thus, A. castellanum belongs to the more primitive clade of anchitheriines, whereas A. matritense and A. cursor are representative of the more derived lineage present in the Madrid Basin.

Table 1.  Teeth examined by EEFJ, reordered with the correct taxonomic determinations.
LocalityCountrySpeciesParacone angleProtocone angleMetacone angleHypocone angleMesial top angleDistal top angle
SandelzhausenGermanyA. aurelianense50.742.1350.2151.0787.278.8
SandelzhausenGermanyA. aurelianense50.8740.9254.4846.9388.378.6
SandelzhausenGermanyA. aurelianense58.5651.5956.2953.569.970.3
SandelzhausenGermanyA. aurelianense50.5543.8454.5248.9585.776.6
SandelzhausenGermanyA. aurelianense46.4850.3154.1147.4383.378.5
SandelzhausenGermanyA. aurelianense40.7546.9333.1847.1292.499.7
SandelzhausenGermanyA. aurelianense48.6941.2751.0345.4490.183.6
SandelzhausenGermanyA. aurelianense52.2756.2348.651.0571.580.4
SandelzhausenGermanyA. aurelianense48.4763.7751.1846.8267.882
SandelzhausenGermanyA. aurelianense57.895142.456.2371.281.4
SandelzhausenGermanyA. aurelianense49.6739.3848.6651.669179.7
La RetamaSpainA. castellanum66.1648.1268.2144.2665.878.5
La RetamaSpainA. castellanum58.543.6757.3249.4177.967.8
La RetamaSpainA. castellanum63.3350.2662.8350.5166.568.9
La RetamaSpainA. castellanum63.5446.0760.6654.4170.469.1
Puente de VallecasSpainA. matritense63.4451.0458.6139.8765.681.6
Puente de VallecasSpainA. matritense67.743.2260.8642.8669.176.3
Puente de VallecasSpainA. matritense59.549.7753.0647.3670.879.6
Puente de VallecasSpainA. matritense63.0343.4262.5845.5473.671.9
Puente de VallecasSpainA. matritense61.4358.7159.4259.2359.961.4
Puente de VallecasSpainA. matritense54.3945.3551.8738.8580.386.3
Puente de VallecasSpainA. matritense60.1843.7654.8941.576.184.5
Puente de VallecasSpainA. matritense60.6144.0754.0753.6375.476.8
Puente de VallecasSpainA. matritense56.2253.949.6248.4369.984.4
Puente de VallecasSpainA. matritense52.0847.1747.1858.3780.853.5
AlhambraSpainA. cursor67.2952.3563.3852.7860.463.9
AlhambraSpainA. cursor50.1847.1856.49-82.7-
Montejo de la VegaSpainAnchitherium sp.58.4348.8657.3446.8172.875.9

To consider the actual taxonomy of the Spanish Anchitherium (Table 1) we corrected the Table 1 of EEFJ, and obtained new plots of the individual cusp slopes (Fig. S1) and top-angles (Fig. S2). Table S1 shows the significance tests. Clearly the new rearrangement of the Spanish data when compared with the single German population gives the same results as those obtained by EEFJ, but we can now see a degree of heterogeneity and a different pattern within the Spanish material, which emphasizes differences in tooth volume by measuring the steepness of the cusp slopes (Figs. S1 and S2). This would indicate subtle differences in hypsodonty according to the method used by EEFJ, so that A. castellanum shows more development toward increased hypsodonty than A. matritense.

Because EEFJ do not provide the data for their mesowear analysis, we cannot test whether these subtle differences in hypsodonty also correspond with visible differences in diet and the relative abrasiveness of the forage. As observed in other contemporaneous Spanish brachydont taxa (DeMiguel et al. 2010) one would expect that the Anchitherium species from Spain employed a wider range of dietary strategies (from strict browsing to grass-dominated mixed feeding) and encountered variations in the abrasiveness of their diet, but this is not evident in the hierarchical cluster analysis of EEFJ. In this respect, the decision to gather together all the Spanish specimens on the assumption that they belong to a single population (Anchi-S) may lead to the erroneous conclusion that dietary homogeneity was exhibited by the genus Anchitherium in the Middle Miocene of Spain.

These adaptive changes, underlined by the differences in hypsodonty, were subtle and it may be difficult to propose a single explanation. Nonetheless, and because the analyzed samples belong to different closed basins, local differences in the environmental conditions may provide an explanation for these changes. However, because samples represent different temporal intervals, these adaptive differences could also reflect environmental changes explicable in terms of changing climatic conditions. In this regard, it should be pointed out that La Retama site (A. castellanum sample) was formed during the global warming of the Miocene Climatic Optimum (MCO; after the 16.8–16.2 million years (my) peak; Shevenell and Kennett 2004), whereas the age of Puente de Vallecas (A. matritense sample) is within the global cooling of the middle Miocene climatic transition (MMCT; ∼14.0 my, see Fig. 1A). This latter process is considered the most abrupt and important change in the temperatures of the Miocene, having enormous consequences on the Spanish ecosystems. In view of its coincidence with both the changes in hypsodonty and the evolutionary scenario suggested by Sánchez et al. (1998) for the Spanish Anchitherium, it is feasible that these phenomena are strongly related.

To sum up, the study of EEFJ on the dietary reconstruction and dental evolution of the Spanish Anchitheriinae remains an interesting contribution. They agree, in general terms, with the scenario proposed by Sánchez et al. (1998), but exhibit mistakes related to the systematics and age of the Spanish samples of Anchitherium included in the analysis. This highlights the need for an extensive knowledge of the taxonomy, systematics, and temporal distribution of a group in any kind of palaeoecological–palaeoclimatic study. Thus, it would be especially valuable for EEFJ to complete the study by considering the actual diversity of the Spanish species of Anchitherium. This would improve our understanding of the ecology and evolution of this Miocene horse, which has revealed as having a much more complex evolutionary history than classically supposed.

Associate Editor: G. Hunt


  1. Top of page
  2. Abstract
  5. Supporting Information

This study is part of the research projects CGL2008-00034/BTE, CGL2008-00325/BTE, CGL2008-05813-C02-01/BTE, and CGL2007-60790/BTE (Dirección General de Investigación-MCI), and the Research Groups CAM-UCM 910607, and EO5 (Gobierno de Aragón). MJS is a contracted researcher within the “Ramón y Cajal” Program (Ministerio de Ciencia e Innovación). IMS acknowledges a CSIC JAE-Doc contract from the program “Junta para la Ampliación de Estudios,” co-funded by the FSE. DD is a postdoctoral researcher in the ICP. We thank Dr. Alan Turner (Liverpool John Moores University) for revising the English. We thank Mauricio Antón for his illustration of Anchitherium cursor.


  1. Top of page
  2. Abstract
  5. Supporting Information
  • DeMiguel, D., B. Azanza, and J. Morales. 2010. Trophic flexibility within the oldest Cervidae lineage to persist through the Miocene Climatic Optimum. Paleogeogr. Paleoclimatol. Paleoecol. 289:8192.
  • Eronen, J. T., A. R. Evans, M. Fortelius, and J. Jernvall. 2010. The impact of regional climate on the evolution of mammals: a case study using fossil horses. Evolution 64:398408.
  • Hernández-Fernández, M., M. J. Salesa, I. M. Sánchez, and J. Morales. 2003. Paleoecología del género Anchitherium (Equidae, Perissodactyla, Mammalia) en España: evidencias a partir de las faunas de macromamíferos. ColPa, 1:253280.
  • Holbourn, A., W. Kuhnt, M. Schultz, and H. Erlenkeuser. 2005. Impacts of orbital forcing and atmospheric carbon dioxide on Miocene ice- sheet expansion. Nature 438:483487.
  • Sánchez, I. M., M. J. Salesa, and J. Morales. 1998. Revisión Sistemática del género Anchitherium Meyer, 1834 (Equidae; Perrissodactyla) en España. Estudios Geol. 54:3963.
  • Shevenell, A. E., and J. P. Kennett. 2004. Paleoceanographic change during the middle Miocene climate revolution: an Antarctic stable isotope perspective. Geophysical Monograph 18;151:235251.
  • Shevenell, A. E., J. P. Kennett, and D. W. Lea. 2004. Middle Miocene Southern Ocean cooling and Antarctic cryosphere expansion. Science 305:17661770.
  • Zachos, J., Pagani, M., Sloan, L., Thomas, E., and K. Billups. 2001. Trends, rhythms, and aberrations in global climate 65 Ma to present. Science 292:686693.

Supporting Information

  1. Top of page
  2. Abstract
  5. Supporting Information

Figure S1. Modified Figure 3 of EEFJ.

Figure S2. Modified Figure 4 of EEFJ.

Table S1. Statistical significance level of the nonparametric statistical tests (Chi-square Kruskal--Wallis) used to examine the differences in cusps slopes and top-angles between the Spanish and German samples of Anchitherium teeth.

EVO_1219_sm_Figure_S1.pdf1436KSupporting info item
EVO_1219_sm_Figure_S2.pdf832KSupporting info item
EVO_1219_sm_Table_S1.pdf528KSupporting 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.