Evolutionary trends of body size and hypsodonty in notoungulates and their probable drivers

Highlights

• Notoungulates increased body sizes and hypsodonty throughout the last 50 Myr.

• Typotherians and toxodonts with low-crowned teeth had higher extinction rates.

• The body mass and the hypsodonty had a coupled long-term evolution.

• The Andean growth appears to be the main driver for the hypsodonty evolution.

Abstract

Members of the Order Notoungulata are among the most diverse and common mammals in South America during the Cenozoic. Several lineages within notoungulates (e.g., suborders Typotheria and Toxodontia) show a tendency for increased body sizes and hypsodonty during the last 50 Myr. However, the timing, evolutionary mode, and drivers of such tendencies are not entirely understood. In this paper, we use an extensive database of notoungulate fossil occurrences and body mass and hypsodonty estimates to characterize the evolutionary mode of these two phenotypic traits over time, test the extent to which several factors (e.g., development of open environments in the south of South America) have influenced it through time, and investigate whether large trait values were selected through elevated origination or reduced extinction rates. Our results demonstrate that most of the major notoungulate clades evolved toward larger body sizes (up 1500 kg) and higher tooth crown, from a small and low-crowned tooth ancestor, in a punctuated mode. We also show that body mass and the hypsodonty in typotherians and toxodonts had a coupled evolutionary history. Species sorting was a relevant macroevolutionary process in some notoungulate clades, as taxa with high teeth crown and body mass had lower extinction rates. Finally, the development of the hypsodonty in notoungulates must reflect repeated and quick instances of adaptive responses to the increased availability of volcanic or other terrigenous particles, within the broad context of the SSA Cenozoic Andean mountain building.

Introduction

During most of the Cenozoic, South America (SA) witnessed the evolution of several groups of mammals in relative geographic isolation (Simpson, 1980). One of the most diverse and common Cenozoic mammal groups that evolved was the notoungulates (Order Notoungulata), achieving an impressive taxonomic diversity (at least 154 genera so far recognized, but now extinct) and morphologic disparity, and filling a wide variety of ecological niches (Croft et al., 2020; Reguero et al., 2010; Reguero and Prevosti, 2010; Scarano et al., 2021). Their relationships within placental mammals remained elusive for more than a century, but recent ancient collagen and protein analyses have identified notoungulates, together with other South American native ungulates like litopterns (Order Litopterna), as the sister taxon of Perissodactyla (Buckley, 2015; Welker et al., 2015). Phylogenetic relationships within notoungulates have supported monophyly and the existence of two relatively well-defined suborders, Toxodontia and Typotheria, and two early-diverging families, Henricosborniidae and Notostylopidae (Billet, 2011; Croft et al., 2020).

High-crowned (hypsodont) teeth are widely found among extant and extinct mammalian herbivores (Damuth and Janis, 2011; Janis and Fortelius, 1988; Kaiser et al., 2013). Notoungulata is one of the clades of SA mammals that evolved high-crowned or hypsodont teeth (Croft et al., 2020; Madden, 2014; Ortiz-Jaureguizar and Cladera, 2006). The development of such kinds of tooth in SA appears precocious compared to other continents (Pascual and Odreman-Rivas, 1971; Patterson and Pascual, 1968). This precocity was first evident as early as the end of the Paleocene. Independent of body size, it was during the Late Oligocene when many families of notoungulates acquired protohypsodont to hypselodont cheek-teeth (Ortiz-Jaureguizar and Cladera, 2006; Pascual and Jaureguizar, 1990; Pascual and Odreman-Rivas, 1971), and the Tinguirirican (Early Oligocene) fauna of SA is the world's oldest fauna dominated by hypsodont herbivores (Flynn et al., 2003). Recent evidence suggests the existence of several periods of relatively intense evolutionary change in hypsodonty in South American mammals (Madden, 2014; Strömberg et al., 2013), and a general trend for an increased hypsodonty throughout the last 40 Myr has been noted in several notoungulates clades (Reguero et al., 2010; Strömberg et al., 2013). Increased hypsodonty appears to be synchronous across several lineages of typotherians by the Tinguirirican South American Land Mammals Age (SALMA) (Reguero et al., 2010). In contrast, hypselodont (i.e., ever-growing teeth) appears to have originated among notoungulates in two pulses: Divisaderan SALMA (late Eocene) in hegetotheriids and mesotheriids and Deseadan SALMA (late Oligocene) in interatheriids and toxodontids (Reguero et al., 2007). Nevertheless, some authors have noted that relatively large errors in hypsodonty estimates are present, and consequently, a potential bias in some of the estimated temporal trends cannot be ruled out (Dunn et al., 2015).

The causes of the development of a precocious hypsodonty in SA notoungulates are not entirely understood and they might be very complex (Dunn et al., 2015; Madden, 2014; Reguero et al., 2010; Strömberg et al., 2013). Its development was initially linked to adaptive shifts given grassland/open habitats expansion (Flynn et al., 2003), consistent with the idea that most modern ungulate clades developed hypsodont teeth in drier and more open habitats (Damuth and Janis, 2011; Janis, 1990; Raia et al., 2010). However, recent works found that grasses only became common in SA after the Miocene, being relatively rare during the Eocene, making it unclear whether hypsodonty evolved in forested or in open but grass-free habitats (Dunn et al., 2015; Raia et al., 2010; Strömberg et al., 2013). In any case, the relevance of exogenous grit or soil's ingestion has driven the necessity to deal with tooth abrasion has been claimed and now are considered a plausible mechanism for hypsodonty evolution in ungulates (Damuth and Janis, 2011; Dunn et al., 2015; Madden, 2014; Semprebon et al., 2019; Strömberg et al., 2013). Indeed, the current view is that hypsodonty represents an adaptation to a worn effect that comprises both diet (e.g., phytoliths in grasses) and environment (dust or grit related to volcanism and erosion) (Damuth and Janis, 2011; Kaiser et al., 2013).

On the other hand, there are possible feedbacks among traits related to hyposodonty. For instance, hypsodont species could be larger-sized than non-hypsodont taxa because the resource that hypsodonty makes available to herbivores is a physiologically demanding one (Raia et al., 2011). Notoungulates show high hypsodonty (brachyodont to hypselodont) and body size (~1–1000 kg) disparity, which likely increases over time (Croft et al., 2020). Body size is one of the essential quantitative traits under evolutionary scrutiny because it influences nearly every aspect of an organism's biology (Peters, 1986). Body size exhibits prominent general trends in both space and time, as the tendency to evolve larger body sizes over evolutionary time (i.e., Cope's rule), or the pattern in which body size among closely related endothermic taxa tends to increase toward colder geographical regions (i.e., Berman's rule) (Ashton et al., 2000; Heim et al., 2015; Hone and Benton, 2005; McNab, 2010; Meiri and Dayan, 2003; Smith et al., 2016). However, none of these trends has been shown to apply generally across taxa (Ashton et al., 2000; Heim et al., 2015; Meiri and Dayan, 2003; Smith et al., 2016). The body size of organisms is often under selection pressure because it provides a direct way to adapt to several different environmental regimes (Kingsolver and Pfennig, 2004; Lyons and Smith, 2010; Smith et al., 2016). Thus, it seems likely that changes in the physical environment should exert a strong influence on the nature and directionality of body size evolution and other phenotypic traits (Clavel and Morlon, 2017; Hunt et al., 2015, Hunt et al., 2010; Hunt and Roy, 2006). In general, both clade's biology and environment must be considered to understand trait evolution fully, and in general, macroevolutionary patterns. Accordingly, characterizing first-order patterns is essential for understanding the potential underlying causes that have shaped traits, like body size or hypsodonty, over evolutionary time (Hunt et al., 2015; Hunt and Roy, 2006; Smith et al., 2016). Characterizing the mode and tempo of the evolutionary change is a major theme in macroevolution since Simpson's earliest days (Simpson, 1944). Nowadays, it is possible to characterize trait evolution based on the fossil record, as each mode of evolution can be expressed as a statistical model (Hunt, 2006a). Three general models have become standard in attempts to understand the nature of evolutionary divergence in fossil lineages: directional change, unbiased random walk, and stasis, and although simplified, these modes of change are useful abstractions that distinguish fundamentally different kinds of evolutionary dynamics (Hunt, 2007, Hunt, 2006a; Hunt et al., 2015; Hunt and Rabosky, 2014).

Even though the unique hypsodonty patterns exhibited by several lineages of notoungulates, to our knowledge, studies dealing with their evolutionary mode of trait evolution are still lacking, and the same situation is uncovered for the body size. Two primary drivers for evolutionary patterns of hypsodonty in notoungulates have been proposed, the tectonic evolution of the Andean orogen (as well as their concomitant complex volcanic history) and the development of open habitats through changes in the disposal of exogenous grit or soil (Gomes Rodrigues et al., 2017; Kohn et al., 2015; Madden, 2014). However, the few studies explicitly testing them have focused on the effects of the availability of open habitats over notoungulate hypsodonty (Strömberg et al., 2013). On the other hand, it has been recently argued that the evolutionary patterns of body size in Protypotherium, a diverse genus of typotherians, could be correlated with global temperature trends (Scarano et al., 2021). Therefore, the potential drivers of the evolutionary change of body mass and hypsodonty through time appear to be mostly related to abiotic or environmental changes. In contrast, additional potential drivers, including feedbacks among own-clade traits, have not been considered, and to our knowledge, no studies have examined multiple hypotheses simultaneously. The last is in need, given the complex interactions of biotic and abiotic factors driving long-term evolutionary patterns (Fraser et al., 2020; Lehtonen et al., 2017; Solórzano et al., 2020).

In the present work, we compiled a comprehensive database of notoungulate fossil species occurrences, dental measurements (as a way to estimate the body mass of each species), and hypsodonty categories from the literature. Within a statistical framework, we analyzed these data to characterize the mode of evolution of phenotypic traits (body size and hypsodonty) over time. The more considerable diversity of notoungulates concentrates toward the south of South America. Therefore, we also test the extent to which assorted environmental factors (e.g., paleobotanical changes) have influenced these traits (body size and hypsodonty). Given that both traits could fluctuate in both space and time, we also test the correlation between body size, hypsodonty, and geography (e.g., latitude, longitude as predictive variables). Moreover, although previous works have advocated an apparent increase of body size and hypsodonty in notoungulates through time (Madden, 2014; Reguero et al., 2010; Strömberg et al., 2013), it is necessary to distinguish whether these general trends are generated by a non-directional diffusive process or by an active directional evolutionary process due to selection for large body size or hypselodont species (Huang et al., 2017; McShea, 1994). Therefore, we also investigate whether large trait values (body size or hypsodonty) were selected through differential origination or extinction rates (Jablonski, 2008).