Hipparion tracks and horses' toes: the evolution of the equid single hoof

**Background:** The traditional story of horse evolution emphasizes digit reduction, culminating in the monodactyl condition of modern *Equus*. A recent hypothesis by Solounias et al. (2018) proposed that the distal portions of digits I, II, IV, and V are retained in the modern horse hoof: digits II and IV forming the V-shaped frog, and digits I and V forming the 'hoof wings' (plantar processes and collateral cartilages). This hypothesis drew on the Laetoli footprints of the tridactyl equid *Hipparion* as evidence. The present study critically evaluates this hypothesis using anatomical, developmental, and ichnological evidence.

**Methods:** The authors analyzed ungual phalanx III measurements (length and width) from 225 specimens (127 tridactyl, 98 monodactyl) and hoof print measurements from 44 prints (24 tridactyl, 22 monodactyl). Ovality index (OI = length/width) was calculated. Statistical comparisons used two-sample Welch's t-tests and Mann–Whitney U-tests. Photographs of footprints from modern equids (tapirs, Icelandic horses, Shetland ponies, Sicilian donkeys, Tennessee walking horses) were taken in various substrates and gaits. Height maps were produced via photogrammetry. The study also reviewed developmental studies of equine embryos and fetuses.

**Key Results:** - Ungual phalanx III OI: tridactyl equids (n=127) mean OI = 1.19; monodactyl equids (n=96) mean OI = 0.88 (p < 0.00001). When comparing only spring-footed equines: tridactyl (n=65) mean OI = 1.15; monodactyl (n=98) mean OI = 0.88 (p < 0.00001). Excluding domestic horses: monodactyl mean OI = 0.91 (p < 0.00001). - Hoof print OI: tridactyl (n=24) mean OI = 1.22; monodactyl (n=22) mean OI = 1.14 (not significant). Excluding hemiones and South American taxa: monodactyl (n=10) mean OI = 0.99 (p = 0.00007). - Ichnological evidence: A clear frog impression is present in a late middle Miocene (approx. 13 Ma) tridactyl equid footprint from Bila Tserkva, Ukraine (*Hippipeda parva*). V-shaped impressions at the rear of footprints from the late Miocene Hoya de la Sima site (Spain) also indicate a frog in tridactyl equids. - Developmental studies (Kavanagh et al. 2002; Franciolli et al. 2009; Barreto et al. 2016) show no evidence of distal segments of digits I, II, IV, or V in equine embryos. The 'folds' described by Solounias et al. in late-stage fetal hooves are interpreted as part of the deciduous hoof capsule (foal slippers), not digit remnants. - The 'hoof wings' in modern horses are identified as a pathological ossification of the collateral cartilage (sidebone), not a normal anatomical remnant of digits I and V. - Arterial and nervous supply of the foot follows the common mammalian pattern for a single digit, with no separate supply to the frog or collateral cartilages.

**Clinical Implications:** This study clarifies that the frog is not derived from retained side digits but is a specialized shock-absorbing structure that evolved as part of the spring foot in equids. Understanding the evolutionary origin of the frog may inform veterinary approaches to hoof care and pathology, such as frog-related lameness or conditions like sidebone. The findings also underscore that the absence of frog impressions in fossil tracks does not indicate absence of a frog, which is relevant for interpreting paleoichnological data. The ovality index provides a quantitative tool for distinguishing tridactyl from monodactyl equid footprints in the fossil record.