1 The evolution of a single toe in horses: causes, consequences, and the way forward
2 Brianna K. McHorse,1,*,†,‡ Andrew A. Biewener,*,† and Stephanie E. Pierce*,‡
3
4 *Department of Organismic and Evolutionary Biology, Harvard University, Cambridge, MA 02138, USA;
5 †Concord Field Station, Harvard University, Bedford, MA 01730, USA; ‡Museum of Comparative Zoology,
6 Harvard University, Cambridge, MA 02138, USA
7
9
10 Abstract
11 Horses are a classic example of macroevolution in three major traits—large body size, tall-
12 crowned teeth (hypsodonty), and a single toe (monodactyly)—but how and why monodactyly evolved is
13 still poorly understood. Existing hypotheses usually connect digit reduction in horses to the spread and
14 eventual dominance of open-habitat grasslands, which took over from forests during the Cenozoic; digit
15 reduction has been argued to be an adaptation for speed, locomotor economy, stability, and/or
16 increased body size. In this review, we assess the evidence for these (not necessarily mutually exclusive)
17 hypotheses from a variety of related fields, including paleoecology, phylogenetic comparative methods,
18 and biomechanics. Convergent evolution of digit reduction, including in litopterns and artiodactyls, is
19 also considered. We find it unlikely that a single evolutionary driver was responsible for the evolution of
20 monodactyly, because changes in body size, foot posture, habitat, and substrate are frequently found to
21 influence one another (and to connect to broader potential drivers, such as changing climate). We
22 conclude with suggestions for future research to help untangle the complex dynamics of this remarkable
23 morphological change in extinct horses. A path forward should combine regional paleoecology studies, Evolution of a single toe in horses 2
24 quantitative biomechanical work, and make use of convergence and modern analogs to estimate the
25 relative contributions of potential evolutionary drivers for digit reduction.
26
27 Total words in text: 7608
28
29 Introduction
30 Horse evolution and grasslands
31 Horses are the only living members of the family Equidae, which today comprises just six species
32 in the genus Equus (including zebras, asses, and caballine horses, the group to which domestic horses
33 belong). In contrast to today’s paucity of species, the equid fossil record includes nearly 50 genera and
34 hundreds of species over the last 58 million years (MacFadden 1994). The earliest equids were only dog-
35 sized, with four toes on the foreleg and three on the hind leg (MacFadden 1994). Today’s horses are
36 large, long-legged grazers with a single toe on each leg, which is enclosed in a hard hoof. An enlarged
37 third digit makes up the bulk of the distal limb, with considerably reduced metapodials II and IV present
38 as splint bones fused to the center metapodial (Figure 1). Recent work has shown that vestiges of digits I
39 and V may still be present as ridges and wings in the proximal metapodial and distal phalanx (Solounias
40 et al. 2018). Despite their large size and long, slender limbs, horses are considerably athletic, reaching a
41 recorded top racing speed of 70 km/hr (“Fastest speed for a race horse” 2019); the highest jump
42 recorded by a domestic horse and rider is 2.47 m (“Highest jump by a horse” 2019). That horses can
43 accomplish such feats on a single toe, which evolved millions of years prior to human influence, is
44 remarkable.
45 Fossil horses played a critical role in both supporting Darwin’s theory of evolution and, later, the
46 Modern Synthesis (Simpson 1951). In the 1870s, O.C. Marsh had made a considerable collection of fossil
47 horses, which he then arranged into a series of small to large, three-toed to one toe, low-crowned teeth Evolution of a single toe in horses 3
48 to high-crowned teeth (Marsh 1874). This proposed evolutionary series was so striking for its time that
49 after seeing it, T.H. Huxley, “Darwin’s Bulldog,” rewrote an address to be given at the New York
50 Academy of Sciences to include these fossil horses as evidence of evolution (Schuchert 1940). At the
51 time, orthogenesis—an evolutionary “progression” in a straight line towards some ideal form—was a
52 popular conception of evolution, and this arrangement of horses supported that view (Figure 2). Thus
53 the classic story of horse evolution was formed: as grasslands took over from forests, the horse
54 gradually evolved larger body size (perhaps to better defend against predators), taller-crowned teeth to
55 handle abrasive grasses, and long, monodactyl limbs to race away from predators in their newly open
56 habitat (Figure 2; Matthew 1926).
57 Despite subsequent recognition that equid evolution was in fact more like a bush than a straight
58 line (Simpson 1951; MacFadden 1994), it is still portrayed in a linear fashion in many museums and
59 textbooks (MacFadden et al. 2012). Some trends in equid evolution do appear to exist by gestalt—
60 today’s horses are indeed much larger, hypsodont, and have reduced digits relative to the earliest
61 horses. Monodactyly had two separate evolutions, one in the Dinohippus/Equus lineage and one in the
62 Pliohippus/Astrohippus lineage, strongly suggesting at least some adaptive utility and selection for this
63 condition. It therefore requires careful attention to discuss the evolution of horses without slipping into
64 verbal orthogenesis by drawing a straight line between the earliest horse and the lone surviving genus
65 today, particularly given that trends of digit reduction and increasing hypsodonty do exist in at least
66 some parts of the horse tree (Janis 2007). But evidence from diet, habitat, tooth morphology, and digit
67 state do not match the orthogenetic pattern: decreasing body size was common in lineages such as the
68 Archaeohippus or Nannippus; not all tridactyl horses browsed; and not all hypsodont, monodactyl
69 horses grazed (MacFadden 1994; MacFadden et al. 2012).
70 Beyond the pattern itself, the classic explanations for why horses evolved the way they did is
71 tremendously “sticky” (Schimel 2012). Long after the complexity of the equid tree and the nonlinearity Evolution of a single toe in horses 4
72 of trait evolution was acknowledged, the initial explanations for each horse trait still held the weight of
73 established fact rather than reasonable hypothesis. The evolutionary story of horses has seen several
74 advances in understanding over the last few decades, particularly as powerful quantitative methods
75 emerge and we accumulate more available specimens through fieldwork, museum cataloguing, and
76 especially digitization (Marshall et al. 2018). With these new data and methods, untested explanations
77 for horse trait evolution have been challenged one by one.
78 The simple causal relationship between abrasive grass and hypsodonty has been shown to be
79 complicated, with grasslands predating hypsodonty by at least four million years in horses, rodents, and
80 lagomorphs (Strömberg 2002, 2006; Jardine et al. 2012). Tooth mesowear, a macroscopic measure of
81 tooth wear that can record information about diet, is highly variable within fossil horse populations and
82 does not always match up directly with grasslands and hypsodonty (Mihlbachler et al. 2011);
83 furthermore, fresh grazing can cause mesowear similar to browsing (Winkler et al. 2019). In extant taxa,
84 hypsodonty correlates more with habitat openness (Mendoza and Palmqvist 2008)than with the
85 proportion of grass consumed, and feeding height (which relates to the amount of soil grit consumed)
86 drives microwear more than diet (Mainland 2003). The study of tooth enamel isotopes has also
87 complicated the relationship between hypsodonty and diet; for example, in one locality, the hypsodont
88 horse species were likely browsers while the species with low-crowned teeth were consuming more
89 grasses (MacFadden et al. 1999), and individual variation in isotope values can be high in large
90 herbivores (Green et al. 2018). Hypsodonty seems to be driven mostly by grit (via phytolith, dirt, or
91 volcanic ash), not grass alone, and evolved under much less of a straightforward evolutionary arms race
92 than initially thought.
93 Increasing body size, which was initially explained as a defense mechanism against predation
94 (Matthew 1926), has also been suggested as an example of Cope’s rule (that lineages tend to increase in
95 maximum body size through time) in equids (Martin 2018), perhaps as a result of ecological Evolution of a single toe in horses 5
96 specialization (Raia et al. 2012). Others have argued that because grass is generally less nutritious than
97 browse (but see Codron et al. 2007), larger body size was beneficial because it increased total digestive
98 capacity, thus allowing the animal to process larger quantities of low-quality food (Demment and Van
99 Soest 1985; Lovegrove and Mowoe 2013). A recent study showed strong evidence for transitions to an
100 unguligrade foot posture being associated with rapid increases in body size, and rate of body size
101 evolution, across mammals (Kubo et al. 2019), pointing to the possibility that unguligrady supports
102 larger body sizes. Kubo et al. (2019) suggested that larger body sizes may provide a release from higher
103 levels of predation, again connecting predator pressures to horse evolution. However, many equid
104 lineages retained similar body sizes through evolutionary time or even became smaller (MacFadden
105 1994), irrespective of expanding grasslands. Recent work suggested that more than 90% of changes in
106 horse body size can be explained simply by diffusion, a random walk of evolution, rather than
107 competition for niches (Shoemaker et al. 2013), so the pattern of increasing body mass may not be a
108 trend at all.
109
110 Modern hypotheses for digit reduction
111 Like hypsodonty and body size, the story of digit reduction is complicated. The classic
112 explanation was that high speeds were necessary for predator escape in open grasslands, with reduced
113 digits on elongated limbs providing this speed (Matthew 1926; Simpson 1951). But high-speed pursuit
114 predators such as wolves did not evolve until approximately 20 million years after many ungulates,
115 including equids, evolved lengthened limbs and reduced digits (Janis and Wilhelm 1993), providing
116 evidence against this hypothesis. Subsequent years have seen three other major hypotheses about the
117 proximate driving force behind digit reduction in horses:
118 1) The locomotor economy hypothesis: Open, arid grasslands required longer travel distances to
119 access patchy resources, such as water, and elongated limbs decreased the energetic cost of Evolution of a single toe in horses 6
120 locomotion by increasing stride length (Janis and Wilhelm 1993). While reduced digits were not
121 explicitly discussed in the referenced paper, reducing mass in the distal limb would decrease
122 moment of inertia and thus the energetic cost of swinging the leg (Hildebrand 1960; Myers and
123 Steudel 1985; Browning et al. 2007; Kilbourne et al. 2016), and has been argued to be a driver of
124 digit reduction in archosaurs (de Bakker et al. 2013).
125 2) The stability hypothesis: Forests required lateral dodging movements on soft ground, whereas
126 grasslands required high-speed, straight-line movements on hard ground; a tridactyl and
127 monodactyl foot were respectively better suited to stability in those environments (Shotwell 1961).
128 Effectively, Shotwell proposed two separate hypotheses: 2a) the soft substrate stability hypothesis,
129 that the tridactyl foot is adapted for stability in lateral dodges on soft ground, and 2b) the hard
130 substrate stability/speed hypothesis, that the monodactyl foot is adapted for stability and speed in
131 straight lines on hard ground. The superior speed of the monodactyl foot on hard ground was also
132 proposed by Matthew (1926), who focused on the rigid hoof and spring-like tendon-ligament system
133 of extant equids.
134 3) The body size hypothesis: Evolutionary increases in body mass produced greater bending forces on
135 the limbs, and a single digit resists bending forces better than several smaller digits of the same total
136 size (Thomason 1986).
137 None of these hypotheses are mutually exclusive. The first two assume that grasslands act as
138 the ultimate driver of digit reduction (via the proximate causes of increased locomotor demands and
139 changing substrate conditions, respectively). The third hypothesis suggests body mass as a proximate
140 cause. As discussed previously, proposed drivers of body mass itself in horses range from evolutionary
141 diffusion (Shoemaker et al. 2013) to grasslands (Illius and Gordon 1992), consistent with the first two
142 hypotheses. Body mass increase in response to the evolution of grasslands might also partially be
143 related to cooling climate (Lovegrove and Mowoe 2013), but climate cooling should also be considered Evolution of a single toe in horses 7
144 as one of many potential drivers of grasslands themselves (Strömberg 2011). Therefore, if the body size
145 hypothesis is correct, the ultimate driver of digit reduction could be any or a combination of other
146 changes in forage and climate.
147 In this review, we survey what evidence exists to support or refute these digit reduction
148 hypotheses. First, we give a brief overview of horse evolution with a focus on digits. Next, we discuss an
149 analytical method for quantifying the degree of digit reduction to provide a continuous metric for
150 evolutionary analyses. We follow this with a review of research from biomechanics, macroevolution, and
151 other subdisciplines that has brought new insights into digit reduction in recent years. Finally, we
152 conclude with a discussion of critical gaps in our understanding and make suggestions for avenues of
153 future study.
154
155 Overview of horse digit evolution
156 The phylogenetic relationships of the earliest equids have seen considerable changes since
157 Hyracotherium/Eohippus was universally considered the first horse (Figure 3). Recent phylogenetic work
158 has split Hyracotherium into H. leporinum, now considered a basal palaeothere (outside of Equidae), and
159 an array of new genera for basal true equids, including Sifrhippus and Arenahippus (Froehlich 2002).
160 Another recent phylogenetic analysis has placed Ghazijhippus (found in Pakistan) and Cymbalophus and
161 Pliolophus (found in Europe) at the base of the equid tree, basal to the North American Sifrhippus and
162 Arenahippus (Bai et al. 2018). Regardless which genus is most basal, the earliest equids were small, dog-
163 sized creatures that had four digits on the forefoot and three on the hind foot (semi-tetradactyl). They
164 had low-crowned (brachydont) teeth indicative of a diet of fruits and soft leaves, and although they
165 seem “primitive” relative to extant horses, a remarkably complete skeleton of the basal horse
166 Arenahippus grangeri (previously H. grangeri; considered a junior synonym of Sifrhippus by Secord et al.
167 2012, but left separate here) shows that it was fairly derived for the time in its foot posture Evolution of a single toe in horses 8
168 (subunguligrade) and somewhat elongate metapodials (Kitts 1956; Wood et al. 2011). Although the
169 shoulder and hip joints had considerable range of movement, the distal limb in A. grangeri was already
170 primarily restricted to parasagittal motion, as in later equids (Wood et al. 2011).
171 Coeval with this Arenahippus was another, similarly small equid, Orohippus, which had even
172 more restricted movements in the distal limb due to more stable carpal and tarsal articulations. Wood
173 et al. (2011) hypothesized that Orohippus may have occupied more open terrain than A. grangeri, which
174 was found in tropical forests—reminiscent of Shotwell’s (1961) stability vs. substrate hypotheses. Wood
175 et al. (2011) suggest the following chain of events: as climate changed in the Paleogene (66-12 Ma),
176 equid diets shifted from high-quality fruits and leaves to lower-quality browse and graze; such a dietary
177 shift drove increases in body size (consistent with results from Secord et al. (2012)) to allow processing
178 greater amounts of food; and finally, increased body size drove a need for more centrally-located,
179 upright limbs (reminiscent of Camp and Smith’s (1942) argument for lineage-scale digit reduction).
180 Like earlier equids, the three-toed (tridactyl) species Mesohippus, Miohippus, and Anchitherium,
181 were also likely subunguligrade, with all distal phalanges contacting the ground and supported by a foot
182 pad (Camp and Smith 1942; Sondaar 1968; Thomason 1986). However, relative to earlier equids, their
183 limbs became more restricted to a pendulum-like motion in the parasagittal plane via limb bone fusion
184 (radius-ulna, tibia-fibula) and changes in joint articulations (Sondaar, 1968). Along with increasingly
185 parasagittal motion came the lengthening of the limb, particularly distally. In later lineages, beginning
186 with Parahippus at the base of the grazing radiation, limb elongation continued, and the lateral digits
187 were reduced; the side toes likely did not touch the ground at rest (Sondaar, 1968). In the monodactyl
188 or nearly-monodactyl lineages, such as Pliohippus and Equus, a tendon-and-ligament suspensory
189 apparatus and a ‘springing’ foot evolved, with markedly elongate phalanges and considerable elastic
190 energy storage (Biewener 1998) that may have benefited them on hard ground in open grasslands
191 (Matthew 1926; Sondaar 1968; Janis and Wilhelm 1993). Evolution of a single toe in horses 9
192
193 Quantifying digit reduction
194 Until recently, one challenge of studying digit reduction in horses was the lack of a quantitative
195 way to measure digit reduction. The discrete categories of semi-tetradactyl (four toes in front and three
196 behind), tridactyl (three toes), and monodactyl (one toe) are useful, but fail to capture a wide variety of
197 morphological (and probably functional) diversity throughout the main body of the equid phylogeny
198 (Figure 3). Furthermore, many modern analyses require continuous variables to reconstruct the mode or
199 rate of evolution (O’Meara and Beaulieu 2014). We addressed this gap in two recent papers, where we
200 introduced the Toe Reduction Index (TRI), a continuous measure of digit reduction for perissodactyls
201 (McHorse et al. 2017; Parker et al. 2018). TRI is measured as the ratio of side digit length to center digit
202 length in the proximal phalanx (Equation 1, Figure 4), taking the average of side digits if they are both
203 available, and is best calculated for each individual before averaging across a species or genus. The index
204 ranges from 0 (no side digits, as in Equus) to 1 (all digits equal in length):
205
푚푒푎푛(푃푃푙푒푛𝑔푡ℎ퐼퐼,푃푃푙푒푛𝑔푡ℎ퐼푉) 206 Equation 1. 푇푅퐼 = 푠푝푒푐푖푒푠 푚푒푎푛 ( ) 푃푃푙푒푛𝑔푡ℎ퐼퐼퐼
207
208 where PPlength refers to the maximum articular length of the proximal phalanx in digit II, IV, or III
209 according to the subscript; the mean of PPlength is first taken for all available side digits (II and IV), then
210 divided by the mean of PPlength for digit III. This provides the individual-level TRI, which is then
211 averaged across all individuals in a species to provide a species-level TRI (Equation 1). Values greater
212 than 1 are theoretically possible and would correspond to lateral digits greater in length than the center
213 digit; however, this seems unlikely to occur. Evolution of a single toe in horses 10
214 With the Toe Reduction Index, we can quantitatively represent the real morphological variation
215 present in equid digits (Figure 5). Whereas previously all tridactyl horses would be coded the same in
216 categorical data, TRI values range from nearly 1 (all three digits of equal size) to less than 0.3 (side digits
217 ⅓ the size of the center digit), illuminating variation that was previously unavailable to quantitative
218 analyses. These new data make it possible to address questions such as whether digit reduction
219 correlates with changes in other traits, e.g., hypsodonty or body mass (Parker et al. 2018). Furthermore,
220 TRI can be used in the future to explore digit reduction in a variety of other groups, including
221 artiodactyls, with appropriate modification to account for paraxonic symmetry vs. mesaxonic symmetry
222 (i.e., artiodactyls have symmetrical enlarged digits III and IV with the axis of symmetry running between
223 them, whereas TRI was developed for a single, symmetrical center digit III). A TRI dataset expanded to
224 other taxa would open up the possibility of both more quantitative taxon-specific studies and more
225 broadly comparative studies.
226
227 Biomechanical investigation of digit reduction
228 As terrestrial quadrupeds, horses primarily use their limbs to interact with the environment
229 through locomotion. The forces that act on bones during an animal’s life can be a powerful source of
230 selection; the geometry of bones can often indicate, for example, locomotor style or other functional
231 uses for that part of the body (Swartz et al. 1992; Anyonge 1996; Doube et al. 2018). In domestic horses,
232 many biomechanical studies have linked skeletal morphology to performance in competition,
233 connecting form to function (Barrey et al. 2002; Gnagey et al. 2006; Weller et al. 2006; Hobbs et al.
234 2010; Kristjansson et al. 2016). An extra load of just 2.4 kg on a horse’s distal limb has been shown to
235 increase cost of transport by nearly 7% (Wickler et al. 2004), providing a direct connection to the
236 locomotor economy hypothesis of digit reduction. Evolution of a single toe in horses 11
237 Studies examining the biomechanical and physiological consequences of limb morphology in
238 extant horses rarely connect to the fossil record and to equid evolution, but a combination of these
239 disciplines offers considerable insight into outstanding questions like the driver of equid digit reduction.
240 Although biomechanical performance data cannot be obtained for extinct animals, musculoskeletal
241 modeling is a powerful tool to reconstruct soft-tissue dynamics, forces, and ultimately provide insight
242 into performance in extinct species (Hutchinson and Garcia 2002; Pierce et al. 2012, 2013; Nyakatura et
243 al. 2019). Such studies usually make use of detailed skeletal data from extinct species, sometimes
244 combining it with experimental biomechanical data on extant taxa. Biomechanical studies can therefore
245 help fill in the relationship between morphology, performance, and the environment, helping to connect
246 patterns evident at the macro-level (morphological and taxonomic change over millions of years) to the
247 individual level (where morphology and performance determine fitness; Figure 6).
248 Beam bending is a mechanical engineering approach used to calculate stresses on structural
249 beams. When applied to skeletons, beam bending analyses determine the stresses experienced by a
250 bone using the bone’s own internal geometry, the forces it experiences from muscle contractions and
251 external sources (such as a food item being bitten or the ground contacting a foot), and the angles and
252 moment arms at which those forces act. While frequently used to explore the effects of bite forces in
253 the skull (e.g., Van Valkenburgh and Ruff 1987; Busbey 1995; Therrien 2005), beam bending has also
254 been used to estimate locomotor forces in the limbs of extant and extinct animals (Biewener 1983;
255 Alexander 1985; Blob and Biewener 2001), including in horses (Biewener et al. 1983). In extinct equids,
256 beam bending was first applied to the center metapodial to explore locomotor stresses in Mesohippus
257 (subunguligrade), Merychippus (unguligrade), and modern Equus (unguligrade), using in-vivo strain
258 gauge data recorded from metapodials of living horses to ground-truth the method (Thomason 1985).
259 Using broken metacarpals to assess internal geometry and reducing forces in Mesohippus by 50% to
260 account for load-bearing side digits, Thomason (1985) found midshaft metacarpal stresses to be similar Evolution of a single toe in horses 12
261 in the extinct and extant horse. Stresses were highest in the unguligrade grazer Merychippus, suggesting
262 that size increase alone did not drive the transition from the subunguligrade to the fully unguligrade
263 foot; Thomason suggested that habitat could have been the major other factor driving this
264 morphological change in the distal limb (Thomason 1985).
265 In a recent study, we used CT scans of fossil metapodials to apply beam bending to the same
266 question in higher anatomical resolution and across much more of the equid phylogeny (McHorse et al.
267 2017). Midshaft stress under high-performance locomotion in the center metapodial was calculated for
268 extinct species in twelve equid genera, first with a full body-weight load on the center digit and then
269 with body-weight load reduced proportional to the size of the side digits using TRI (scaled as in Equation
270 2):
271
1 272 Equation 2. 푙표푎푑 = 푙표푎푑 ∗ ( ) 푇푅퐼 푏표푑푦 (2 ∗ 푇푅퐼) + 1
273
274 where loadTRI is the load on the center digit, loadbody is a species average body weight or the body weight
275 of the individual, and TRI is the Toe Reduction Index as calculated by Equation 1. The scaling is such that,
276 at a TRI of 1, the body weight is distributed equally among the three toes, and at a TRI of 0, the entire
277 body weight is on the center digit.
278 Our results supported and expanded on Thomason’s (1985) work, showing that when the side
279 digits bear load proportional to their size, bone safety factors (ratio of failure stress to peak locomotor
280 stress) were in the range of those found for extant mammals from mice to elephants (2 to 4; Biewener
281 1991). In contrast, when the side digits did not reduce the load on the center metapodial, stresses close
282 to or surpassing the tensile fracture stress of bone were reached in taxa as late as Parahippus, indicating
283 that side digits were mechanically critical for resisting stress until at least the beginning of the grazing Evolution of a single toe in horses 13
284 radiation (Figures 2 and 4). Furthermore, the center metapodial is positively allometric relative to body
285 mass in its cross-sectional geometry (i.e. resistance to compressive and bending forces), meaning that
286 the center digit compensated for reduced digits through evolutionary time (McHorse et al. 2017). This
287 positive allometry lends some support to the body size hypothesis, that a single large digit is a response
288 to increasing body sizes and can better resist the increased loads. The evolution of unguligrady, which
289 has been connected to increased rates of body size evolution (Kubo et al. 2019), could have spurred
290 these changes indirectly. The allometry results of this study (and the timing of side digits becoming
291 unnecessary for load-bearing) are also consistent with the locomotor economy hypothesis; it is possible
292 that as longer strides and thus longer limbs were favored by selection, the inertial costs of maintaining
293 side toes began to outweigh any remaining stabilizing or load-bearing benefit.
294
295 Evidence from macroevolution, biogeography, and ecology
296 Selection for digit reduction on an evolutionary scale requires morphological changes to
297 influence fitness by changing how the animal performs in its environment (Figure 6; Arnold 1983). In
298 other instances of digit reduction, new selective pressures frequently come from new ecologies or
299 locomotor modes, such as the cetacean transition into water (Shapiro et al. 2007) or the evolution of
300 ricochetal locomotion coupled with out-in-the-open foraging in jerboas (Moore et al. 2015). In horses,
301 various proximate causes of digit reduction have been suggested, as illustrated by the hypotheses set
302 forth earlier in this paper. Yet the most generally accepted ultimate cause of digit reduction in this
303 group, the one that makes for a new relationship between morphology and fitness, is the evolution of
304 grasslands.
305 Virtually no macroevolutionary work has explicitly addressed digit reduction in horses, but many
306 studies focus on horse macroevolution more generally. Of particular interest is the Miocene grazing
307 radiation of horses in North America (18 to 15 Ma), which began with Parahippus—the same genus Evolution of a single toe in horses 14
308 found to be among the last in which side toes were critical for mechanical support (Figures 2 and 4;
309 McHorse et al. 2017). Diversification rates were high and at least 19 new species originated quite
310 rapidly, although rates of morphological evolution were not elevated (MacFadden and Hulbert 1988;
311 Cantalapiedra et al. 2017). Though the radiation was suggested to be in response to grasslands, rapid
312 diversification lagged grasslands by several million years (Strömberg 2006; Cantalapiedra et al. 2017).
313 Most speciation events during the Miocene radiation were in fact via dispersal into new regions
314 (Maguire and Stigall 2008). Because dispersal was the main driver of speciation, factors that facilitated
315 movement—such as habitat fragmentation due to tectonic and climatic events—promoted speciation
316 (Stigall 2013). If digit reduction promoted greater economy of locomotion, it could therefore have
317 indirectly supported speciation.
318 There is no denying the scope of environmental change that accompanied the approximately
319 58-million-year history of horse evolution. In North America, temperatures swung from the warmth of
320 the Paleocene-Eocene thermal maximum, through a gradual, bumpy cooling spanning the Eocene,
321 Oligocene, and Miocene (periodically interrupted by warmer peaks lasting a few million years), and
322 finally dropped into the cyclical chill of Ice Ages (Figure 7; Zachos et al. 2001). These thermal changes
323 accompanied precipitation changes, from the wet tropical forest of warmer periods through increasing
324 aridity as the climate cooled (Janis 1993). It is against this climatic backdrop that grasslands evolved in
325 North America, becoming regionally dominant ecosystems approximately 22 million years ago and
326 dominant across North America by 7-11 million years ago (Strömberg 2005, 2006).
327 As the environmental landscape changed, varied habitats appeared. This variety is the basis of
328 Shotwell’s (1961) hypothesis that tridactyl horses and monodactyl horses were better suited to different
329 habitats—woodland-savanna and grasslands, respectively—and thus partitioned habitats accordingly.
330 Shotwell (1961) tracked the biogeography of the genera Hipparion (tridactyl) and Pliohippus
331 (monodactyl) in the Pliocene, connecting it to patterns of faunal change in the Northern Great Basin of Evolution of a single toe in horses 15
332 North America. In most faunas, including the Southern Great Basin, the monodactyl grazer Pliohippus is
333 found first in the middle Miocene (Clarendonian, approximately 13.6 to 10.3 million years ago), but in
334 the Northern Great Basin, it is not found until its immigration in the late Miocene and early Pliocene
335 (Hemphillian, approximately 10.3 to 4.9 million years ago). Shotwell connected this late appearance of
336 Pliohippus to the coeval spread of semi-arid plains and prairie grasslands from the Southern into the
337 Northern Great Basin, arguing that Pliohippus migration tracks this habitat. Similarly, he suggested that
338 Hipparion tracked woodland-savanna habitat, going locally extinct at the end-Hemphillian as savanna
339 habitats were reduced or eliminated but persisting longer where such habitats remained a major feature
340 of the landscape for a longer time (Shotwell 1961). To further support his claim of partitioning by
341 habitat, Shotwell noted that the relative abundance of hipparionines was not different in regions with or
342 without Pliohippus present, supporting the idea that the genera were not in direct competition.
343 In a recent study that aimed to investigate Shotwell’s (1961) hypothesis more quantitatively, we
344 tested for niche partitioning among different groups—in this case, tridactyl and monodactyl equids—
345 using site occupancy (Parker et al. 2018). With approximately 3500 fossil horse occurrences that could
346 be assigned to a North American Land Mammal Age and to a paleohabitat (forest/swamp, forest,
347 woodland, woodland-savanna, savanna, grassland-savanna, or grassland), we tested whether tridactyl
348 and monodactyl genera were found in the same habitat type significantly less often than by random
349 chance (which would support habitat partitioning). In fact, overlap between these groups was higher
350 than expected by chance in all North American Land Mammal Ages except the Blancan, where it was
351 indistinguishable from random (Parker et al. 2018). Rather than partitioning by habitat, tridactyl and
352 monodactyl horses were found together more often than by chance. Consequently, at the spatial and
353 temporal scale examined, horses were more similar by shared ancestry than different by digit state, and
354 digit state did not correlate with hypsodonty or body size (Parker et al. 2018). These results suggest that
355 the three classic equid traits did not coevolve under a single, grassland-specific selective regime, but Evolution of a single toe in horses 16
356 rather were the product of multiple selective pressures that varied across the diverse habitats available
357 to equids in the Miocene and Pliocene. This lack of correlation or habitat partitioning points to the
358 conclusion that whatever drove digit reduction was not identical to the driver of other important equid
359 traits, so the cause of digit reduction was probably multifaceted (because, e.g., grasslands as the driving
360 factor of all three traits would likely lead to correlated evolution among them).
361
362 Insights from convergent evolution of digit reduction
363 Digit reduction is widespread in tetrapods, including theropod dinosaurs, marsupials, rodents,
364 squamates, and ungulates. Although some forms of digit reduction are arguably related to very different
365 drivers than in equids (e.g., in hopping, bipedal rodents), others may be more closely related. Some of
366 the most ecologically and morphologically convergent examples to horses are in the Artiodactyla, the
367 second major clade of North American ungulates (along with the Perissodactyla, in which Equidae is a
368 family). Many artiodactyl taxa evolved elongated limbs and reduced digits throughout the Cenozoic
369 (Janis and Wilhelm 1993), and they have been characterized as a competitor group to perissodactyls,
370 perhaps partially responsible for the decline of the perissodactyl order by outcompeting them (Illius and
371 Gordon 1992), although a qualitative comparison has found that competition leading to replacement
372 was not at play (Cifelli 1981). As with hypsodonty and limb evolution in horses, parallel arguments have
373 been made in artiodactyls for the evolutionary benefit of hypsodonty (DeMiguel et al. 2014) and the
374 unguligrade distal limb (Clifford 2010). However, recent work has shown that the small herbivorous
375 “condylarths,” which were replaced by artiodactyls and perissodactyls, also had cursorial species,
376 suggesting that cursoriality was not the only driver of success in ungulates (Gould 2017).
377 Developmental studies have also shown that digit reduction is accomplished via different
378 mechanisms in these two ungulate clades. For example, horses (Perissodactyla), camels (Artiodactyla),
379 and jerboas (Rodentia) form the beginnings of all five digits in early embryonic limb patterning but then Evolution of a single toe in horses 17
380 show a spike in cell death around the reduced digits in later post-patterning stages (Cooper et al. 2014).
381 In contrast, pigs (Artiodactyla) and cattle (Artiodactyla) developmentally reduce digits via restricted
382 expression of Ptch1, with no noticeable increase in cell death (Sears et al. 2011; Cooper et al. 2014;
383 Lopez-Rios et al. 2014). These results suggest that, developmentally at least, there may be more than
384 one way to evolve reduced digits, encouraging a comparative approach to further determine what
385 similarities and differences exist among convergent evolutions of digit reduction.
386 In convergence with the foot structure of early horses, two genera of caviomorph rodents,
387 Hydrochaeris and Cavia, have three toes on the hind foot (digits II, III, and IV) and in the forefoot have
388 eliminated digit I, reduced digit V to nonfunctionality, and evolved a digit-III-dominant foot (Rocha-
389 Barbosa et al. 2007). Ground reaction forces and effective mechanical advantage of the limbs have been
390 characterized in capybara (Biewener 2005); they were found to have a less erect posture than goats, and
391 so may be a suitable comparison for basal equids, whose posture was not so erect as extant horses.
392 Even more striking convergence can be found in the South American Litopterna, an order
393 recently found to be sister to Perissodactyla (Buckley 2015; Welker et al. 2015; Westbury et al. 2017).
394 The litoptern genus Thoatherium evolved a monodactyl condition even more extreme than equids by
395 eliminating even the remnant “splint” metapodials II and IV. However, litopterns and horses show an
396 opposite relationship between digit reduction and body mass. Equids with the most reduced digits are
397 the largest species and, in general, less reduced digits are found on smaller species (McHorse et al.
398 2017). Conversely, the monodactyl litoptern species are the smallest and those with extremely robust
399 side digits are the largest (Janis 2007). This difference suggests that even if the body size hypothesis
400 remains plausible as a driver of digit reduction in horses, body size is likely not driving digit reduction in
401 litopterns. Despite these divergent body mass patterns, tracing the convergent evolution of digit
402 reduction in litopterns, together with studies of habitat and climate in South America (e.g., Strömberg et
403 al. 2013), has the potential to reveal whether other potential drivers are parallel in the two groups. Evolution of a single toe in horses 18
404
405 Future directions
406 The future of studying digit reduction in equids is promising. Here we lay out the steps we
407 believe are necessary to support, or reject, existing hypotheses, as well as new ideas for what may have
408 driven the evolution of such a remarkable trait as monodactyly. It is important to recognize that most of
409 the working hypotheses are not mutually exclusive, and as has become clear from previous work, we
410 should expect interrelationships between the potential causes of digit reduction (e.g., cooler climate
411 may affect body size directly, which is a hypothesized driver of digit reduction, but cooler climate also
412 leads to more open habitats, which is a different hypothesized driver). However, we argue that the way
413 that these selective drivers interact has a significant effect on how we conceptualize the “why” behind
414 digit reduction, and therefore it is a valuable endeavor to uncover the primary driver or drivers of
415 monodactyly and digit reduction as a whole.
416
417 Locomotion and biomechanics
418 If the primary driver of digit reduction was the need for better economy while covering long
419 distances, as hypothesized by Janis and Wilhelm (1993), then digit reduction should virtually always go
420 hand-in-hand with limb elongation. Research manipulating moment of inertia (MOI) in limbs has shown
421 that decreasing MOI does reduce cost of transport (Martin 1985; Myers and Steudel 1985; Wickler et al.
422 2004), and added distal limb mass increases cost of transport in extant horses (Wickler et al. 2004), so
423 some degree of energetic savings is almost certainly a consequence of digit reduction. The logical next
424 step is a theoretical exploration of the magnitude of energetic savings from 1) reduction of MOI at the
425 distal limb due to loss of digits and 2) elongation of the limb over 3) a range of body sizes and taxa. The
426 results from these calculations could be used to create a theoretical cost of transport morphospace that
427 connects changes in limb length, relative differences in segment MOI, estimated limb swing frequency, Evolution of a single toe in horses 19
428 and body mass to a resulting cost of transport. That morphospace could be used to calculate, e.g., how
429 much locomotor economy was improved between different taxa and whether the difference gained by
430 digit reduction constitutes a significant energetic savings. To evaluate significance of energetic savings, it
431 would be necessary to relate the results of the calculations to known costs associated with swinging legs
432 in living animals (e.g., Fedak et al. 1982). Tracking these changes through time would further allow
433 testing of whether quantitative improvements in locomotor economy over evolutionary time tracked (or
434 slightly lagged) aridification and the spread of grasslands. This analysis is ideally suited to include
435 artiodactyls as a comparative group, because they also evolved longer limbs, reduced digits, and would
436 have experienced the same pressures at the same times where they overlapped spatially with horses.
437 The idea that a tridactyl foot is more stable for lateral dodging on soft substrate, whereas a
438 more rigid single hoof is more stable for and provides faster straight-line locomotion on hard substrate
439 (Shotwell 1961), remains untested. The soft substrate hypothesis is concerned primarily with stability,
440 whereas the hard substrate may have considerably more to do with elastic energy storage and energy
441 dissipation. Whether theoretical or practical (as in biorobotics), this hypothesis requires a test of
442 whether tridactyl feet are indeed more stable on softer substrates. Examination of locomotor
443 performance in extant taxa with reduced digits would provide an interesting first step, which could be
444 complemented by studies using modeling and simulation to explore the effect of digit number on
445 stability in the equid distal limb; a combination of simulation and biorobotics would offer considerable
446 flexibility (e.g., Nyakatura et al. 2019). Biomechanical modeling work such as this can be more powerful
447 when ground-truthed with locomotion data, such as speed, joint kinematics, and forces from extant
448 animals—a challenge when the vast majority of equid diversity is extinct. Monodactyl equids are
449 straightforward to study in that domestic horses are anatomically extremely similar, particularly in the
450 distal limb, to both wild Equus and to other extinct monodactyl taxa. Evolution of a single toe in horses 20
451 Several taxa are potential modern analogs for tridactyl or semi-tetradactyl horses, and would
452 therefore be suitable for study of locomotor biomechanics (directly testing dodging stability on softer
453 substrates), for ground-truthing the proposed simulation and biorobotics work, and for comparative
454 studies of distal limb anatomy (including internal geometry, i.e., resistance to bending, torsion, and
455 compressive forces). Tapirs (Tapiridae) are one of three extant families in the perissodactyl order, the
456 others being equids and rhinoceroses, and are similar to basal equids in their digit state: four digits on
457 the front leg and three in the back. Tapirs may therefore offer a convenient semi-tetradactyl species for
458 locomotion studies of substrate-based stability; they have been used for comparative anatomical and
459 biomechanical studies (McHorse et al., 2017), and we have collected kinetics and kinematics data from
460 Baird’s tapir (Tapirus bairdii) that will provide locomotor forces to scale for future finite element
461 modeling work. A recent study has argued that the semiaquatic tapir benefits from lateral splaying in
462 phalanges II and IV, which could allow for greater stability on soft, muddy surfaces beneath water (Endo
463 et al. 2019), echoing the substrate stability hypothesis itself.
464 While tapirs provide the closest phylogenetic match for a study of digits and stability, they may
465 not provide the best biomechanical one. In contrast to equids, tapirs are semiaquatic (Nowak and
466 Paradiso 1999), and although some tapir species may provide morphological analogs for other basal
467 perissodactyls such as palaeotheres (MacLaren and Nauwelaerts 2019), the range of body sizes differs
468 considerably in the two families. Most basal equids had a body mass between 5 kg and 20 kg
469 (MacFadden 1986; Secord et al. 2012), but even the smallest extant tapir species have body masses over
470 130 kg (Nowak and Paradiso 1999). Therefore, caviomorph rodents and small artiodactyls may provide a
471 closer biomechanical analog in terms of the forces generated during locomotion. The two genera of
472 caviomorph rodents with remarkable convergence towards the equid foot condition are much closer in
473 size: species tend to average 30-60 kg and 1 kg in Hydrochaeris and Cavia respectively (Biewener 2005;
474 Ferraz et al. 2005). Similarly, some species of artiodactyl are quite small (e.g., tragulids and moschids are Evolution of a single toe in horses 21
475 generally less than 10 kg; Nowak and Paradiso 1999) and may provide a more suitable biomechanical
476 comparison than tapirs. Furthermore, extant artiodactyls are considerably more taxonomically diverse
477 than perissodactyls, and they underwent many similar changes in the evolution of unguligrady and
478 reduced digits—often argued to be an adaptation for fast and efficient locomotion (Clifford 2010). As
479 with all non-domesticated animals, the greatest challenge here may lie in access, and the most suitable
480 biomechanical comparison might require a compromise between the anatomy, preferred habitat, and
481 the availability (and behavioral temperament) of various species.
482
483 Macroevolutionary and regional analyses
484 While biomechanical studies can help untangle why different morphologies might be adaptive
485 for different substrates, macroevolution can give us perspective on how it happened. If grasslands were
486 the ultimate driver of monodactyly, we would expect genera with more reduced digits to be found more
487 often and/or to be more evolutionarily successful in open grassland-dominated habitats. Conversely, we
488 would expect genera that retained considerable side digits, such as hipparionines, to be found more
489 often in softer, forested environments and/or to be more successful there. Differential habitat use has
490 been shown in several taxa to provide diverging selective demands and thus to influence diversification
491 (Losos 2009; Collar et al. 2011; Price et al. 2011), although the groups in those studies (anoles, monitor
492 lizards, and labrid fishes) are all much smaller than equids. A recent spatially and temporally broad study
493 of habitat partitioning found no such evidence for tridactyl and monodactyl horses dividing habitats
494 where they overlapped (Parker et al. 2018).
495 Explicitly hypothesis-testing the niche partitioning aspect of Shotwell’s (1961) argument was an
496 important first step in tackling these ideas quantitatively, but the study by Parker et al. (2018) had some
497 limitations in scope. First, both temporal and spatial bins were coarse; North American Land Mammal
498 ages can be several million years long, which represents considerable time-averaging, and the spatial Evolution of a single toe in horses 22
499 resolution of habitat identification was typically at the county level. Since modern counties certainly
500 cover sufficient land to contain multiple habitat types, it is likely that fossil localities within the same
501 county could have had very different habitats. Second, the relative distribution of habitats in the study
502 was fairly constant through time, despite evidence that, on the whole, grasslands became the far more
503 dominant ecosystem. It is possible that the data analyzed by Parker et al. (2018) do not sufficiently
504 capture ecological reality, perhaps reflecting a bias in preservational environments that led to the
505 fossilization of horses.
506 Regional analyses may offer a solution to the limitations of spatial and temporal data,
507 particularly given how regionally-specific climate and habitat can be (Chen et al. 2015). We propose that
508 the Great Plains region of North America, which includes parts of several states including Montana,
509 Wyoming, North and South Dakota, and Nebraska, offers an ideal place to test more carefully the
510 relationship between horses, climate, and habitat (Figure 8). The horse fossil record is dense,
511 particularly in Nebraska (Figure 8; MioMap search for “Equidae” returns more than 1000 specimens,
512 Carrasco et al. 2005). The phytoliths (grass species indicators) and paleosols (C3 vs. C4 grass indicators)
513 are also well-characterized in the area (Figure 8; Fox and Koch 2003; Strömberg 2005, 2011). Other
514 potential candidate regions include the John Day region of Oregon and the state of Florida, both with
515 remarkable fossil records of horse evolution and extensive research into climatic and habitat change
516 through time (Stock 1946; Macfadden and Cerling 1996; MacFadden et al. 1999; Retallack 2004;
517 Maguire and Stigall 2008; Maguire 2015). However, there are tradeoffs to choosing regional-scale
518 studies: in exchange for better-controlled data and more power to detect trends locally, one gives up
519 some amount of power to explain global trends. In other words, trends at one scale cannot necessarily
520 be extrapolated to others—a critical challenge of macroevolutionary studies in general (Jablonski 2008).
521 Although qualitatively the trend appears to point toward monodactyly being ‘optimal,’ at least
522 in grasslands, this pattern has yet to be quantitatively tested—and as we know from work showing that Evolution of a single toe in horses 23
523 body size evolution was likely not directional in horses, apparent trends can be deceiving. Phylogenetic
524 comparative methods offer a way to explicitly test the evolutionary mode of trends like these.
525 Evolutionary model-fitting can compare the fit of models such as Brownian Motion (a random walk), an
526 Ornstein-Uhlenbeck (OU) process (a model where the trait is being pulled with some strength towards
527 an adaptive peak of some ‘optimal’ value), or a multi-peak OU model, which allows for multiple optima
528 that may correspond to another feature such as habitat (Hansen 1997; Butler and King 2004). In an in-
529 progress study, we explicitly test digit reduction in this framework, investigating whether digit reduction
530 is pulled to some adaptive optimum for all of equids (i.e., some degree of digit reduction is “optimal”) or
531 whether that optimum varies based on habitat type (e.g., forest-dwelling species are pulled towards
532 some moderate value of TRI whereas grassland dwellers are pulled towards monodactyly). Alternatively,
533 if different habitats drive different rates of digit evolution but there is no trait optimum, a multi-rate
534 model would be more appropriate (Collar et al. 2010). A study such as this would also be suited to a
535 more broadly comparative context, evaluating whether the mode of digit reduction evolution is similar
536 in other taxa (e.g., artiodactyls or litopterns).
537
538 Conclusion
539 The evolution of monodactyly in horses is remarkable and is unique among extant animals, but
540 fortunately for scientists, the themes of digit reduction, habitat change, and body size change are
541 repeated many times in the fossil record. Reviewing the available evidence makes it clear that we are
542 unlikely to find a single evolutionary driver to be solely responsible for the evolution of monodactyly,
543 because open habitat, changes in substrate, changes in foot posture, and changes in body size can all tie
544 to one another and to broader ecological drivers such as changing climate. However, we argue that by
545 combining finer-scale regional studies, quantitative biomechanical studies, and careful analysis of
546 convergent clades, it will be possible to estimate the relative contributions of these evolutionary drivers. Evolution of a single toe in horses 24
547 Even if digit reduction is ultimately not driven by the same factors in each clade (as may be the case with
548 horses vs. litopterns), such a discovery would be a considerable leap forward in our understanding of
549 how—and why—horses evolved a single toe.
550
551 Funding
552 This work was supported by the National Science Foundation [DGE-1144152 to B.K.M, DEB-
553 1701656 to B.K.M and S.E.P].
554
555 Acknowledgments
556 The authors would like to thank Zachary Morris and other members of the Pierce and Biewener
557 labs for productive discussion; Talia Moore for help conceptualizing Figure 6; Samantha Hopkins and
558 Edward Davis, who first encouraged the equid evolutionary line of thinking and have provided ongoing
559 thoughts; and Abigail Parker and Tristan Reinecke, whose work contributed to ideas mentioned in this
560 review. Hayley O’Brien suggested looking into caviomorph rodents. Zhijie Jack Tseng and an anonymous
561 reviewer provided helpful comments that improved the manuscript. Finally, the authors would like to
562 thank Samantha Price and Martha Muñoz, who organized the symposium on Biomechanics in the Era of
563 Big Data, to which this paper is a contribution.
564
565 References
566 Alexander RM. 1985. Mechanics of posture and gait of some large dinosaurs. Zoological Journal of the
567 Linnean Society 83:1–25.
568 Anyonge W. 1996. Locomotor behaviour in Plio-Pleistocene sabre-tooth cats: a biomechanical analysis.
569 Journal of Zoology 238:395–413.
570 Arnold SJ. 1983. Morphology, performance and fitness. Integr Comp Biol 23:347–61. Evolution of a single toe in horses 25
571 Bai B, Wang Y-Q, Meng J. 2018. The divergence and dispersal of early perissodactyls as evidenced by
572 early Eocene equids from Asia. Communications Biology 1:115.
573 Barrey E, Desliens F, Poirel D, Biau S, Lemaire S, Rivero JLL, Langlois B. 2002. Early evaluation of dressage
574 ability in different breeds. Equine Veterinary Journal 34:319–324.
575 Biewener AA. 1983. Allometry of quadrupedal locomotion: the scaling of duty factor, bone curvature
576 and limb orientation to body size. Journal of Experimental Biology 105:147–171.
577 Biewener AA. 1991. Musculoskeletal design in relation to body size. Journal of Biomechanics,
578 Proceedings of the NASA Symposium on the Influence of Gravity and Activity on Muscle and
579 Bone 24:19–29.
580 Biewener AA. 1998. Muscle-tendon stresses and elastic energy storage during locomotion in the horse.
581 Comparative Biochemistry and Physiology Part B: Biochemistry and Molecular Biology 120:73–
582 87.
583 Biewener AA. 2005. Biomechanical consequences of scaling. Journal of Experimental Biology 208:1665–
584 1676.
585 Biewener AA, Thomason J, Goodship A, Lanyon LE. 1983. Bone stress in the horse forelimb during
586 locomotion at different gaits: a comparison of two experimental methods. Journal of
587 Biomechanics 16:565–576.
588 Blob RW, Biewener AA. 2001. Mechanics of limb bone loading during terrestrial locomotion in the green
589 iguana (Iguana iguana) and American alligator (Alligator mississippiensis). Journal of
590 Experimental Biology 204:1099–1122.
591 Browning RC, Modica JR, Kram R, Goswami A. 2007. The effects of adding mass to the legs on the
592 energetics and biomechanics of walking. Med Sci Sports Exerc 39:515–25.
593 Buckley M. 2015. Ancient collagen reveals evolutionary history of the endemic South American
594 ‘ungulates.’ Proceedings of the Royal Society B: Biological Sciences 282:20142671. Evolution of a single toe in horses 26
595 Busbey AB. 1995. The structural consequences of skull flattening in crocodilians. In: Thomason JJ, editor.
596 Functional Morphology in Vertebrate Paleontology Cambridge: Cambridge University Press. p.
597 173–192.
598 Butler MA, King AA. 2004. Phylogenetic comparative analysis: a modeling approach for adaptive
599 evolution. The American Naturalist 164:683–95.
600 Camp CL, Smith N. 1942. Phylogeny and functions of the digital ligaments of the horse. Memoirs of the
601 University of California 3:69–124.
602 Cantalapiedra JL, Prado JL, Fernández MH, Alberdi MT. 2017. Decoupled ecomorphological evolution
603 and diversification in Neogene-Quaternary horses. Science 355:627–30.
604 Carrasco, MA, Kraatz BP, Davis EB, Barnosky AD. 2005. Miocene Mammal Mapping Project (MIOMAP).
605 University of California Museum of Paleontology, http://www.ucmp.berkeley.edu/miomap/
606 Chen ST, Smith SY, Sheldon ND, Strömberg CA. 2015. Regional-scale variability in the spread of
607 grasslands in the late Miocene. Palaeogeography, Palaeoclimatology, Palaeoecology 437:42–52.
608 Cifelli RL. 1981. Patterns of evolution among the Artiodactyla and Perissodactyla (Mammalia). Evolution
609 35:433–40.
610 Clifford AB. 2010. The evolution of the unguligrade manus in artiodactyls. Journal of Vertebrate
611 Paleontology 30:1827–39.
612 Codron D, Lee-Thorp JA, Sponheimer M, Codron J. 2007. Nutritional content of savanna plant foods:
613 implications for browser/grazer models of ungulate diversification. Eur J Wildl Res 53:100–111.
614 Collar DC, Schulte II JA, Losos JB. 2011. Evolution of extreme body size disparity in monitor lizards
615 (Varanus). Evolution 65:2664–80.
616 Collar DC, Schulte JA, O’meara BC, Losos JB. 2010. Habitat use affects morphological diversification in
617 dragon lizards. Journal of Evolutionary Biology 23:1033–1049.
618 Cooper KL, Sears KE, Uygur A, Maier J, Baczkowski K-S, Brosnahan M, Antczak D, Skidmore JA, Tabin CJ. Evolution of a single toe in horses 27
619 2014. Patterning and post-patterning modes of evolutionary digit loss in mammals. Nature
620 511:41–45.
621 de Bakker MAG, Fowler DA, den Oude K, Dondorp EM, Navas MCG, Horbanczuk JO, Sire J-Y, Szczerbińska
622 D, Richardson MK. 2013. Digit loss in archosaur evolution and the interplay between selection
623 and constraints. Nature 500:445–48.
624 DeMiguel D, Azanza B, Morales J. 2014. Key innovations in ruminant evolution: a paleontological
625 perspective. Integrative Zoology 9:412–433.
626 Demment MW, Van Soest PJ. 1985. A nutritional explanation for body-size patterns of ruminant and
627 nonruminant herbivores. The American Naturalist 125:641–72.
628 Doube M, Felder AA, Chua MY, Lodhia K, Kłosowski MM, Hutchinson JR, Shefelbine SJ. 2018. Limb bone
629 scaling in hopping macropods and quadrupedal artiodactyls. Royal Society Open Science
630 5:180152.
631 Endo H, Yoshida M, Nguyen TS, Akiba Y, Takeda M, Kudo K. 2019. Three-dimensional CT examination of
632 the forefoot and hindfoot of the hippopotamus and tapir during a semiaquatic walking.
633 Anatomia, Histologia, Embryologia 48:3–11.
634 Famoso NA, Davis EB. 2014. Occlusal enamel complexity in Middle Miocene to Holocene equids
635 (Equidae: Perissodactyla) of North America. PLOS ONE 9:e90184.
636 Fastest speed for a race horse. 2019. Guinness World Records.
637 (http://www.guinnessworldrecords.com/world-records/fastest-speed-for-a-race-horse/).
638 Fedak MA, Heglund NC, Taylor CR. 1982. Energetics and mechanics of terrestrial locomotion. II. Kinetic
639 energy changes of the limbs and body as a function of speed and body size in birds and
640 mammals. Journal of Experimental Biology 97:23–40.
641 Ferraz KMPM de, Bonach K, Verdade LM. 2005. Relationship between body mass and body length in
642 capybaras (Hydrochoerus hydrochaeris). Biota Neotropica 5:197–200. Evolution of a single toe in horses 28
643 Fox DL, Koch PL. 2003. Tertiary history of C4 biomass in the Great Plains, USA. Geology 31:809–812.
644 Fraser D, Gorelick R, Rybczynski N. 2015. Macroevolution and climate change influence phylogenetic
645 community assembly of North American hoofed mammals. Biol J Linn Soc Lond 114:485–94.
646 Froehlich DJ. 2002. Quo vadis eohippus? The systematics and taxonomy of the early Eocene equids
647 (Perissodactyla). Zool J Linn Soc 134:141–256.
648 Gnagey L, Clayton HM, Lanovaz JL. 2006. Effect of standing tarsal angle on joint kinematics and kinetics.
649 Equine Veterinary Journal 38:628–633.
650 Gould FDH. 2017. Testing the role of cursorial specializations as adaptive key innovations in Paleocene-
651 Eocene ungulates of North America. J Mammal Evol 24:453–63.
652 Green DR, Olack G, Colman AS. 2018. Determinants of blood water δ18O variation in a population of
653 experimental sheep: Implications for paleoclimate reconstruction. Chemical Geology 485:32–43.
654 Hansen TF. 1997. Stabilizing selection and the comparative analysis of adaptation. Evolution 51:1341–
655 51.
656 Highest jump by a horse. 2019. Guinness World Records.
657 (http://www.guinnessworldrecords.com/world-records/highest-jump-by-a-horse).
658 Hildebrand M. 1960. How animals run. Scientific American 202:148–160.
659 Hobbs SJ, Levine D, Richards J, Clayton H, Tate J, Walker R. 2010. Motion analysis and its use in equine
660 practice and research. Wiener Tierarztliche Monatsschrift 97:55–64.
661 Hutchinson JR, Garcia M. 2002. Tyrannosaurus was not a fast runner. Nature 415:1018–1021.
662 Illius AW, Gordon IJ. 1992. Modelling the nutritional ecology of ungulate herbivores: evolution of body
663 size and competitive interactions. Oecologia 89:428–34.
664 Jablonski D. 2008. Biotic interactions and macroevolution: extensions and mismatches across scales and
665 levels. Evolution: International Journal of Organic Evolution 62:715–739.
666 Janis CM. 1993. Tertiary mammal evolution in the context of changing climates, vegetation, and tectonic Evolution of a single toe in horses 29
667 events. Annual Review of Ecology and Systematics 24:467–500.
668 Janis CM. 2007. The horse series. In: Regal B, editor. Icons of evolution Westport: Greenwood Press. p.
669 257–80.
670 Janis CM, Wilhelm PB. 1993. Were there mammalian pursuit predators in the Tertiary? Dances with wolf
671 avatars. Journal of Mammalian Evolution 1:103–125.
672 Jardine PE, Janis CM, Sahney S, Benton MJ. 2012. Grit not grass: concordant patterns of early origin of
673 hypsodonty in Great Plains ungulates and Glires. Palaeogeography, Palaeoclimatology,
674 Palaeoecology.
675 Jones K. 2016. New insights on equid locomotor evolution from the lumbar region of fossil horses.
676 Proceedings of the Royal Society B: Biological Sciences 283:20152947.
677 Kilbourne BM, Andrada E, Fischer MS, Nyakatura JA. 2016. Morphology and motion: hindlimb
678 proportions and swing phase kinematics in terrestrially locomoting charadriiform birds. Journal
679 of Experimental Biology 219:1405–16.
680 Kitts DB. 1956. American Hyracotherium (Perissodactyla, Equidae). Bulletin of the American Museum of
681 Natural History 110:1–60.
682 Kristjansson T, Bjornsdottir S, Albertsdóttir E, Sigurdsson A, Pourcelot P, Crevier-Denoix N, Arnason T.
683 2016. Association of conformation and riding ability in Icelandic horses. Livestock Science
684 189:91–101.
685 Kubo T, Sakamoto M, Meade A, Venditti C. 2019. Transitions between foot postures are associated with
686 elevated rates of body size evolution in mammals. PNAS 116:2618–23.
687 Lopez-Rios J, Duchesne A, Speziale D, Andrey G, Peterson KA, Germann P, Ünal E, Liu J, Floriot S, Barbey
688 S, others. 2014. Attenuated sensing of SHH by Ptch1 underlies evolution of bovine limbs. Nature
689 511:46–51.
690 Losos JB. 2009. Lizards in an evolutionary tree: ecology and adaptive radiation of anoles Univ of Evolution of a single toe in horses 30
691 California Press.
692 Lovegrove BG, Mowoe MO. 2013. The evolution of mammal body sizes: responses to Cenozoic climate
693 change in North American mammals. Journal of Evolutionary Biology 26:1317–1329.
694 MacFadden BJ. 1986. Fossil horses from “Eohippus” (Hyracotherium) to Equus: scaling, Cope’s Law, and
695 the evolution of body size. Paleobiology 12:355–69.
696 MacFadden BJ. 1994. Fossil Horses: Systematics, Paleobiology, and Evolution of the Family Equidae
697 Cambridge University Press.
698 Macfadden BJ, Cerling TE. 1996. Mammalian herbivore communities, ancient feeding ecology, and
699 carbon isotopes: A 10 million-year sequence from the Neogene of Florida. Journal of Vertebrate
700 Paleontology 16:103–15.
701 MacFadden BJ, Hulbert RC. 1988. Explosive speciation at the base of the adaptive radiation of Miocene
702 grazing horses. Nature 336:466–468.
703 MacFadden BJ, Oviedo LH, Seymour GM, Ellis S. 2012. Fossil horses, orthogenesis, and communicating
704 evolution in museums. Evo Edu Outreach 5:29–37.
705 MacFadden BJ, Solounias N, Cerling TE. 1999. Ancient diets, ecology, and extinction of 5-million-year-old
706 horses from Florida. Science 283:824–827.
707 MacLaren JA, Nauwelaerts S. 2019. Modern tapirs as morphofunctional analogues for locomotion in
708 endemic Eocene European perissodactyls. J Mammal Evol.
709 Maguire KC. 2015. Dietary niche stability of equids across the mid-Miocene Climatic Optimum in
710 Oregon, USA. Palaeogeography, Palaeoclimatology, Palaeoecology 426:297–307.
711 Maguire KC, Stigall AL. 2008. Paleobiogeography of Miocene Equinae of North America: A phylogenetic
712 biogeographic analysis of the relative roles of climate, vicariance, and dispersal.
713 Palaeogeography, Palaeoclimatology, Palaeoecology 267:175–84.
714 Mainland IL. 2003. Dental microwear in grazing and browsing Gotland sheep (Ovis aries) and its Evolution of a single toe in horses 31
715 implications for dietary reconstruction. Journal of Archaeological Science 30:1513–27.
716 Marsh OC. 1874. Fossil horses in America. The American Naturalist 8:288–294.
717 Marshall CR, Finnegan S, Clites EC, Holroyd PA, Bonuso N, Cortez C, Davis E, Dietl GP, Druckenmiller PS,
718 Eng RC. 2018. Quantifying the dark data in museum fossil collections as palaeontology
719 undergoes a second digital revolution. Biology Letters 14:20180431.
720 Martin PE. 1985. Mechanical and physiological responses to lower extremity loading during running.
721 Medicine and Science in Sports and Exercise 17:427–433.
722 Martin RA. 2018. Scaling of species diversity and body mass in mammals: Cope’s rule and the
723 evolutionary cost of large size. Historical Biology 0:1–14.
724 Matthew WD. 1926. The evolution of the horse: a record and its interpretation. The Quarterly Review of
725 Biology 1:139–85.
726 McHorse BK, Biewener AA, Pierce SE. 2017. Mechanics of evolutionary digit reduction in fossil horses
727 (Equidae). Proceedings of the Royal Society B: Biological Sciences 284.
728 Mendoza M, Palmqvist P. 2008. Hypsodonty in ungulates: an adaptation for grass consumption or for
729 foraging in open habitat? Journal of Zoology 274:134–42.
730 Mihlbachler MC, Rivals F, Solounias N, Semprebon GM. 2011. Dietary change and evolution of horses in
731 North America. Science 331:1178–81.
732 Moore TY, Organ CL, Edwards SV, Biewener AA, Tabin CJ, Jenkins Jr. FA, Cooper KL. 2015. Multiple
733 phylogenetically distinct events shaped the evolution of limb skeletal morphologies associated
734 with bipedalism in the jerboas. Current Biology 25:2785–94.
735 Myers MJ, Steudel K. 1985. Effect of limb mass and its distribution on the energetic cost of running.
736 Journal of Experimental Biology 116:363–73.
737 Nowak RM. 1999. Walker’s Mammals of the World. 6th ed Baltimore: Johns Hopkins University Press.
738 Nyakatura JA, Melo K, Horvat T, Karakasiliotis K, Allen VR, Andikfar A, Andrada E, Arnold P, Lauströer J, Evolution of a single toe in horses 32
739 Hutchinson JR. 2019. Reverse-engineering the locomotion of a stem amniote. Nature 565:351.
740 O’Meara BC, Beaulieu JM. 2014. Modelling stabilizing selection: the attraction of Ornstein–Uhlenbeck
741 models. In: Garamszegi LZ, editor. Modern Phylogenetic Comparative Methods and Their
742 Application in Evolutionary Biology: Concepts and Practice Berlin, Heidelberg: Springer Berlin
743 Heidelberg. p. 381–93.
744 Parker AK, McHorse BK, Pierce SE. 2018. Niche modeling reveals lack of broad-scale habitat partitioning
745 in extinct horses of North America. Palaeogeography, Palaeoclimatology, Palaeoecology.
746 Pierce SE, Clack JA, Hutchinson JR. 2012. Three-dimensional limb joint mobility in the early tetrapod
747 Ichthyostega. Nature 486:523.
748 Pierce SE, Hutchinson JR, Clack JA. 2013. Historical perspectives on the evolution of tetrapodomorph
749 movement. Integrative and Comparative Biology.
750 Price SA, Holzman R, Near TJ, Wainwright PC. 2011. Coral reefs promote the evolution of morphological
751 diversity and ecological novelty in labrid fishes. Ecology letters 14:462–469.
752 Raia P, Carotenuto F, Passaro F, Fulgione D, Fortelius M. 2012. Ecological specialization in fossil
753 mammals explains Cope’s Rule. The American Naturalist 179:328–37.
754 Retallack GJ. 2004. Late Miocene climate and life on land in Oregon within a context of Neogene global
755 change. Palaeogeography, Palaeoclimatology, Palaeoecology 214:97–123.
756 Rocha-Barbosa O, Loguercio MFC, Renous S, Gasc J-P. 2007. Comparative study on the forefoot and
757 hindfoot intrinsic muscles of some cavioidea rodents (Mammalia, Rodentia). Zoology 110:58–65.
758 Schimel J. 2012. Writing science: how to write papers that get cited and proposals that get funded
759 Oxford: Oxford University Press.
760 Schuchert C. 1940. O.C. Marsh, pioneer in paleontology New Haven: Yale University Press.
761 Sears KE, Bormet AK, Rockwell A, Powers LE, Cooper LN, Wheeler MB. 2011. Developmental basis of
762 mammalian digit reduction: a case study in pigs. Evolution & Development 13:533–41. Evolution of a single toe in horses 33
763 Secord R, Bloch JI, Chester SG, Boyer DM, Wood AR, Wing SL, Kraus MJ, McInerney FA, Krigbaum J. 2012.
764 Evolution of the earliest horses driven by climate change in the Paleocene-Eocene Thermal
765 Maximum. Science 335:959–962.
766 Shapiro MD, Shubin NH, Downs JP. 2007. Limb diversity and digit reduction in reptilian evolution. In: Hall
767 BK, editor. Fins into limbs: evolution, development, and transformation. Chicago and London:
768 University of Chicago Press. p. 225–245.Shoemaker L, Clauset A, Hector A. 2013. Body mass
769 evolution and diversification within horses (family Equidae). Ecology Letters 17:211–20.
770 Shotwell JA. 1961. Late Tertiary biogeography of horses in the northern Great Basin. Journal of
771 Paleontology 35:203–217.
772 Simpson GG. 1951. Horses: The Story of the Horse Family in the Modern World and through Sixty Million
773 Years Oxford University Press.
774 Solounias N, Danowitz M, Stachtiaris E, Khurana A, Araim M, Sayegh M, Natale J. 2018. The evolution
775 and anatomy of the horse manus with an emphasis on digit reduction. Royal Society Open
776 Science 5:171782.
777 Sondaar PY. 1968. The osteology of the manus of fossil and recent Equidae, with special reference to
778 phylogeny and function. Verhandelingen der Koninklijke Nederlandse akademie van
779 wetenschappen 25:1–76.
780 Stigall AL. 2013. Analysing links between biogeography, niche stability and speciation: the impact of
781 complex feedbacks on macroevolutionary patterns. Palaeontology 56:1225–38.
782 Stock C. 1946. Oregon’s Wonderland of the Past--The John Day. The Scientific Monthly 63:57–65.
783 Strömberg CA. 2002. The origin and spread of grass-dominated ecosystems in the late Tertiary of North
784 America: preliminary results concerning the evolution of hypsodonty. Palaeogeography,
785 Palaeoclimatology, Palaeoecology 177:59–75.
786 Strömberg CA. 2006. Evolution of hypsodonty in equids: testing a hypothesis of adaptation. Paleobiology Evolution of a single toe in horses 34
787 32:236–58.
788 Strömberg CA. 2011. Evolution of grasses and grassland ecosystems. Annual Review of Earth and
789 Planetary Sciences 39:517–544.
790 Strömberg CAE. 2005. Decoupled taxonomic radiation and ecological expansion of open-habitat grasses
791 in the Cenozoic of North America. PNAS 102:11980–84.
792 Strömberg CAE, Dunn RE, Madden RH, Kohn MJ, Carlini AA. 2013. Decoupling the spread of grasslands
793 from the evolution of grazer-type herbivores in South America. Nat Commun 4:1478.
794 Swartz SM, Bennett MB, Carrier DR. 1992. Wing bone stresses in free flying bats and the evolution of
795 skeletal design for flight. Nature 359:726–29.
796 Therrien F. 2005. Feeding behaviour and bite force of sabretoothed predators. Zoological Journal of the
797 Linnean Society 145:393–426.
798 Thomason JJ. 1985. Estimation of locomotory forces and stresses in the limb bones of recent and extinct
799 equids. Paleobiology 11:209–20.
800 Thomason JJ. 1986. The functional morphology of the manus in the tridactyl equids Merychippus and
801 Mesohippus: paleontological inferences from neontological models. Journal of Vertebrate
802 Paleontology 6:143–161.
803 Van Valkenburgh B, Ruff CB. 1987. Canine tooth strength and killing behaviour in large carnivores.
804 Journal of Zoology 212:379–397.
805 Welker F, Collins MJ, Thomas JA, Wadsley M, Brace S, Cappellini E, Turvey ST, Reguero M, Gelfo JN,
806 Kramarz A. 2015. Ancient proteins resolve the evolutionary history of Darwin’s South American
807 ungulates. Nature 522:81.
808 Weller R, Pfau T, Verheyen K, May SA, Wilson AM. 2006. The effect of conformation on orthopaedic
809 health and performance in a cohort of National Hunt racehorses: preliminary results. Equine
810 Veterinary Journal 38:622–627. Evolution of a single toe in horses 35
811 Westbury M, Baleka S, Barlow A, Hartmann S, Paijmans JL, Kramarz A, Forasiepi AM, Bond M, Gelfo JN,
812 Reguero MA. 2017. A mitogenomic timetree for Darwin’s enigmatic South American mammal
813 Macrauchenia patachonica. Nature Communications 8:15951.
814 Wickler SJ, Hoyt DF, Clayton HM, Mullineaux DR, Cogger EA, Sandoval E, McGuire R, Lopez C. 2004.
815 Energetic and kinematic consequences of weighting the distal limb. Equine Veterinary Journal
816 36:772–777.
817 Winkler DE, Schulz-Kornas E, Kaiser TM, Cuyper AD, Clauss M, Tütken T. 2019. Forage silica and water
818 content control dental surface texture in guinea pigs and provide implications for dietary
819 reconstruction. PNAS 116:1325–30.
820 Wood AR, Bebej RM, Manz CL, Begun DL, Gingerich PD. 2011. Postcranial functional morphology of
821 Hyracotherium (Equidae, Perissodactyla) and locomotion in the earliest horses. J Mammal Evol
822 18:1–32.
823 Zachos J, Pagani M, Sloan L, Thomas E, Billups K. 2001. Trends, rhythms, and aberrations in global
824 climate 65 Ma to present. Science 292:686–693.
825
826 Figure Captions
827 Figure 1. The anatomy of modern Equus metapodials; proximal articular views are of the metacarpal (A)
828 and metatarsal (B), and the metacarpal and phalanges are shown in anterior (C) and posterior (D) views.
829 Abbreviations: digits II, III, and IV are shown for the metacarpal and metatarsal; PP is proximal phalanx;
830 MP is medial phalanx; DP is distal phalanx. Sesamoids are indicated with lines.
831 Figure 2. The linear progression of horses (small to large, many toes to one toe, low-crowned teeth to
832 high-crowned teeth), a view that dominated early narratives about equid evolution. Modified from
833 Matthew (1926). Evolution of a single toe in horses 36
834 Figure 3. A simplified cladogram of horse genera with tapir as an outgroup. Topology after Froehlich
835 (2002), Fraser et al. (2015), Jones (2016), and Bai et al. (2018). Subclades are highlighted by color and
836 are after Famoso and Davis (2014) and Cantalapiedra et al. (2017), but are frequently paraphyletic (e.g.,
837 the Merychippus-Grade Equinae). Note that “Merychippus” is a known polyphyletic group, and here we
838 include only one of several phylogenetic positions for taxa called Merychippus; see Fraser et al. (2015).
839 Digit state is shown by lines (solid black for semi-tetradactyl, thin gray for tridactyl, dotted black for
840 monodactyl). Size is indicated by a circle, scaled based on the base-10 logarithm of body mass (where
841 available). Extant taxa are marked with an asterisk.
842 Figure 4. An illustration of the measurements used to calculate TRI. First the lengths of the proximal side
843 phalanges (PPlengthIV and PPlengthII) are averaged, then this value is divided by the length of the
844 proximal center phalanx (PPlengthIII). Illustration modified from Matthew (1926).
845 Figure 5. A phylogenetic tree of some horse genera showing discrete categories of digit state (left) vs.
846 Toe Reduction Index (TRI), a continuous measure of digit reduction (right). TRI captures considerable
847 variation within tridactyl horses that is missed by discrete categories. Modified from Parker et al. (2018).
848 Figure 6. Morphology interacts with the environment to create a given performance, which then
849 (modulated by competition) determines fitness in that environment. Selection acts according to fitness,
850 driving evolutionary change in morphology.
851 Figure 7. Global temperature through time, with significant biotic and abiotic events highlighted.
852 Temperature, climatic event, and ice sheet data from Zachos et al. (2001); equid data from Bai et al.
853 (2018), MacFadden and Hulbert (1988), and Janis (2007), with equid and litoptern data from MacFadden
854 (1994); grassland and hypsodonty data from Strömberg (2005, 2011); pursuit predator data from Janis
855 and Wilhelm (1993). Ages indicated by annotations are approximate, and in many cases (e.g., the spread
856 of grasslands) are ± several million years. Note that oxygen isotope to degrees Celsius relationships are Evolution of a single toe in horses 37
857 calculated for an ice-free ocean, so temperature estimates are only valid until approximately 35 Ma
858 (Zachos et al. 2001).
859 Figure 8. Equid occurrences (orange, Paleocene, Eocene and Oligocene; yellow, Miocene, Pliocene, and
860 Pleistocene) across a section of North America. Size of the circle is scaled to number of occurrences. The
861 densely sampled Great Plains region is highlighted in pale orange. Sites characterized for C3 vs. C4
862 grasses (Fox and Koch 2003) are shown by magenta circles; sites characterized for grassland indicator
863 phytoliths (Strömberg 2005, 2011) are shown by blue and purple circles.