**Contents**


Reprinted from: *Animals* **2019**, *9*, 1058, doi:10.3390/ani9121058 .................... **117**


## **About the Editors**

#### **Matilde Lombardero Fern´andez**

Matilde Lombardero Fernandez, DVM, PhD, is a Professor at the Faculty of Veterinary Sciences ´ of Lugo, University of Santiago de Compostela, Spain. She teaches subjects related to the anatomy of domestic and exotic species, and veterinary citology and histology. Her previous academic positions were Associate Professor (1998–2005) and Assistant Professor (2006–2010). Her research activity started 30 years ago, and her scientific interests include studies on various topics from macroscopic to microscopic morphology, as well as new non-toxic preservation methods for anatomical specimens used in practical sessions of veterinary anatomy. She has undertaken various postdoctoral research stays at the Edinburgh University (Scotland, UK), Mayo Clinic (Rochester, Minnesota, USA), and at Toronto University (Toronto, Ontario, Canada).

#### **Mar´ıa del Mar Yllera Fern´andez**

Mar´ıa del Mar Yllera Fernandez, DVM, PhD, studied Veterinary Medicine at Complutense ´ University of Madrid (Spain) from 1979 to 1984. She received a PhD in Veterinary Medicine in 1988 from the University of Santiago de Compostela (Spain). Her research is focused on the study of the embryonic development of domestic animals and the anatomy of both domestic and exotic mammals, as well as new non-toxic preservation methods for anatomical specimens used in practical sessions of Veterinary Anatomy. At present, she is a Professor in the Department of Anatomy, Veterinary Medicine and Animal Production at the Faculty of Veterinary Sciences of Lugo, University of Santiago de Compostela (Spain). She teaches both the anatomy of domestic and exotic mammals, embryology and histology. Her previous academic positions include the following: pre- and postdoctoral fellow of the Spanish government (1985–1988), Assistant Professor (1988–1992) and Associate Professor (1992–1994). She has undertaken various postdoctoral research stays at the Sorbonne University (1989) and at the Institute National pour la Recherche Agronomique (INRA) in Jouy-en Josas (France) (1989–1990). She has also worked as the Veterinarian in Charge for the research animal facilities of the University of Santiago de Compostela and held the position of the President of the Bioethics Committee of the same institution.

## **Preface to "Advances in Animal Anatomy"**

During the last few years, the subject of veterinary anatomy has been gradually reduced in the successive academic programs. However, the number of species that require veterinary attention is increasing. In an attempt to bridge this gap, this Special Issue entitled Advances in Animal Anatomy comprises a set of research articles and reviews focused on the importance of the applied anatomy of diverse anatomical structures in a wide range of species, from domestic animals to wild species, including new companion animals. Therefore, this Special Issue is mainly directed at veterinary students and professionals, researchers and technicians who want to update their knowledge in their daily practice in both veterinary clinics and conservation centers.

#### **Matilde Lombardero Fern´andez and Mar´ıa del Mar Yllera Fern´andez** *Editors*

### *Editorial* **Advances in Animal Anatomy**

**Matilde Lombardero \* and María del Mar Yllera**

Department of Anatomy, Animal Production and Clinical Veterinary Sciences, Unit of Anatomy, Faculty of Veterinary Sciences, Campus of Lugo, University of Santiago de Compostela, 27002 Lugo, Spain **\*** Correspondence: matilde.lombardero@usc.es

This Special Issue was the result of reviewing Leonardo da Vinci's anatomical drawings of the bear foot and the horse trunk (among others) [1]. Since then, we were challenged to propose an interesting topic under the title "Advances in Animal Anatomy". We were convinced that we had to opt for more than just descriptive anatomy; we chose applied anatomy since, at present, the time dedicated to teaching anatomy has been drastically reduced in the undergraduate programs of veterinary degrees in favor of subjects from the clinical field/area. We also decided to address both domestic and exotic animals, which are so prevalent today in veterinary practice, as well as those animals that must be kept in rehabilitation and conservation centers. Unfortunately, veterinarians are not very familiar with the anatomy of the latter species (which always tends to be considered similar to that of domestic animals), their physiology and more frequent pathologies, since these are not part of the core training in the veterinary degrees. This deficiency is probably not exclusive to the veterinary undergraduate programs. Researchers and specialized technicians who work in wildlife recovery centers and zoos and are involved in exotic medicine and the welfare of these species that are under their surveillance, also need specialized information regarding these species' anatomies and physiologies.

In order to shed light on the applied anatomy of domestic, exotic, and wild species, we introduce the fourteen manuscripts (eleven research articles and three reviews) that are compiled in this Special Issue (Figure 1), covering current research trends in applied anatomy by using a wide range of techniques. Computed-tomographic imaging was used in four articles, magnetic resonance imaging in three of them, radiography in two articles, and ultrasound techniques and endoscopy in addition to the traditional gross dissections, morphological, and morphometrical studies were also employed (Figure 2).

The horse is the species that most articles, including three research articles and one review, in this Special Issue study. Phalanges from the equine hand were studied by Gündemir et al. [2] by using X-ray imaging. They took radiogrametic measurements of the forelimb phalanges of Arabic and throughoutbred horses. Based on the left manus data, they stated that the proximal and middle phalanges could show sexual dimorphism. These two phalanges can also be used to differentiate breeds, especially, considering the depth of the caput of the proximal phalanx, which can reaching an accuracy level of breed classification of almost 90%. Moreover, all the radiometric measurements made to the thoracic limb phalanges of 75 horses are available as valuable reference data/material when evaluating the manus digital bones in horses.

The second research article about horses focuses on equine dentistry. The most accessible maxillary teeth of the horse, the incisors, were studied by Miró et al. [3] in 25 skulls from horses aged between 12–42 months. Combining visual inspection, radiographic study and computed-tomography imaging, they investigated the development of the deciduous incisors, the dental germs of permanent incisors, and the surrounding bone, just before and up to the beginning of teeth shedding. Measurements of three lengths and two proportions or relative lengths were also analyzed. The development of deciduous and permanent maxillary incisors, in addition to their alveoli, was described in detail at different ages. It

**Citation:** Lombardero, M.; Yllera, M.d.M. Advances in Animal Anatomy. *Animals* **2023**, *13*, 1110. https://doi.org/10.3390/ ani13061110

Received: 17 March 2023 Accepted: 20 March 2023 Published: 21 March 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

was concluded that the dental germ of the first permanent incisors appears at 12 months, although their crown mineralization starts a few months later.

**Figure 1.** A tag cloud including the titles and keywords of the articles published in this Special Issue.

**Figure 2.** A mind map of this Special Issue summarizing the techniques used to study the different anatomic structures from the diverse species.

The third manuscript that focuses on horses addresses the study of the thorax in neonatal foals. Authored by Arencibia et al. [4], their study was based on computedtomography angiography (CTA) complemented with the gross dissections and sectional anatomy of the same specimens used as anatomical references. With intravascular contrast, CTA gives an overview of the thoracic morphology, providing information on the size and position of the heart, as well as the heart chambers, in addition to the information on the main thoracic blood vessels. This information could be used as reference data when comparative morphology is needed to reach a diagnosis in foals with thoracic disease. A dorsal view is preferable to visualize midline thoracic vascular structures (including pulmonary vessels and the brachiocephalic trunk), while lateral views are better to reveal the relationship between the heart chambers and the main blood vessels.

The other species of domestic mammals studied in three of the articles were the dog and the cat. The research article by Ariete et al. [5], which referred to the canine internal vertebral venous plexus (IVVP) of the lumbar segment, used contrast-enhanced computedtomography (CT) to conduct a morphometrical study of the aforementioned vessels, the dural sac, and the vertebral canal. As part of the vertebral venous plexus, this IVVP is made of a paired thin-walled and valveless venous vessels with a rhomboidal pattern, placed on the floor of the vertebral canal, and included in the lumbar epidural space. It drains blood from the vertebral column, spinal cord and its meninges, including the paravertebral muscles. As a result, when bursting, spontaneous spinal epidural hematomas are produced and could generate spinal cord compression and progressive neurological disorders. IVVP is also supposed to be involved in the pathogenesis of many other pathologies. Consequently, the accurate morphometric study of the lumbar IVVP (and surrounding structures) in healthy and alive animals is necessary to obtain reference values and to, afterwards, assess their alterations in pathological states. The measurements of the cross sectional area of the vertebral canal, dural sac, epidural space, and the right and left IVVP of six dogs were obtained, in addition to the percentage occupied by the IVVP in relation to the vertebral canal and epidural space, and the proportion of the vertebral canal occupied by the dural sac. Regarding all the measurements, there was a turning point between L4 and L5, showing a clear change in the trends at that level, which is in concurrence with the emergence of the nervous roots of the lumbosacral plexus.

Regarding the domestic cat, two review articles assessed its mandible anatomy and its manipulation when a mandibular pathology had to be treated. The first of them, Part I by Lombardero et al. [6], analyses the mandible anatomy in detail in order to obtain a deep knowledge of its morphology and avoid iatrogenic damage, caused by the fact that they are usually treated as those of small dogs. The cat mandible has fewer dental pieces than the mandible of dogs, and most of its mandibular body is filled up with dental roots and the mandibular canal (with the neurovascular supply), leaving little bone surface for the safe placement of screws in a fracture resolution. It should be emphasized that the mandibular canal is not a medullary canal and intramedullary pins should be avoided. The innervation areas of the different branches of the inferior alveolar nerve (with afferences to both skin/mucosal and tooth/periodontal structures) were also considered to serve as a reference when a mental nerve blocking must be performed. In addition, other specific considerations include the angular process in the ventrocaudal part of the mandibular ramus when the mouth remains wide open (with mouth-gags) for a considerable time, which presses on the maxillary artery. This compression reduces its blood flow (mainly directed to the brain since a functional internal carotid artery is missing in cats) producing temporary or permanent neurological disorders due to a cerebral ischaemia. Consequently, in order to prevent this complication, the use of spring-loaded mouth gags to keep their mouth wide open should be avoided in cats. Complementary to this information, Part II, also by Lombardero et al. [7], described different mandibular fractures and temporomandibular joint dislocations and how they could be solved when possible non-invasive techniques should be considered first. Otherwise, it was recommended that simple jaw fractures should be used to repair caudal to rostral, preferably, using a ventral approach. Diverse surgical methods were discussed to maintain biomechanical functionality when repairing mandibular fractures. However, taking into account that the use of rigid fixation methods, such as osteosynthesis plates, are challenging due to the scarce availability of bone surface to fix the screws onto, a new prosthesis design was proposed by the authors

to repair simple mandibular body fractures. This prosthesis proposal was a conceptual design to provide an acceptable rigid biomechanical stabilization while minimizing dental root and neurovascular damage, hence reducing patient suffering and speeding up their recovery. This prothesis would be custom-designed and would have to be manufactured in a biocompatible and resistant material, such as titanium. Its shape would look like two horizontal "Y" partially overlapped, and it would support three points of fixation with small screws, each one strategically placed to avoid damaging any tooth, periodontal structure or branch of the inferior alveolar nerve. The fourth fixation point was a flat hook-like flap embracing the body's ventral border, caudal to the mandibular fracture, thus keeping the integrity of the mandibular canal and its neurovascular supply, thus contributing to the patients' welfare. Custom-designed prostheses are utterly dependent on reliable technology, from diagnostic imaging (such as virtual surgical planning—VSP—or cone-beam computed tomography—CBCT) to computer-aided design/computer-aided manufacturing (CAD/CAM) technology. All together, they are the current trend and the trend of the near future.

The Bengal Tiger (*Panthera tigris tigris*) belongs to the same family as the domestic cat, *Felidae*, but it is exotic and much bigger. The elbow joints of the Bengal Tiger (*Panthera tigris tigris*) were studied by Encinoso et al. [8] using magnetic resonance imaging (MRI) in combination with traditional gross dissections. The elbow joint is complex and consists of a hinge joint between the humerus (whose distal end is a trochlea) and the proximal ends of the radius and ulna (with a reciprocal shape), as well as the pivot joint between the proximal ends of the antebrachial bones, all enclosed by a single capsule filled with synovial fluid. Unfortunately, not much information is currently available regarding the regional anatomy of the tiger elbow. Thus, gross dissections of the elbow joint, complemented by the brachial and antebrachial muscles and tendons, and their visualization/identification in MRI images would help us understand the normal tiger elbow anatomy, which can help us discern whether there is any elbow joint disorder in this species. The musculoskeletal system is usually studied by MRI, as it avoids ionizing radiation and it provides good image resolution and good contrast, even in soft tissues. In addition to bones, muscles, ligaments, and articular cavities filled with synovial fluid were observed with MRI, and all of these joint structures were confirmed by gross dissection. The study of the tiger elbow joint by means of MRI is essential for the proper training of experts involved in Bengal tiger conservation (such as veterinarians and researchers), as it will allow them to acquire a deep knowledge of the healthy elbow joint, and consequently, to be able to identify any alteration, reaching an accurate diagnosis, treating the patient successfully, minimizing suffering, and promoting animal welfare.

Additionally, the wild relative of the domestic dog, the Iberian wolf, was studied in this Special Issue in a study of their teeth, using a morphological and morphometric approach. This interesting dental analysis, led by Toledo et al. [9], is a comprehensive study of the dental type (incisors, canines, premolars, and molars from maxilla and mandible) morphology and morphometry, including any sexual dimorphism, and was carried out on 45 skulls of Iberian wolves (males and females) with permanent dentition. Up to 36 dental variables (including superficial and deep bite marks) were assessed and statistically analyzed to obtain a valuable morphometric data collection to establish their age and sex in a population control. In addition, the results of the analysis of their bite mark patterns (based on the tooth mark dimensions, distribution, and proportions) could be used as a database reference to differentiate between domestic dogs and Iberian wolves in forensic cases when an identification of a bite mark is needed in cases of livestock attack. This can aid in receiving economic compensation. From here on we will address the topic of mammals that live in the sea. This Special Issue includes three research articles that address the anatomical study of the whole head of dolphin fetuses and newborn dolphins, and the developmental and comparative study of some species of dolphins focused on different regions of the head (oral, pharyngeal and nasal cavities). Regarding García de los Ríos et al.'s [10] the developmental study on the head of striped dolphins during the fetal

and perinatal periods, they used different techniques, from gross dissection and sectional anatomy to diagnostic imaging (CT and MRI), to study the anatomical and physiological inferences affecting the heads of marine and land mammals. The development of the head reveals the evolutionary changes of Cetaceans that helped them adapt to the marine environment. These changes include modifications in the feeding apparatus, the caudal relocation of the nasal opening, and other structures that were reduced or even removed. Currently, there are not many methods to estimate the age of dolphin specimens during the fetal period, thus they used CT and MRI in correlation with transversal sections to study the anatomical changes of the heads of the four specimens, which occurred in chronological order during gestation. The results were organized into nine subsections covering oral cavity, rostrum, melon, nasal cavity and paranasal sinuses, orbit and eyeball, central nervous system, ear, larynx, and cranial cavity. This article is richly illustrated with photographs, so they can be used as a head morphology atlas for these species of odontocetes, giving valuable information about the changes produced in the head during prenatal and postnatal development, which are discussed in detail under an anatomical and functional perspective. This general study serves as an introduction to the next two articles, which address different structures of the head in more detail. The second article by García de los Ríos et al. [11] studied the comparative anatomy of the nasal cavity in three species of dolphins, covering the external nose and nasal cavity in different stages of development. They used endoscopy, MRI, anatomic sections and dissections, and histology, as well as the CT to generate 3D reconstructions of the nasal cavity and the bones that delimit it. The external nose is protected by two nasal lips (rostral and caudal) becoming gradually more waterproof during the development to adulthood. Their nasal cavity, usually with vertical orientation, has a single nostril continuous with a vestibule (with two vestibular folds or phonic lips, two diverticula, and two incisive recesses) and two nasal cavities (right and left) at both sides of the nasal septum, conducting the air towards the choanae. Their nasal cavity has no nasal cornettes or conchae, and it is divided in two parts: respiratory and olfactory. However, no olfactory epithelium was identified. All findings were profusely illustrated with 30 figures, each of them composed of a panel of photographs. The latest article by García de los Ríos et al. conducted a developmental study of the oral and pharyngeal cavities [12] of five species of dolphins, also based on endoscopy, MRI, anatomic sections and dissections, and histology. They increased the number of species compared to their previous work on the nasal cavity [11] and studied specimens from a wide range of ages, from fetuses, through to newborns, juveniles, and adults. They described in detail an oral cavity divided in a vestibule and an oral cavity proper, with a roof and a floor, the tongue, and teeth. They reported the pharyngeal cavity divided in three parts and confirmed the existence of the pair of pharyngeal diverticula of the auditory tubes from the early stages of development, with two areas inside: the cavity lined by respiratory epithelium (in contact with air) and its walls with a pharyngeal vascular plexus that would help to eject the air into the adjoining nasopharynx, relevant in decompression. Taken together, these three articles [10–12] present valuable descriptive and visual information about the anatomy of the head structures studied in some species of dolphins during development, relevant in veterinary medicine and some fields of biology.

Moving from mammals to reptiles, this Special Issue contains a study of the head of the monitor lizard by Pérez et al. [13], which includes advanced imaging techniques, such as CT and 3D head reconstruction in order to explore the complex anatomy of the head and associated structures of these animals. They used sagittal and transverse CT images, volume-rendered reconstructed CT images, as well as maximum intensity projection (MIP) images to map the head of the Komodo dragon. This is valuable visual work, with all the main structures of the head identified, including the comparative images from CT bone and soft tissue windows at different transversal levels, the 3D volume-rendered reconstruction images from dorsal, ventral, and lateral views, complemented with the dorsal and ventral MIP images, display the head bone structure. Apart from the bones and other structures, including muscles, nervous structures (brain, cerebellum, and brain stem), eye (different

structures, as well as sclera ossicles), glands (labial, nasal and Harderian glands), and air-filled structures (oral and nasal cavities, larynx, and trachea), the existence of cephalic osteoderms was also verified with the techniques used. This kind of morphological study based on CT imaging is very useful for people in charge of wildlife conservation centers, such veterinarians, researchers, and technicians, who have to deal with the day-to-day prevention and treatment of threatening diseases in this species in captivity.

Turning to amphibians, and to conclude the list of research articles, Ruiz-Fernández et al. [14], using non-invasive methods, such as benchtop MRI (BT-MRI) and high-resolution ultrasound (HR-US) techniques, managed to identify the biological sex of two species of Anuran in sexually mature specimens when sexual dimorphism is not apparent. BT-MRI is based on expensive equipment and the patients should be under anesthesia. The resulting images were more accurate, allowing for the identification of their sexual organs. Nevertheless, the equipment for HR-US is more affordable for veterinary clinics and zoo facilities, and has some advantages, such as that patients do not need anesthesia and that it is less time consuming. Using it, they identified the gonads of both species. The ovaries were clearly distinguished (as they appeared hyperechoic, with plenty of hypoechoic foci follicles); however, the testes were not so evident due to their homogeneous echotexture. Therefore, amphibian sexing using non-invasive techniques is a significant advance in the conservation of these species and in reproduction programs for them.

To conclude the editorial of this Special Issue, we return to the initial article about Leonardo da Vinci's drawings representing the anatomy of the foot of the bear and the horse [1]. The bear foot was considered one of his early scientific drawings and, apart from being a very common animal in Italian mountains, he probably chose to dissect the bear foot because of its plantigrade gait. Later on, his masterpieces on human anatomy were based on his lifelong learning. He was an outstanding Renaissance scientist, as he expressed throughout his life a continuous interest in learning and exploring the world around him. He also did not hesitate to go a little further. This is the purpose of this Special Issue as well, moving forward, displaying the advances of applied anatomy. We do hope to have achieved this and we wish that we all maintain our interest in broadening our knowledge in applied animal anatomy, parallel to the advances in diagnostic technology.

**Acknowledgments:** We are very grateful to all the authors for submitting their research to this Special Issue. In the same way, we thank the reviewers and Academic Editors for their expertise and wise comments that helped improve all the manuscripts.

**Conflicts of Interest:** The author declares no conflict of interest.

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

## *Review* **Leonardo da Vinci's Animal Anatomy: Bear and Horse Drawings Revisited**

#### **Matilde Lombardero \* and María del Mar Yllera**

Unit of Veterinary Anatomy and Embryology, Department of Anatomy, Animal Production and Clinical Veterinary Sciences, Faculty of Veterinary Sciences, University of Santiago de Compostela—Campus of Lugo, 27002 Lugo, Spain

**\*** Correspondence: matilde.lombardero@usc.es

Received: 10 April 2019; Accepted: 16 June 2019; Published: 10 July 2019

**Simple Summary:** Leonardo da Vinci was an outstanding artist of the Renaissance. He depicted numerous masterpieces and was also interested in human and animal anatomy. We focused on the anatomical drawings illustrating different parts of bear and horse bodies. Regarding Leonardo's "bear foot" series, the drawings have previously been described as depicting a bear's left pelvic limb; however, based on the anatomy of the *tarsus* and the *digit* (finger) arrangement, they show the right posterior limb. In addition, an unreported rough sketch of a dog/wolf *antebrachium* (forearm) has been identified and reported in detail in one of the drawings of the "bear's foot" series. After a detailed anatomical analysis, the drawing "The viscera of a horse" has more similarities to a canine anatomy than to a horse anatomy, suggesting that it shows a dog's trunk. Besides, the anatomies of the drawings depicting the horse pelvic limb and the human leg were analyzed from the unprecedented point of view of movement production.

**Abstract:** Leonardo da Vinci was one of the most influencing personalities of his time, the perfect representation of the ideal Renaissance man, an expert painter, engineer and anatomist. Regarding Leonardo's anatomical drawings, apart from human anatomy, he also depicted some animal species. This comparative study focused only on two species: Bears and horses. He produced some anatomical drawings to illustrate the dissection of "a bear's foot" (Royal Collection Trust), previously described as "the left leg and foot of a bear", but considering some anatomical details, we concluded that they depict the bear's right pelvic limb. This misconception was due to the assumption that the bear's *digit I* (1st toe) was the largest one, as in humans. We also analyzed a rough sketch (not previously reported), on the same page, and we concluded that it depicts the left *antebrachium* (forearm) and *manus* (hand) of a dog/wolf. Regarding Leonardo's drawing representing the horse anatomy "The viscera of a horse", the blood vessel arrangement and other anatomical structures are not consistent with the structure of the horse, but are more in accordance with the anatomy of a dog. In addition, other drawings comparing the anatomy of human leg muscles to that of horse pelvic limbs were also discussed in motion.

**Keywords:** bear pelvic limb; dog antebrachium; horse trunk; horse and human comparative anatomy

#### **1. Introduction**

Leonardo da Vinci was one of the most important renaissance personalities of his time, and the fifth centenary of his death will be commemorated in 2019. Being the illegitimate son of a notary, he did not continue the family saga and was educated privately. He had no formal education, thereby not conditioning his curiosity about the world around him. The erudite texts of his time were written in Latin and Greek, languages he did not master, and his access to the literature was therefore limited.

He was an artist and a scientist. As a painter, scientist, engineer and theorist, he produced thousands of drawings [1], personifying the 'Renaissance man' skilled and versed in arts and sciences [2].

His interest in anatomy was overwhelming, proven by the numerous sheets dedicated to his anatomical studies, with abundant notes and drawings, exemplifying Leonardo's principle that anatomic parts and organs should be represented in multiple views. Considering that dissections of human corpses outside Universities were not considered appropriate by the ecclesiastical authorities, he performed some dissections of animals. According to the Royal Collection Trust [3], at the outset of Leonardo's anatomical investigations, he was unable to procure much human material. Hence, many of his dissections were therefore of animals.

Practically his entire collection of anatomical drawings was compiled in the Windsor Codex, property of Her Majesty Queen Elizabeth II. These drawings of the human body were exhibited in an unprecedented exhibition in 2012 at the Queen's Gallery, Buckingham Palace (London, UK). Although previous access to the collection was highly restricted, nowadays, the Royal Collection Trust offers the possibility of free access to these drawings in high resolution on its website, which greatly enables the observation of these masterpieces and their details.

Several works have been published based on these anatomical drawings, the most exhaustive ones are those from the collection of three volumes from Clark [4,5], compiling all the inventory information, the book from O'Malley and Saunders [6] and its posterior editions in 1983 and 2003 [7,8], and the official book of the exhibition. Clayton and Philo [9] and another book published in 2013 [10], the two latter reviews, mainly referred to human anatomy, although they also include comments on some animal anatomy drawings. Apart from books, there are numerous scientific articles sharing the same subject: Leonardo da Vinci's anatomical drawings, mainly intended to some areas of expertise, such as those from Schultheiss et al. [11], Jose [12], Ganseman and Broos [13], Pasipoularides [2], Sterpetti [14], Bowen et al. [15] and West [16], among others.

It is well known that Leonardo dissected numerous animals [17]. As a result, many endeavors have been made to identify the animal of which the individual anatomical drawings have been made. In some cases, such identification is easy, while in others it is impossible [17]. Leonardo da Vinci's methods of acquiring knowledge were observation and experiment, and for him, the study of anatomy became a science, combining both the study of structure and function [12].

Reviewing the work of several authors on the description of Leonardo's animal anatomy drawings, and comparing them with the high-resolution images available on the website of the Royal Collection Trust [3], it can be noted that some of them were not properly described elsewhere, with some inaccuracies or misunderstandings that deserve to be discussed, probably due to the fact that the consulted authors were experts in human anatomy and, therefore, had no deep understanding of animal anatomy. Hence, it is important to point out that human anatomy could be considered similar, although with some differences, to animal anatomy. For major details, all of Leonardo da Vinci´s anatomical drawings can be accessed on the Royal Collection Trust website [3].

Regarding Leonardo's anatomical drawings, apart from human anatomy, he also depicted some animal species such as dogs, bears, pigs, horses, oxen and monkeys. The main aim of this comparative study on anatomy was focused only on two species: Bears and horses.

According to Forlani-Tempesti [1], da Vinci mentioned bears in his notes for his anatomical treatise: "I will discourse of the hands of each animal to show in what they vary; as in the bear which has the ligatures of the toes joined above the instep", and again: "Here is to be depicted the foot of the bear or ape or other animals to show how they vary from the foot of man or, say, the feet of certain birds" [1]. Bears are also the protagonists of other drawings of da Vinci: A bear walking (Robert Lehman Collection) and three other studies of a bear's (or a wolf's or dog's) paws (1490–1495) and head (Colville Collection) [1]. However, those images of paws cannot be from a bear, simply because bears have a *manus* (hand) with five *digits* (fingers) with their *distal phalanxes* (claws) in contact with

the floor. In contrast, a dog and wolf *manus* only have four *digits* ending in *unguicula* (claws) in contact to the ground, plus the *digit I* (dewclaw) medially placed at the level of the metacarpal bones.

#### **2. Flaws of the Anatomy of the Bear's** *Pes* **(Foot) and the Hidden** *Antebrachium* **(Forearm) and** *Manus* **from a Dog**/**Wolf**

The set of drawings that made us realize that some inaccuracies were made in terms of their description was that of the bear's foot (Royal Collection Inventory Number—RCIN 912372-5). Regarding RCIN 912372 (Figure 1), the first reference to it was stated by Professor William Wright in 1919 [18], in a section entitled 'Leonardo as an Anatomist', published by the Burlington Magazine to commemorate the Quartercentenary of Leonardo da Vinci [18] as 'one of the finest of Leonardo's anatomical drawings, the hind foot of a plantigrade carnivorous animal—probably a bear, a view supported by the fact that in one of the manuscripts, a reference is made to a bear's foot'.

**Figure 1.** Bear's foot series—Number 1. Bear distal right pelvic limb/*pes*, medial aspect. A bear's foot c.1488–1490. Modified from www.rct.uk/collection/912372 (Royal Collection Trust [3]). This image is credited as Royal Collection Trust/© Her Majesty Queen Elizabeth II 2019.

The reference made to this figure (RCIN 912372) by Castiglioni [19] describes it as 'Studio di anatomia del piede umano, con artigli al posto delle unghie', a statement that clearly evidences its misinterpretation. Previously, it was catalogued as the foot of a monster, as reported later by Clark [4]. The description of Leonardo's bear's foot drawings (1488–1490), made by Clayton and Philo [9] (RCIN 912372; Figure 1), indicated that "Leonardo dissected the left hind leg of a bear ... The drawings show the bones, muscles and tendons of the lower leg and foot", in accordance to O'Malley and Saunders [6], whose Comparative Anatomy Chapter states that "the drawings represent a dissection of the left leg and foot of a bear as originally pointed out by the anatomist, William Wright. There can be no question that the identification is correct". It seems, indeed, to be a bear's foot. In accordance, the description of the same drawing at the Royal Collection Trust website [3] states "this drawing shows with some accuracy the bones, muscles and tendons of a bear's lower leg and foot, with the big toe, claw raised, away from the viewer". The bear, as a plantigrade animal, walks like humans, with the whole plantar face of the *pes* (the sole with the heel) touching the ground. However, in contrast to humans, the shortest *digit* (toe) is not the fifth one (the lateral one), but the medial one, that is the first *digit* [20,21]. Hence, to the best of our knowledge, we support that the bear's foot depicted by Leonardo corresponds to the right hind limb instead of the left one, as previously reported by O'Malley and Saunders [6] and Clayton and Philo [9] as well as at its description at the Royal Collection Trust website [3], maybe in resemblance to humans. In addition, the *calcaneus* bone of the *tarsus* is always in a lateral position to the *talus*, and the medial projection of the *calcaneus* bone to support the *talus* is quite visible, the so-called *sustentaculum tali*, with a groove to the tendon of the muscle *flexor digitalis lateralis* [22]. Figure 1 also shows, on the left-centre, a rough sketch of some muscles that continues beneath the represented bear's *pes*. If observed upside-down (Figure 2), it seems an *antebrachium* (forearm) and *manus* of another animal. Considering all the elements depicted (muscles' shape, the bone and the carpal orientation as well as the preliminary draft of the *manus*), we think it corresponds to a caudomedial/palmar view of the left *antebrachium* and the palmar view of a *manus* provided of short *digits* (at least *digits* shorter than those of humans, without an opposable *digit I* (thumb)). At the *carpus* level, the *retinaculum flexorum* is still patent to some extent, as a thick transverse fascial band from the medial carpal bones to the *os carpi accessorium*, forming the *canalis carpi* (canal between the proximal row of carpal bones and *retinaculum flexorum*) [22], although it has partially been removed to liberate the tendon of the muscle that seems to be the *flexor carpi radialis*, which appears to be attached at the base of the medial metacarpal bones. However, its deepest part still keeps the tendon of what seems to be the *M. flexor digitalis superficialis* at its place. This *antebrachium* and *manus* sketches are quite slim, suggesting being depicted from a dog or a wolf more than from a bear, with a more robust *antebrachium*. The proportions of the *antebrachium*/*manus* length are also more coincident with those of a dog/wolf than of a bear (with a relatively short *antebrachium*, but longer *manus*). The oblique line crossing the medial face of the radius might represent the *Vena cephalica*, which joins the *V. cephalica accessoria* (it runs dorsomedial to the *metacarpus* and *carpus*) to conform the *V. cephalica*, which continues cranially along the antebrachium. However, if that line represents the *V. cephalica*, it should have been a bit more distal (closer to the carpal joint). To the best of the authors' knowledge, this sketch was not previously described in any of the consulted literature, maybe because the ultimate detailed representation of the bear *pes* drawing catches the entire attention of the observer.

**Figure 2.** An *antebrachium* of a dog/wolf. Dog/wolf *antebrachium* and *manus*, caudomedial/palmar view. A bear's foot c.1488–1490. Modified from www.rct.uk/collection/912372 (Royal Collection Trust [3]). This image is credited as Royal Collection Trust/© Her Majesty Queen Elizabeth II 2019.

Another one of Leonardo da Vinci's bear *pes* drawings, RCIN 912373 recto (Figure 3), described at the Royal Collection Trust [3] as "a study of the dissection of the leg and foot of a bear, viewed in profile to the left", corresponds to the bear's right *pes* shown in a medial aspect (from the left). This sheet also includes an inset in the upper left, representing one *digit* viewed in profile, where the tendon of *M. flexor digitalis superficialis* fixes at two points in the second *phalanx*, between which a hole is provided, allowing the tendon of the *M. flexor digitalis profundus* to fix distally on the flexor surface of the third *phalanx* of the *pes digit*, a structure perfectly depicted by Leonardo da Vinci.

**Figure 3.** Bear's foot series—Number 2. Upper left: Bear *pes digit*. Down right: Bear distal right pelvic limb/*pes*, medial aspect. A bear's foot c.1488–1490. Modified from www.rct.uk/collection/912373 Recto (Royal Collection Trust [3]). This image is credited as Royal Collection Trust/© Her Majesty Queen Elizabeth II 2019.

The metalpoint drawing RCIN 912374 (1488–1490; Figure 4), "a study of the dissection of the leg and foot of a bear, viewed in profile to the right", shows the lateral side, with the perfectly defined *fibula* and three tendons associated to its distal end (*malleolus*), one in the lateral face belonging to the *M. peroneus longus* and two sliding caudal tendons of *M. peroneus brevis* and *M. extensor digitalis lateralis*. In addition, the lateral *digit* is one of the longest. Taking all this information into account, it could be concluded that the bear's *pes* represented in this drawing is again the right one.

**Figure 4.** Bear's foot series—Number 3. Bear distal right pelvic limb/*pes*, lateral view. A bear's foot c.1488–1490. Modified from www.rct.uk/collection/912374 (Royal Collection Trust [3]). This image is credited as Royal Collection Trust/© Her Majesty Queen Elizabeth II 2019.

The last one of da Vinci's bear *pes* drawings of this series, catalogued as RCIN 912375: "A bear's foot" (Figure 5), shows an "outside view of the foot, partially from below" [9], where the tibia is visible and the *calcaneus* is lateral to it, corresponding also to the right bear's *pes*, an aspect not revealed in former descriptions.

**Figure 5.** Bear's foot series—Number 4. Bear distal right pelvic limb/*pes*, plantaromedial oblique view. A bear's foot c.1488–1490. Modified from www.rct.uk/collection/912375 (Royal Collection Trust [3]). This image is credited as Royal Collection Trust/© Her Majesty Queen Elizabeth II 2019.

In summary, all bear's *pes* sheets depicted by Leonardo da Vinci correspond to the right pelvic limb. However, the preliminary draft of the *antebrachium* and *manus* of a dog/wolf illustrates the left one. It is clear that Leonardo da Vinci, on many occasions, used the sheets without an order. This fact leads to the interpretation made by Keele [17] relative to Leonardo's method of research: "he used pages of notes on particular subjects as we would use a filing system, returning to the same page at intervals of weeks, months, in some cases as long as twenty years later to record further drawings or verbal notes on the same subject".

#### **3. Anatomy of the Horse Trunk that Turned into a Dog's Trunk**

Later on, according to Clayton and Philo [10] "Writing in the mid-sixteenth century, the biographer Giorgio Vasari stated that Leonardo compiled a treatise on the anatomy of the horse. One drawing of the viscera of a large quadruped, probably a horse, does survive from this period, suggesting that Leonardo conducted full dissections to investigate the internal anatomy of the beast. But Vasari also stated that the treatise on the horse was lost when Milan was invaded by French forces in 1499. Ludovico Sforza was overthrown, and soon afterwards, Leonardo left the city and returned to Florence" [10]. This text refers to the drawing RCIN 919097-recto, entitled 'The viscera of a horse' (1490–1492; Figure 6), and described at the Royal Collection Trust [3] as: "an anterior view of the arteries, veins and the genito-urinary system of an animal, probably a horse," implying that Leonardo did not name this

drawing. The drawing represents the ventral aspect of the trunk of an animal (supposedly, a horse) with the lungs and the *canalis alimentarius* (esophagus, stomach and intestines) removed. The large vessels depicted at the centre, all the figure down, represent the *aorta* (on the right of the figure) and the vena *cava caudalis* (on the left of the image). The way they ramify helps us to discard the idea that this drawing represents a human being. In humans, the *aorta* ends up dividing into two branches: *Aa. iliaca commune* (*dextra* and *sinistra*). In contrast, in animals, the end of the *aorta abdominalis* (at the level of the pelvic limbs) produces two branches (*external iliac arteries*—*dextra* and *sinistra*), well depicted, and subsequently continues and produces two more (*internal iliac arteries*), well represented in the drawing, ending as the *arteria sacral median*, not depicted. Regarding the veins, the *vena cava caudalis* is formed by the confluence of two *Vv. Iliaca commune*—*dextra* and *sinistra*, each one resulting from the junction of the *V. iliaca externa* and the *V. iliaca interna*, following a similar pattern both in humans and domestic mammals.

**Figure 6.** *Cont.*

**Figure 6.** The viscera of a horse? (**A**) Whole drawing representing the ventral aspect of the trunk of an animal with the *canalis alimentarius* and lungs removed. (**B**) Inset at higher magnification depicting the lumbar and pelvic regions. The viscera of a horse c.1490–1492. Modified from www.rct.uk/collection/ 919097 recto (Royal Collection Trust [3]). This image is credited as Royal Collection Trust/© Her Majesty Queen Elizabeth II 2019.

Regarding the blood vessels, the drawing (Figure 6) provides three key points: (a) The first huge vessel (on the left of the image), reaching the heart, could be the *Vena cava cranealis*, and the other curved vessel going down is the *aorta* and its *arcus aortae*, with two big arteries leaving the aortic arch and some smaller ones (2–3) once the arch finishes and continues to the descendent aorta (*aorta descendens*). Large domestic mammals (horses—Eq, and ruminants—Ru) only have one artery deriving from the aortic arch, the *truncus brachiocephalicus*, which is then divided into a *truncus bicaroticus* (could be absent in carnivores—dogs and cats) and two *Aa. subclaviae*. In contrast, carnivores and pigs (Su) have two arteries leaving the *arcus aortae*: The *truncus brachiocephalicus* first and secondly the *A. subclavia sinistra*. (b) On the other hand, at the kidney level, there are two arteries perfectly outlined in the drawing stemming from the *aorta*: The *A. circumflexa ilium profunda* (*dextra* and *sinistra*), exclusive to carnivores [22] and dividing into the *rami craniales* and *caudales*. In contrast, the *Aa. circumflexa ilium profunda* derives from the *A. iliaca externa* in Su, Ru and Eq [22], similar to humans [23], not stemming directly from the aorta. (c) The arteria and vena *circumflexa ilium superficiales*, the first branches of the *A. femoralis* and *V. femoralis*, respectively, are exclusive to carnivores [22]. They leave their main vessels cranially oriented, at the medial and proximal part of the thigh. These vessels (a–c) in this drawing are the main clue to determine the species. Consequently, the horse representation/provenance of this drawing could be discarded. However, the horse is the unique domestic species in which the aorta does not end caudally as an *arteria sacral median*, which is not represented in the illustration.

In order to elucidate the identity of the depicted specimen, more elements were analyzed. The testicles in that position (*regio urogenitalis*-ventral pelvic region) could be mainly those of a dog. Male cats have the scrotum placed at the perineum, similar to boars, while the scrotum position of horses and bulls is more inguinal. The kidneys are not depicted as those from a horse (heart-shaped the *ren dexter* and more irregular the *ren sinister*). Those from pigs are more symmetrical and flattened, while those depicted are very similar to those from a dog. In contrast, cat kidneys have some vessels on the surface, with a radial arrangement toward the renal hilus, called capsular veins (*venae capsulares*), exclusive to cats [22], not represented in the drawing.

The hanging organ below the heart must be the liver. Accordingly, it has some very deep *incisurae interlobares* common for pigs and dogs/cats. At first glance, in between the liver and the renal arteries, it seems to be the lumbar part of the diaphragm (*crura*), but at higher magnification, it looks like an odd branch stemming from both the *aorta* and the *V. cava caudalis* (it is not clear in the drawing, maybe da Vinci had his doubts about this issue) that splits into three. There is a possibility that this vessel represents the *A. celiaca* with its three branches illustrated (*A. gastrica sinistra*, *A. hepatica* and *A. lienalis*). In that case, the organ below the heart cannot be the liver or was represented misplaced to give leadership to other, more relevant structures. Similarly, the penis has also been removed to expose the pelvic organs.

In conclusion, these details led us to confirm the hypothesis that this drawing does not represent the anatomy of a horse, as previously reported, since most of the anatomical elements are consistent with the open chest, abdomen and pelvis of a carnivore, probably a dog rather than a cat.

#### **4. More on the Comparative Anatomy of Humans and Horses: The Case of the Horse and Human Anatomy of Their Pelvic Limb and Leg, Both Standing and Walking Forward**

Continuing with horses, Leonardo da Vinci drew some sketches comparing the horse and human anatomy in terms of their pelvic limbs and legs, both standing and walking forward. The drawing entitled 'The leg muscles and bones of man and horse' (RCIN 912625; Figure 7) is described as "The muscles of a man's legs are here studied in several views, together with the bones of the pelvis and legs, with 'cords' indicating the lines of action of the muscles". At the lower centre is a diagram of the same structures in the horse, with the astute note that "to match the bone structure of a horse with that of a man you will have to draw the man on tip-toe" [3]. In 1919, and according to Wright [18], these drawings "serve to illustrate Leonardo's methods, referred to previously, of analysing a region into its elements and of making use of comparative anatomy". However, Wright described those illustrations as "the left hindlimb of an animal, probably a dog, and the left lower limb of a man, both drawn in the natural standing posture and both showing in the upper parts strips of corresponding muscles" [18]. Although the mentioned species of dog does not properly match with the skeleton morphology represented in this sheet, mainly due to the coxal bone (*os coxae*) shape, the presence of the third trochanter and the long metatarsal bone. Referring to these drawings, Keele [17] reported "Leonardo's interest in movement extended from those of man to animals", and relative to these figures, Leonardo "compares the bones of a man's leg with those of a horse when standing, saying that to compare the two, the man must be shown standing on tiptoe. When they walk forward, this becomes even more evident". Later on, Clayton and Philo [9] stated that "Leonardo's study of the horse here was to some degree compromised by his knowledge of human anatomy: the pelvis is too upright and not long enough, and the femur is too long and thin". We are in accordance with those authors: The pelvis is represented shorter than it should be, and its natural position should be a bit more horizontal. In addition, the femur should be more robust (Figure 7).

**Figure 7.** Series of comparative anatomy of man's leg and horse's pelvic limb—Number 1. (**A**) Whole drawing. (**B**) Inset at higher magnification showing a horse left pelvic limb (on the left) and a human left leg (centre and right), lateral aspect. The leg muscles and bones of man and horse c.1506–1508. Modified from www.rct.uk/collection/912625 (Royal Collection Trust [3]). This image is credited as Royal Collection Trust/© Her Majesty Queen Elizabeth II 2019.

Besides, the lumbar vertebrae, which should be five or six, have large horizontal transverse processes that are not illustrated, neither are the high spinous processes present at the lumbar vertebrae and at the *os sacrum* shown, composed of five fused sacral vertebrae. In addition, the number of the lumbar and sacral vertebrae is not accurately illustrated: Three to four at the lumbar region and also at the sacrum. Moreover, the tibia and fibula are represented as bones of the horse's leg (crus), but quite inaccurately, showing the fibula as long as the tibia, although horses only have a head and rudimentary body of the fibula (*caput* and *corpus fibulae*) [22] in a lateral position of the tibia, which barely reaches the half tibia and never articulates with the tarsus. O'Malley and Saunders [6] stated that "These three figures are for comparative purposes ... The supposedly corresponding muscles of the horse are shown ... but very inexactly. Presumably the cords represent the adductor, sartorius, tensor fasciae latae, gluteus superficialis or gluteus medius and gluteo-biceps of the horse". Clayton and Philo [9] also stated that "A few muscles are represented by threads: rectus femoris from the anterior iliac spine to the patella; tensor fasciae latae from roughly the same point towards the lesser trochanter of the femur (in the horse this trochanter is much more prominent than in the human); and the gluteal muscles, represented by a number of threads (two in the horse, four in the human) running from the iliac crest towards the greater trochanter". It is indeed the caudal part of a horse skeleton in profile because of the *os coxae* morphology, and its femur has a third trochanter (*Trochanter tertius*) specific for horses [22] (Figure 7). Hence, Clayton and Philo's description [9] is quite inaccurate, because the lesser trochanter is placed medially with respect to the major trochanter (below the head and neck of the femur, so it cannot be seen from this lateral view), and the detail they described as lesser trochanter is, in fact, the third trochanter (*Trochanter tertius*), in which only one muscle attaches: The *M. gluteus supeficialis* [22]. On the other hand, in horses, the *M. gluteus supeficialis* should extend from the *fascia glutea* and the *os sacrum* to the *trochanter tertius*, it has a cranial head originating from the *M. tensor fasciae latae* [22]. As the unique muscle that attaches to the third trochanter is the *M. gluteus superficialis*, it is obvious that Leonardo has represented only one of its first attachments (at the *tuber coxae*, where the *M. tensor fasciae latae* comes from). Consequently, in Eq, there could be some confusion in terms of the extent of the *M. tensor fasciae latae* as it starts together with one of the heads of the *M. gluteus superficialis* (fixing at the third trochanter) to the *fasciae latae*, but in quadrupeds, never to a bone fixation as reported by Clayton and Philo [9], because it tenses the *fasciae latae*, which is one of the *M. biceps femoris* fixations.

According to Schaller [22], the *M. gluteobiceps* described by O'Malley and Saunders [6] is inexact because only Su and Ru present it (resulting from the fusion of the *M. gluteus superficialis* and the cranial portion of the *M. biceps femoris*); consequently, it does not appear in horses.

In addition, in domestic mammals, the *M. rectus femoris* (as part of the *M. quadriceps femoris*) arises from cranial to acetabulum areas [22], but not from the ventral area of the *tuber coxae* (*os ilium*), as Leonardo depicted and as Clayton and Philo stated [9]. However, it could be the *M. sartorius* as reported by O'Malley and Saunders [6], whose attachments, from the *fascia iliaca* (in Ungulates) to the medial side of the proximal portion of the tibia and *fascia cruris* [22], match Leonardo's drawing. In humans, the *M. sartorius* runs from the *spina iliaca anterior superior* to medial to the *tuberositas tibiae* [23]. Hence, the longest thread depicted by Leonardo does not perfectly match the description. Regarding the vertical inner muscle (thread), it seems to start at the *symphysis pelvina* and to end medially and distally in the femur. This location is compatible with (a) the *M. gracilis* (flat superficial adductor of the thigh) from the *symphysis pelvina* (by *tendo symphysialis*) to the *fascia cruris* on the medial surface of the proximal portion of the crus [22], but it is placed caudal to the femur or, (b) the possibility that the *M. adductor* 'undivided in ungulates, from the *tendo symphysialis*, *ramus caudalis ossis pubis* and *ramus ossis ischia* to the *fascies aspera*, in Eq also *condylus medialis femoris*' [22]. In accordance to O'Malley and Saunders [6], it seems the latter (*M. adductor*) is the muscle that better fits to Leonardo's representation. In humans, there are the *M. adductor longus* and *magnus* from near the *symphysis* and the *tuber ischiadicum* and *ramus ischii*, respectively, to the *labium mediale* of *linea aspera* both, and the *magnus* additionally to the *condylus medialis* of the femur [23]. As in a profile view from lateral, it is not possible to see the distal attachments of the represented threads; it is therefore difficult to infer which ones of the adductor muscles are depicted.

Clayton and Philo [9] also stated that "the drawing at lower right has been called an 'anatomical fantasy', blending the bones of a horse with those of a man. It is much more likely that Leonardo intended it to be purely human, but incorporated errors (in particular, the extended ischium below the coccyx; cf. no. 64b, in which the error is corrected), derived from his superior knowledge at that date, of equine anatomy-the study of human bones and muscle threads to the left display the same errors". Should it say: ... because of his superior knowledge of human anatomy. O'Malley and Saunders [6] referred to this figure as 'the elongation of the innominate bone and the length of the coccyx suggest the pelvis of an animal rather than of a man. In addition, the observer will note a trochanter tertius below the greater trochanter. From these appearances, this figure seems to have been derived by the expansion of animal bones to the approximate proportions of the human'. Regarding RCIN 919012 recto: The skeleton c.1510–1511 (figure not included), drawn 2 to 5 years later than RCIN 912625 recto (Figure 7), Clayton and Philo [9] stated that "Leonardo has here corrected the length of the ischium", referring to the lower right figure. According to Wright [18], these drawings "serve to illustrate Leonardo's methods, referred to previously, of analyzing a region into its elements and of making use of comparative anatomy".

There are other illustrations depicting the comparative anatomy between the man and the horse, belonging to the 'Manuscrit K' from the Bibliothèque de l'Institut de France [24]. One of them, folio 102 recto (Figure 8), compares the right horse hind limb (in a lateral view) with the right human leg in a frontal view. The drawing at the left centre (Figure 8B) is without doubt the laterocaudal view of the right pelvic limb of the horse, and again he has drawn a fibula that extends much too far distally along the caudal aspect of the tibia for the horse (this fact adds to the evidence that the limb depicted in Figure 7 is indeed an equine limb).

Nevertheless, the most impressive drawing in 'Manuscrit K' about this issue corresponds to Folio 109 verso (Figure 9), in which both sketches are reported to compare a man's leg with a horse´s hind limb when walking forward, for which the man should be on tiptoes [17]. The illustration shows part of the trunk and the left leg of a man seen in left profile and compared with the left posterior limb of a horse, but not keeping a relative size/proportion.

Here, however, it is not clear that the represented man is walking forward, because the flexed knee during walking is concurrent with tiptoe contact to the ground, but not the trunk inclination/angle as depicted. This position is similar to that when a man or a horse is taking impulse from below (flexing their legs or hind limbs) to be able to jump. These illustrations exemplify what O'Malley [7] stated about Leonardo's interests: "they are directed towards the structure of the body in relationship to its workings". Maybe Leonardo tried to show a parallelism between the relative positions of bones and muscles of the horse, captured and transferred to a human skeleton in order to describe and understand the combination of the flexion of the joints together with the muscle action to accumulate the energy needed on the impulse of jumping, as it is described in the text of the following page of the manuscript K [24] (Folio 110 (30) recto: 'saut de l'homme'; compare Figure 7 on standing position vs. Figure 9 on flexion; Figure 8 seems to be a previous sketch to Figure 9).

The main difference between the drawing at the Royal Collection Trust [3] (Figure 7) and those from 'Manuscrit K' [24] (Figures 8 and 9 ) is that the muscle is represented as an inverted 'V', illustrated as different threads in the Royal Collection Trust. The inverted 'V' muscle has its apex at the *tuber coxae*, and the shorter part fixes laterally at the third trochanter. The longest part goes to the *condylus medialis* of the tibia in one of them (Figure 9), but it is not clear in the other folio (Figure 8) because its fixation should have been out of the sheet (lower sketch) or is not drawn with enough detail (upper sketch), indicating a rough study previous to Figure 9. In addition, just analyzing the depicted muscles and threads of these illustrations, it is clear that in Manuscript K [24] (Figures 8 and 9), the muscles are represented as a wide thread with a sort of 'belly'. However, in the drawing RCIN 912625 (Figure 7), slimmer threads are depicted; hence, it could be deduced that the sketches from Manuscript K were made previously to the drawing of the Royal Collection Trust, just as an evolving progress in his studies, a trend in all artists´ lives, as a result of a lifelong development in their conceptualization and

abstraction ability/aptitude. However, comparing the contents of both illustrations (Figures 7 and 9), it is not likely that Leonardo started to study the flexed hind limb to approach its standing position later. These facts should be taken into account when dating back to Leonardo's work, as those sheets are assigned more or less to the same periods of c.1506–1508 (Figure 7) and c.1503–1508 (Figures 8 and 9).

**Figure 8.** Series of the comparative anatomy of a man's leg and horse's pelvic limb—Number 2. (**A**) Two pages. (**B**) Inset: Higher magnification of Folio 102(22) recto: 'Anatomie, cheval'. Modified from Ms 2181-folio 102(22)-manuscrit K (Bibliothèque de l'Institut de France [24]) (1503–1508). This image presentation is authorized in limited time period by Réunion des Musées Nationaux-Grand Palais (R.M.N.-Grand Palais), and the Bibliothèque de L'Institut de France. Photo © RMN-Grand Palais (Institut de France)/René-Gabriel Ojéda.

**Figure 9.** Series of the comparative anatomy of a man's leg and horse's pelvic limb—Number 3. (**A**) Two pages: Folio 109 verso (red) and folio 110 recto (violet inset with da Vinci notes about 'Leviers, movements, saut de l'homme'). (**B**) Folio 109 verso: 'Anatomie comparée de l'homme et des animaux (chevaux)'. Red inset upside down at higher magnification displaying the comparative between the man left leg (bottom, left) and the horse left pelvic limb, lateral aspect. Modified from Ms 2181-folio 109(29-30) and 110 (30)-manuscrit K (Bibliothèque de l'Institut de France [24]) (1503–1508). This image presentation is authorized in limited time period by Réunion des Musées Nationaux-Grand Palais (R.M.N.-Grand Palais), and the Bibliothèque de L'Institut de France. Photo © RMN-Grand Palais (Institut de France)/René-Gabriel Ojéda.

#### **5. Conclusions**

Leonardo da Vinci was an outstanding artist and scientist, looking for the reasons and mechanisms of all the aspects he had studied during his entire life. He tried to find the common clues shared by human and animal anatomy through comparative anatomy. He produced an impressive collection of anatomical drawings, which were, in general, quite accurate. However, some later studies have misunderstood some of them. Hence, a deep anatomical insight into the bear and horse anatomical drawing collections revealed some inaccuracies, which, for some anatomical elements, we identified and amended. Such is the case of the depicted 'bear's foot' series, described as the left hind limb, but based on their *digits* and *tarsus*, it is the right one. Regarding the horse anatomy, one of Leonardo's drawings illustrating the internal and dorsal parts of the horse trunk, if considered the different blood vessels depicted and other viscera, seems to be from a carnivore (probably a dog) rather than that of a horse, as previously reported. Other drawings illustrating the comparative anatomy of the horse hind limb and the human leg were also reconsidered in a new approach, assuming that Leonardo da Vinci used the comparative anatomy in order to understand the process to produce movement, especially when jumping.

**Author Contributions:** M.L. and M.d.M.Y. participated equally in the conception/design, interpretation, drafting the manuscript, critical review of the manuscript and approval of its final version.

**Funding:** The authors declared they received no financial support for their research and/or authorship of this article.

**Acknowledgments:** The authors are thankful to Royal Collection Trust website, Biblioteca Digitale from Museo Galileo-Firenze-Italia and Bibliothèque de l'Institut de France website for the possibility of free access to their Leonardo da Vinci´s digital collection. In addition, we thank the Royal Collection Trust for granting their images for free.

**Conflicts of Interest:** The authors declared that they had no conflict of interests with respect to their authorship or the publication of this article.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Radiogrametric Analysis of the Thoracic Limb Phalanges in Arabian Horses and Thoroughbred Horses**

**Ozan Gündemir 1, Tomasz Szara 2,\*, Gülsün Pazvant 1, Dilek Ol ˘gun Erdikmen 3, Sokol Duro <sup>4</sup> and William Perez <sup>5</sup>**


**Simple Summary:** The aim of the research was to determine the radiogrametric features differentiating the phalanges of the thoracic limbs of Arabian horses and Thoroughbred horses including sexual dimorphism. Nine traits and three indexes were analyzed. Radiological measurements of phalanges showed that sexual dimorphism is not clearly marked and differences between breeds manifest themself mainly in the proximal phalanx measurements. None of the parameters tested in Thoroughbred horses differed significantly between males and females. The discriminant analysis enabled the correct classification of 89.33% of the proximal phalanx samples to the exact breed. This percentage was 77.33% in the case of the middle phalanx and 54.67% for the distal phalanx, respectively. The data obtained from this study can be used as a reference material for the radiogrametric evaluation of the skeleton of the manus in horses.

**Abstract:** In this study, it was aimed to determine the statistical differences between Arabian horses and Thoroughbred horses based on X-ray images of forelimb digital bones. Latero-medial X-ray images of digital bones of thoracic limbs were taken of 25 Arabian horses and 50 Thoroughbred healthy horses. The difference between males and females within the breed was statistically analyzed as well. Nine measurements and three indexes taken from phalanges of thoracic limbs were used. Thoroughbred horses did not differ significantly between sexes, as indicated by the ANOVA. For the Arabian horses, the length of the middle of the proximal phalanx (*p* < 0.05), the length of the middle of the middle phalanx (*p* < 0.001), and the length of the dorsal surface of the distal phalanx (*p* < 0.05) measurement points were found to be differentiated between sexes. In the analysis made between Thoroughbred horses and Arabian horses with no respect to sex, the critical measurement was the depth of the caput of the proximal phalanx. The discriminant analysis enabled the correct classification of 89.33% of the proximal phalanx samples to the exact breed. The correct classification rate was 77.33% in the case of middle phalanx and 54.67% in the case of distal phalanx. Measurement results of the distal phalanx were found to be insignificant between both breeds and sexes. The radiological measurements of digital bones showed that sexual dimorphism was not too expressed and that decisive differences were found between the breeds.

**Keywords:** horse; phalanx; radiogrametric study; veterinary anatomy

#### **1. Introduction**

Radiographic imaging techniques have become more readily available, and are now widely used in veterinary environments. Beside clinical use, these image data are used in

**Citation:** Gündemir, O.; Szara, T.; Pazvant, G.; Erdikmen, D.O.; Duro, S.; Perez, W. Radiogrametric Analysis of the Thoracic Limb Phalanges in Arabian Horses and Thoroughbred Horses. *Animals* **2021**, *11*, 2205. https://doi.org/10.3390/ani11082205

Academic Editor: Chris W. Rogers

Received: 27 May 2021 Accepted: 23 July 2021 Published: 26 July 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

veterinary anatomy as educational tools, and for differentiating breeds [1–4]. Quantitative radiography allows not only subjectively assessment of the condition of the digital organ in particular individuals, but to learn about the changes that the phalanges undergo with age [5]. Radiological examination is crucial in revealing the source of orthopedic problems in horses [6,7]. In particular, a significant number of cases of lameness in thoracic limbs may be due to the non-optimal alignment of phalanges [8]. Various studies have been carried out to examine the position of phalanges relative to each other and the angular positions between them radiogrametrically [9–11]. In addition, X-ray images have been used to evaluate the general joint conformation of healthy animals, not just in suspicion of lameness [12]. Radiographic techniques can provide information regarding relationships between the hoof capsule conformation and the geometry of the distal phalanx [13,14]. Cohen et al. [15] attempted to determine prospectively the association between the abnormal radiographic findings and performance indicators in English Thoroughbred horses.

As stated above, the positions and conformations of phalanges are very important for clinical diagnosis. In this regard, it is thought that the reference information obtained from healthy animals will be valuable for future studies on this subject [16]. In addition, while collecting radiometric reference data on digital bones of the thoracic limbs of healthy animals, we tried to answer the following study questions:

Is there any difference in radiometric measurements of digital bones of the thoracic limbs of Arabian horses and Thoroughbred horses?

Can gender determination be made in radiometric measurements of digital bones of the thoracic limbs?

#### **2. Materials and Methods**

The study was conducted with 25 Arabian horses (7 females, 18 males) and 50 (15 females, 35 males) Thoroughbred horses aged between 2 and 8 years, brought to the Jockey Club of Turkey for control purposes. We used the unit of age as years. Horses that had no musculoskeletal or orthopedic problems before and actively participated in races and training were used in the study. Latero-medial radiological images of the digital bones of their thoracic limbs were acquired with Gierth X-ray (TR9030). The difference between the right–left manus was ignored, and only the images of the digital bones of the left thoracic limbs were used for analyses.

Three measurements of each of the proximal phalanx, middle phalanx, and distal phalanx were taken (Figure 1). Measurements were recorded using the Radiant DICOM Viewer (version 2020.2.2) software:

Proximal phalanx:


Middle phalanx:


Distal phalanx:


**Figure 1.** Measurement (4-year-old male Arabian horse). (**A**): Depth of the basis of the proximal phalanx (DBP), depth of the caput of the proximal phalanx (DCP), depth of the basis of middle phalanx (DBM), depth of the caput of the middle phalanx (DCM), length of the caudo-dorsal surface of the distal phalanx (FA). (**B**): Length of the middle of the proximal phalanx (LMP), length of the middle of the middle phalanx (LMM), length of the dorsal surface of the distal phalanx (LD), greatest solear length of the distal phalanx (GSL).

SPSS (version 22) was used for statistical calculations. Differences between Arabian and Thoroughbred horses in terms of the morphometric features were evaluated separately for proximal phalanx, middle phalanx, and distal phalanx. Male and female samples of the two breeds were evaluated separately. In addition, measurement differences between the two horse breeds were analyzed regardless of gender. ANOVA was used for these analyses. Homogeneity of variances was examined, and *p* values were obtained. Discriminant function analysis was applied to reveal the differences between Arabian horses and Thoroughbred horses. Eigenvalue and Wilks' lambda values were examined. A confusion table was used to evaluate the validity of DFA results. The correctly classified values and a discriminant distribution table were obtained with the Past (4.01) statistics program. Correlations between variables of the phalanges and between phalange traits and horse age were examined as well.

#### **3. Results**

The mean values of particular traits and standard deviation determined separately for the males and females of Arabian horses and Thoroughbred horses are presented in Table 1. The measurement points selected for Thoroughbred horses had no significant effect on the difference between sexes, as indicated by ANOVA results. The distribution of the samples between both breeds and sexes was homogeneous. Trait values in males were found to be higher for the proximal phalanx and middle phalanx. In Arabian horses, LMP (*p* < 0.05), LMM (*p* < 0.001), and LD (*p* < 0.05) were found to be significantly different between sexes. Values of all traits and indexes were found to be higher in the males of Arabian horses. According to the index results used, index 3 was a determinant between sexes only in Arabian horses (*p* < 0.05).

Table 2 shows the mean values, standard deviations, and statistical differences in phalangeal traits between Arabian and Thoroughbred horses regardless of sex. All traits and index values of the proximal phalanx and middle phalanx can be used in breed distinction. The difference in phalangeal traits was significant. The proximal phalanx and middle phalanx values were higher in Thoroughbred horses. The distal phalanx values were not statistically significantly different between the breeds. The GSL and FA values were found to be higher in Arabian horses, unlike other length measurements.


**Table 1.** Mean values of measurements, standard deviations, and *p* values determined between sexes of Arabian horses and Thoroughbred horses (ANOVA).

> **Table 2.** Mean values of measurements, standard deviations, and *p* values determined for Arabian horses and Thoroughbred horses (ANOVA).


Results of the discriminant function analysis between Arabian and Thoroughbred horses are given in Table 3. Separate tests were performed for proximal phalanx, middle phalanx, and distal phalanx. Separate formulas were obtained for each bone. The highest canonical correlation value in the breed discrimination was obtained for the proximal phalanx. Multivariate discriminant function analysis score equations were as follow:

Proximal phalanx: (DBP × −16.584) + (DCP × −28.755) + (LMP × 6.686) + (Index 1 × 50.261) + 25.973.

Middle phalanx: (DBM × 0.931) + (DCM × 2.555) + (LMM × 2.187) + (Index 2 × −1.026) − 17.503.

Distal phalanx: (LD × −1.969) + (GSL × 1.706) + (FA × 0.454) + (Index 3 × 0.012) − 4.331.


**Table 3.** Stepwise discriminant function analysis for Arabian horses and thoroughbred horses.

V: variables, UC: unstandardized coefficient, SM: structure matrix, WL: Wilks' lambda, E: eigenvalue, GC: group centroids, CC: canonical correlation. A: Arabian horses T: thoroughbred horses.

For the proximal phalanx, the most distinctive feature in the discriminant function analysis was DCP (structure matrix: −0.801). In turn, LMM (structure matrix: 0.951) was the most distinguishing feature for the middle phalanx and FA (structure matrix: 0.666) for phalanx distalis. The eigenvalue value for the proximal phalanx was −0.909. It allowed correct classification of 89.33% of the samples used in the study (Table 4). The distinguishing features were the most tangible in the proximal phalanx; which was indicated by the highest percentage of the correctly classified bones. In the case of middle phalanx and distal phalanx, the correct classifications reached 77.33% and 54.67%, respectively. Traits of the distal phalanx were insignificant in the breed distinction. Wilks' lambda value was very high (0.958).


**Table 4.** Confusion matrix. Percentage of initial classifications that were correct, shown by breed.

Percentages within rows sum to 100%.

The distribution frequency of Arabian horses and Thoroughbred horses is shown in Figure 2. This figure demonstrates that the division into breeds based on the proximal phalanx data was more distinct. It can be seen that overlaps were more frequent in the distal phalanx. Almost none of the characteristics of the distal phalanx in Arabian horses differed from those of Thoroughbred horses.

**Figure 2.** Distribution frequency of Arabian horses and Thoroughbred horses after discriminant analysis.

Correlation values are presented in Table 5. There was usually a significant positive correlation between the individual measurement values. The length measurements of all three phalanges correlated strongly with the other parameters. The strongest correlation was found between the DCP and DBM.

**Table 5.** Coefficients of correlation between measurements of digital bones and between these measurements and horse age.


\*\* Correlation is significant at 0.01 level. \* Correlation is significant at 0.05 level.

#### **4. Discussion**

The phalanges of thoracic limbs of 75 horses were examined in the study. The LMP, LMM, and LD values were statistically different between males and females in Arabian horses, while no significant difference was found between males and females of Thoroughbred horses. Between the breeds, DCP was found to be more determinant for the proximal phalanx. In turn, LMM was the most distinguishing morphometric feature for the middle phalanx. It was observed that in Thoroughbred horses, most length measurement results were higher than those of Arabian horses (DBP, DCP, LMP, DBM, DCM, LMMLD, and FA). The reason for this is that Thoroughbred horses are generally taller than Arabian horses and this might be related to the length of individual sections of the limb skeleton [17]. The traits of the distal phalanx provide limited possibilities for breed differentiation.

Only radiogrametric measurements of digital bones of the left thoracic limbs of Arabian horse and Thoroughbred horse were used in this study. The difference between the right and left bones was ignored; only bone measurements were used to examine whether there was a sex and breed distinction. Previous studies have proved the lateralization of the skeleton of the manus. Alrtib et al. [15] compared right and left proximal phalanx measurements in their study using 10 Thoroughbred horses, five Standardbred horses and eight ponies. It was said that the medial length of proximal phalanx on the right side was larger than the left side. This difference was statistical. According to Kummer et al. [11], the left LMM is greater than the right and this difference is statistically significant. He reported that the left LMM before trimming was 4.6 ± 0.24 cm. In addition, he stated that the LMM

correlates positively with the height at the withers. In contrast, Linford et al. [12] did not observe significant differences between the left and right phalanges in any radiographic determination. In our study, LMM was 3.9 ± 0.27 cm in Arabian horses and 4.3 ± 0.26 cm in Thoroughbred horses.

Measurements and 3D-modeling using imaging systems has become an alternative to traditional anatomical morphometry. In the study of Dos Reis et al. [2], it was stated that the 3D models successfully reflect the anatomical characteristics of bones and that measurements taken using these models approximate the actual bone measurements. In their study using 3D printing, the length of the proximal phalanx was 8.42 cm and the length of the middle phalanx was 4.48 cm. No information was given about the breed of horses used in this research, but the length of phalanx media was close to the respective value found in this study for Thoroughbred horses.

The mean value of the length of the proximal phalanx was 9.28 cm and the length of the middle phalanx media was 3.9 cm in Arabian horses, whereas in the Thoroughbred horses the respective values reached 9.89 cm and 4.3 cm. We examined whether there was a difference between the two breeds in radiogrametric measurements taken of phalangeal bones. Results of the proximal phalanx were found to be more distinctive than those of the other digital bones in breed differentiation. Dzierz ˛ecka and Komosa [18] also revealed a statistically significant (*p* ≤ 0.01) difference in the greatest length of the proximal phalanx between warmblood horses and coldblood horses. They stated that the length of the proximal phalanx plays an important role in a certain morphological classification of animals.

In this study performed with Arabian horses and Thoroughbred horses, a correlation between the age of animal and osteometric features of phalanges was also evaluated. Analyses conducted demonstrated a positive correlation between the greatest solear length of the distal phalanx and age. Some studies can be found in literature addressing the effects of increasing age in horses on digital radiological measurements. Mullard et al. [19] proved that the ratio of the hoof distal phalanx distance to the length of the palmar aspect of the distal phalanx in horses decreases with increasing age, and that this correlation is statistically significant. In this study, the correlation of age with GSL was positive and only this value was statistically significant (R = 0.381). Several traits were negatively correlated with age, but these values were statistically insignificant. Anderson et al. [20] proved that there was no significant increase of length measurements in horses from age 2 to 3 years, suggesting that the growth rate either slowed or reached a plateau between ages 2 and 3 years. The lack of relationship of most analyzed parameters with age in our study is presumably due to the fact that most of the growth and development in the distal limbs of the horses ended at the age before the youngest of the measured horses. According to Sadek et al. [21], correlations between body measurements in Arabian horses vary a lot for both genders. Cruz et al. [22] examined the magnetic resonance images taken on these bones and stated that the exercise caused a positive change in the width parameters of the distal phalanx. In turn, Turek et al. [23] reported that different factors, including age, feeding, exercise and breed, might determine mechanical properties of the proximal phalanx in horses.

Osteometric measurements of phalanges were taken on equine bone remains obtained from excavations. Forsten [24] measured bones of horses in the Steinheim region. The length of the proximal phalanx found in this study was 9.13 ± 0.08 cm, and the length of the middle phalanx was 5.1 ± 0.14 cm. In another study, horses' remains were examined in Üzüür Gyalan Tomb, and measurements of proximal phalanx and middle phalanx were taken of these bone remains. The greatest length of the proximal phalanx from this excavation was 8.47 cm and the greatest length of the phalanx media was 5.00 cm [25]. In the present study conducted with Arabian horses and Thoroughbred horses, reference measurement values were obtained radiogrametrically from contemporary samples. These reference data may suggest that in terms of taxonomy, they can be helpful in interpreting osteoarchaeological studies. These data are particularly comparable with bone remnant

measurements obtained from excavations, and can provide information on animals of ancient periods. Although, when using these data, it should be kept in mind that there may be differences between measurements taken of bones and measurements taken via radiological imaging, as shown in the reference study [26].

The authors are aware that the lack of significance of differences between breeds and genders is partly due to the low abundance of research material. Another limitation is the lack of zoometric data, including the height at the withers, concerning the tested horses.

#### **5. Conclusions**

In this study, the statistical significance of radiogrametric measurements of the skeleton of the manus was examined in terms of sex and breed. Measurements of male individuals were generally greater. However, it can be concluded that the proximal phalanx and the middle phalanx manifest sexual dimorphism. The traits of the distal phalanx used in the study did not discriminate sex or breed. These data can serve as a reference material for radiogrametric studies in animals. The present study allowed obtaining reference radiogrametric measurements of the digits of the manus of Arabian horses and Thoroughbred horses.

**Author Contributions:** Conceptualization, O.G. and G.P.; methodology, D.O.E., O.G. and G.P.; software, O.G., S.D. and T.S.; investigation, D.O.E. and O.G.; validation, T.S., W.P.; writing—original draft preparation, S.D., T.S., W.P. and O.G.; writing—review and editing, T.S. and W.P. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** The study was approved by the Local Ethics Committee of the Faculty of Veterinary Medicine, ˙ Istanbul University-Cerrahpa¸sa (approval number: (2019/43)).

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** We would like to thank Turkey's Jockey Club and Hülya Hartoka for their support in conducting the study.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Article* **Maxillary Incisors of the Horse before and at the Beginning of the Teeth Shedding: Radiographic and CT Study**

#### **Francisco Miró 1,\*, Carla Manso 2, Andrés Diz <sup>1</sup> and Manuel Novales <sup>3</sup>**


Received: 24 August 2020; Accepted: 31 August 2020; Published: 10 September 2020

**Simple Summary:** Although much is known about equine dentistry, there is a period of the horse's life, prior to teeth shedding, in which there is lack of knowledge related to the development of deciduous incisors and dental germs of permanent incisors. To gain insight into the radiographic appearance of maxillary deciduous incisors and dental germs of maxillary permanent incisors during this period, a radiographic and computed tomography study of 25 horse skulls was made. Data regarding morphology and development were obtained. The results of the present study indicate that radiographic intraoral images are suitable to identify the grade of development of the dental germs of permanent incisors in horses before dental change. A detailed description of the radiographic appearance of deciduous incisors and dental germs of permanent incisors will help clinicians to expand their knowledge for diagnostic and treatment purposes.

**Abstract:** To gain insight into the radiographic appearance of maxillary deciduous incisors and dental germs of maxillary permanent incisors in the period prior to teeth shedding, radiographs and computed tomography (CT) of 25 horse skulls, with an estimated age of between 12 and 42 months, were studied. Data regarding morphology and development were obtained. Dental germs of first maxillary permanent incisors were identified radiographically as rounded radiolucent areas at the level of the apical parts of the first deciduous incisors, in skulls with an estimated age of twelve months. The first sign of crown mineralization of these dental germs appeared in skulls supposedly a few months older. Before teeth shedding, the unerupted, mineralized crowns of the first permanent incisor could be identified radiographically relatively caudal to the corresponding first deciduous incisors. The results of the present study indicate that radiographic intraoral images are suitable to identify the grade of development of the dental germs of maxillary permanent incisors. A detailed description of the radiographic appearance of deciduous incisors and dental germs of permanent incisors will help clinicians to expand their knowledge for diagnostic or treatment purposes.

**Keywords:** computed tomography; development; incisor; radiograph

#### **1. Introduction**

In horses, the grade of development of deciduous and permanent teeth, including incisors, has always been of great interest for veterinary clinicians to determine their age. Besides, knowledge of dental and periodontal regions is needed when dental and periodontal disorders, such as malocclusions, dental fractures, persistent deciduous teeth, supernumerary teeth, traumas, etc., are present in young animals [1,2]. Although visual examination of the mouth and radiography have always been the most-used methods by veterinary clinicians [3], recent diagnostic imaging procedures, such as computed tomography (CT), provide complementary and more precise information on dental examination in horses [4–9].

Some authors have reviewed and summarized the current knowledge of equine dental and periodontal anatomy [10–13] and the three-dimensional appearance of teeth, as well as their individual composition of dental hard structures [14]. Throughout the life of a horse, specific changes occur to the appearance of its teeth, and dental examination therefore provides the most convenient mean for age determination and diagnostic purposes. In this sense, there is information relating morphological characteristics of deciduous and permanent incisors observed by visual examination with the age of the horse [11,15–17]. Equine teeth, pertaining to the high crowned hypsodont teeth, are subjected to continuous dental wear [18], and the process of change from deciduous to permanent dentition involves a complex mechanism of development of the dental germs. Time of radiographic appearance and grade of development of dental germs are well-known for deciduous and permanent cheek teeth [19,20], but very few data are available for incisor teeth [21,22]. The latest studies focus on the disorders involving the incisive bone, maxillary incisors, and periodontal structures and show very few radiographic data of dental germs of the permanent incisors. These few data and the clinicians' experience indicate that there is a period of the horse's life in which the deciduous incisors and the dental germs of permanent incisors undergo different stages that could be identified radiographically. The referred period would extend from the complete eruption of deciduous incisors, some months before twelve months of age, to the eruption of the first permanent incisor, at the age of thirty months [11,15,23]. High quality intraoral dorsoventral radiographs are an excellent tool to study the maxillary incisors [1,23]. Moreover, modern imaging techniques such as computed tomography complement and overcome the limitations of two-dimensional radiographic images. To accurately diagnose and formulate a treatment plan, evaluation of dental disorders requires a complete maxillofacial–oral examination and supplemental imaging means [21]. Some authors have compared and validated the accuracy of CT and radiographic imaging in detecting cheek teeth disorders in horses [24]. As stated above, in horses, there is a lack of radiographic data on deciduous incisors and dental germs of permanent incisors in the period before and at the beginning of the teeth shedding. Hence, a detailed description of the radiographic appearance of the referred structures in the referred period will help clinicians to expand their knowledge for diagnostic or treatment purposes.

We hypothesized that, in young horses before shedding, the different grade of development of deciduous incisors and dental germs of permanent incisors can be identified radiographically. The purpose of the present study was to gain further insight into the radiographic appearance of deciduous incisors and dental germs of permanent incisors of horses in the period prior to and at the beginning of teeth shedding.

#### **2. Materials and Methods**

Twenty–five horse skulls were studied in the present study. Neither ponies nor draft horses were included in the study. Equine skulls, obtained from the Anatomy Department of the Veterinary School of Cordoba University, Spain, kept their complete dentition and surrounding bone without noticeable abnormalities.

#### *2.1. Visual Inspection*

Based on published aging guidelines for horses [11,15,23], the whole skulls were visually examined by one of the authors, a specialist on equine dentistry (C.M., board-certified specialist, European Veterinary Dental College). Eruption of deciduous and permanent maxillary and mandibular incisors, morphological characteristics of their erupted crowns, and the presence of some of the permanent cheek teeth were the main criteria for determining the estimated age. Incisors were named according to

the anatomical nomenclature, such as first, second, and third [25]. Correspondence with the modified Triadan system, worldwide used in veterinary dentistry [26], was made, appearing between en-dash after the corresponding anatomical name. Skulls with all deciduous incisors erupted and their occlusal surfaces in the same plane, assessed as between 12 and 30 months, were included in group 1. Skulls maintaining their second and third deciduous incisors—02 and 03; Quadrants 5,6,7, and 8—and with first permanent incisors—01; Quadrants 1,2,3, and 4—as well as second and maybe third permanent premolars—06 and 07; Quadrants 1,2,3, and 4—erupted, assessed as between 30 and 42 months of age, were included in group 2.

#### *2.2. Radiographic Study*

After visual inspection the maxillary incisor arcades were radiographed with an X-ray machine (Odel model C306-20®, Monza, Italy). Radiographs were processed by computerized radiology (Fuji Computed Radiography, Capsule XL, CR-IR 356® Tokyo, Japan). Radiographs were obtained according to the standard procedures of intraoral dorsoventral technique, being the X-ray beam directed 90◦ to the plane that bisects the angle between the 1st maxillary incisors and the imaging plate [1,3,4,22,27]. Radiographs, being the incisors crown down and the horse's left side presented to the viewer's right, were then assessed [3]. Maxillary deciduous incisors were identified and the radiological characteristics of their erupted and nonerupted portions and surrounding bone were assessed. Dental germs of the maxillary permanent incisors were identified, and their radiological characteristics and surrounding bone were analyzed. In skulls in which maxillary permanent incisors were identified, their radiological characteristics of the erupted and non-erupted portions and surrounding incisive bone were assessed. A first analysis of the intraoral radiographs of group 1 skulls allowed us to identify the dental germs of the 1st maxillary permanent incisors—101 and 201—according to different grades of development. Some of the skulls—classified from now on as *group 1a*—revealed dental germs like round quite radiolucent areas and apparently no other notorious characteristics. Several skulls—organized from now on within *group 1b*—showed circumscribed radiopaque images within the respective radiolucent zones of the dental germs. In radiographic studies of some other skulls—from now on as *group 1c*—unerupted 1st permanent incisors—101 and 201—were identified as being short, conic, wide, and with hardly any observable radiopacity. A group of skulls—from now on *group 1d*—showed larger unerupted 1st permanent incisors—101 and 201—with greater radiopacity. Table 1 shows the distribution of the skulls in the above referred groups and some of the most remarkable radiographic characteristics of 1st permanent maxillary incisors—101 and 201—used to classify them.


**Table 1.** In 25 skulls of Spanish horses, most remarkable radiographic characteristics of 1st permanent maxillary incisors—101, 201—and number of skulls studied. Groups 1a to 1d, estimated age of between 12 and less than 30 months. Group 2, estimated age between 30 and 42 months.

#### *2.3. Computed Tomography Imaging*

Subsequently, skulls were scanned with a helical CT scanner (CT Hi Speed CT/e Dual, General Electric Yokogawa Medical Systems LTD®, Hino, Japan) to obtain CT images from the incisors to the canines. Skulls were placed with the mandible on the CT scanning table. After image acquisition, CT slices from different planes of the incisors and adjacent structures were chosen, exported in DICOM format, and analyzed in Horos © software (open source for Apple, 64 bit, version 3.3.6). CT characteristics of deciduous incisors, dental germs of permanent incisors, permanent incisors (when they were present), and surrounding bone were analyzed.

#### *2.4. Length Measurements and Proportions*

Some length measurements and proportions were analyzed as a complementary study. Due to the small number of skulls per group, no statistical analysis was performed. Based on certain anatomical points (Figure 1), the following distances in mm were measured in sagittal scans at midlevel of the 1st deciduous incisors—501 and 601—by using Horos © software (open source for Apple 64 bit, version 3.3.6):

**Figure 1.** Sagittal computed tomography (CT) scan at midlevel of the 1st left maxillary deciduous incisor—601—of a skull of group 1d (supposedly the oldest among those with estimated age of between 12 and less than 30 months) showing the reference points for length measurements in the left maxillary 1st deciduous incisor—601— and in the unerupted 1st left permanent incisor—201. **1** and **6**, most occlusal points of the labial side of the teeth; **2** and **7**, most apical points of the labial enamel cover of the teeth; **3**, most apical point of the labial side of the 1st left maxillary deciduous incisor—60—; **4** and **6**, most occlusal points of the infundibular enamel, labial side; **5** and **8**, most apical points of the infundibular enamel, labial side.

Length of the tooth (LTOOTH): distance between the most occlusal point and the most apical point of the labial side of the tooth.

Length of the crown (LCR): distance between the most occlusal point of the labial side and the most apical point of the labial enamel cover (i.e., dental crown) of the tooth.

Length of the infundibulum (LINF): distance between the most occlusal point and the most apical point of the infundibular enamel, both on labial side.

Based on the above measurements, the following percentage measurements were also calculated: Relative length of the crown (%CR): percent of length of the crown with respect to the total length of the same tooth.

Relative length of the infundibulum (%INF): percent of length of the infundibulum with respect to the length of the same tooth.

When the unerupted 1st first maxillary permanent incisors—101 and 201—were identified, the same measurements were taken. Until they erupt, the 1st permanent incisors—101 and 201—were

considered as a whole tooth or crown. Once these teeth had erupted, the same measurements were taken on sagittal scans at midlevel of the tooth. Using the measurements for each group, the mean and standard deviations were calculated based upon the corresponding values of the right tooth and left tooth.

#### **3. Results**

#### *3.1. Visual Inspection*

Examination of the teeth of the skulls of group 1, classified as having an estimated dental age of between 12 and less than 30 months, revealed that all the deciduous incisors were clearly erupted and with wear on the occlusal surface of the 3rd incisors—03, Quadrants 5, 6, 7, and 8. They were of small size, with the oval occlusal surfaces in a mesiodistal orientation, and possessed shallow infundibula. Upon visual inspection, the most notable result available from the skull incisors of group 2, classified as having an estimated dental age of between 30 and less than 42 months, was the presence of the 1st permanent incisors—01, upon erupted Quadrants 1,2,3, and 4. In this group of skulls all other incisors were deciduous. In addition to the above, the skulls of group 2 had erupted the second permanent premolars or the second and third permanent premolars. In some of the skulls of group 2, there were the remains of the 1st deciduous incisors—01, Quadrants 5–7 and 8—in contact with the labial surface of the corresponding permanent tooth. Small oval openings, termed as gubernacular canals of the permanent incisors, were identified in the palatal region of the alveolar bone. They were present close to the 1st maxillary deciduous incisors—501 and 601—in groups 1a, 1b, and 1c (Figure 2A). In one of the skulls of group 1c the gubernacular canals were also present close to the 2nd deciduous incisors—502, 602. In group 1d the gubernacular canals were present close to the 1st and 2nd deciduous incisors—501, 502, 601, and 602—(Figure 2B). In group 2, there were gubernacular canals close to the 2nd and 3rd deciduous incisors—502, 503, 602, and 603—and in one of the skulls there was also a gubernacular canal close to the erupted 1st left permanent incisors—201.

**Figure 2.** Ventral view of maxillary dental arcade, incisive part, of skull of group 1a (**A**) and of group 1d (**B**). **1D**, right 1st deciduous incisor—501; **2D**, right 2nd deciduous incisor—502; **3D**, right 3rd deciduous incisor—503; **Gc1P**, gubernacular canal of right 1st permanent incisor—101; **Gc2P**, gubernacular canal of right 2nd permanent incisor—102; **Ic**, incisive canal; **INF**, infundibulum of the right 2nd deciduous incisor—502.

#### *3.2. Radiographic and CT Studies*

In radiographic studies of skulls of *group 1a* (Figure 3A), the 1st deciduous incisors—501 and 601—were identified with an elongate shape. These teeth stick out approximately one third from the incisive bones. This part was widened in a mesiodistal direction. Approximately two thirds of these teeth were embedded in the incisive bone. The embedded parts of these teeth comprised a central elongated and radiolucent area with two collateral radiopaque fringes. They may be identified, respectively, as the pulp cavity (the cavity of the tooth that contains the pulp) and the surrounding hard component of the dental root. The apical portion of these teeth end at the level of the incisive canal. Lateral to the image of the 1st deciduous incisors—501 and 601—overlapping images of the second and third deciduous incisors—502, 503, 602, and 603. In the apical parts of the 1st deciduous incisors—501 and 601—rounded radiolucent zones identify the dental germs of the 1st permanent incisors—101 and 102. These zones almost medially reached the interincisive suture, and their caudal limits were just at the level of the incisive canal. Sagittal CT images at midplane of 1st deciduous incisors—501 and 601—(Figure 3B) showed these teeth convexly curved on the labial face and concavely curved on the lingual face. They tapered evenly from their occlusal part to their apex. Analysis of the structural density allowed us to identify in them the anatomical crowns (parts with peripheral enamel) and dental roots (parts without superficial enamel). The enamel cover (i.e., crown) extended more apically on the labial side than in the palatal side. Most parts of the crowns had erupted, and the all dental roots were enclosed within the alveolus. Scans showed shallow infundibula filled at their bottom with hyperdense tissue. Transversal and sagittal images (Figure 3B,C) showed round hypodense areas touching the palatal side of the root of the corresponding 1st deciduous incisors—501 and 601. These areas corresponded with the dental germs of the 1st permanent incisors—101 and 201. The dorsal tip of the dental germs reached the end of the apical height of the root of the 1st deciduous incisors—501 and 601—and it is limited caudally and ventrally by the more rostral portion of the hard palate.

**Figure 3.** Intraoral radiographic image (**A**) and sagittal and transverse scans (**B**,**C**) of maxillary incisor arcade of a skull representative of group 1a. **1D**, left 1st deciduous incisor—601; **2D**, left 2nd deciduous incisor—602; **3D**, left 3rd deciduous incisor—603; **Pc**, pulp cavity; **dg**, dental germ of left 1st permanent incisor—201; **hd**, hard palate; **Ic**, incisive canal; **INF**, infundibulum of the left 1st deciduous incisor—601; **is**, interincisive suture.

Intraoral radiographs of skulls of *group 1b* (Figure 4A) did not show notable differences in images of the 1st deciduous incisors with that of the same teeth in group in 1a. However, it was notable that the radiolucent zones of dental germs of the 1st permanent incisors—101 and 201—were in this group smaller, more elongated, and caudally went beyond the incisive canal. The most remarkable aspect of the radiographic study in this group was the presence of circumscribed radiopaque images within the respective radiolucent zones of the dental germs (white arrow, Figure 4A). CT scans of skulls in this group (Figure 4B,C) showed, as in the group 1a, most of the crown of the 1st deciduous incisors—501 and 601—already erupted, and their dental roots enclosed within the alveolus. As it was in the previous group, the enamel cover extended more apically in the labial side than in the palatal side. There were characteristic hypodense and oval areas of dental germs in permanent incisors—101 and 201—touching the palatal sides of the roots of the corresponding deciduous teeth. Within the dental germs circumscribed hyperdense images could be clearly distinguished (white arrows, Figure 4B,C). Sagittal scans showed that the dorsal limit of the dental germs did not dorsally reach the more apical point of the root of the corresponding deciduous tooth.

**Figure 4.** Intraoral radiographic images (**A**) and sagittal and transverse scans (**B**,**C**) of the maxillary incisor arcade of a skull representative of group 1b. **1D**, left 1st deciduous incisor—601; **2D**, left 2nd deciduous incisor—602; **3D**, left 3rd deciduous incisor—603; **dg**, dental germ of left 1st permanent incisor—201; **Ic**, incisive canal; **white arrows**, circumscribed radiopaque (**A**) and hyperdense (**B**,**C**) images within the dental germs of left 1st permanent incisor—201.

In intraoral radiographs of skulls of *group 1c* (Figure 5A), 1st deciduous incisors—501 and 601—showed a trapezoidal occlusal surface area. Wide radiolucent zones of dental germs of 1st permanent incisors—101 and 201—were visible in the body of every incisive bone. Their caudal limits exceeded the level of the incisive canal. Fine radiopaque lines enclosed the radiolucent areas of the germs in some of the skulls. They were considered the lamina dura of the alveolus. Wide and conic images of the unerupted crowns of 1st permanent incisors—101 and 201—could be identified within the radiolucent areas of the germs with hardly any observable radiopacity. Sagittal and transverse CT scans of skulls of this group (Figure 5B,C) showed the crown of the 1st deciduous incisors—501 and 601—to be erupted with the dental root within the alveolus. Images of the dental germs of the 1st permanent incisors—101 and 201—were touching the palatal sides of the roots of their corresponding deciduous teeth. In sagittal scans the dental germs of 1st permanent incisors—101 and 201—showed oval hypodense areas containing parallel hyperdense structures, which correspond to the unerupted crowns during early mineralization. The surrounding limits of the germs clearly exceeded dorsocaudally the apical point of the root of the corresponding deciduous tooth. It is noticeable that the ventral limits of the germs were disrupted caudally by a few millimeters to the corresponding deciduous incisors (Figure 5, Gc). This structure was identified as the gubernacular canal of the 1st permanent incisors—101 and 201. As stated above, within the hypodense area of the germs, the short crowns of the 1st permanent incisor—101 and 201—were identified. In them, the enamel did not reach the apical point of the infundibulum.

**Figure 5.** Intraoral radiographic images (**A**) and sagittal and transverse scans (**B**,**C**) of maxillary incisor arcade of one skull representative of group 1c. **1D**, left 1st deciduous incisor—601; **2D**, left 2nd deciduous incisor—602; **3D**, left 3rd deciduous incisor—603; **1P**, left unerupted 1st permanent incisor—201; **dg**, dental germ of left 1st permanent incisor—201; **Gc**, gubernacular canal of left 1st permanent incisor—201; **Ic**, incisive canal; **Ld**, lamina dura of the alveolus of the right 1st permanent incisor—101.

In intraoral radiographs of skulls of *group 1d* (Figure 6A), those images of the 1st deciduous incisors—501 and 601—had notably shorter lengths than those in all previously reported groups. The area corresponding to the occlusal surface was trapezoid. Relatively caudal to the 1st deciduous incisors—501 and 601—the crowns of unerupted 1st permanent incisors—101 and 201—were identified clearly (Figure 6, 1P). They were larger and of greater radiopacity than in the radiographs of skulls of group 1c. They tapered apically, where they were surrounded by radiolucent areas, smaller than in previous groups. Most parts of these radiolucent areas were caudal to the incisive canal. As it was also reported in some of the skulls of group 1c, a thin radiopaque line, the lamina dura, enclosed the radiolucent area of the dental germs in two of the three skulls of this group. The lamina dura was not identified in the radiograph shown in Figure 6. It was noticeable that lateral to the crowns of the unerupted 1st permanent incisors—101 and 201—two radiolucent images can be easily identified. They corresponded with the dental germs of the 2nd permanent incisors—102 and 202—(Figure 6, dg2). Sagittal CT scans at midplane of 1st deciduous incisors—501 and 601—(Figure 6A) showed the roots of these teeth hardly enclosed by osseous tissue. Hence, a great part of the roots were not included within the alveolus. As shown in previous groups, the enamel cover of the 1st deciduous incisors—501 and 601—i.e., crowns, extended more apically on the labial side than in the palatal side. Caudal to the root of the 1st deciduous incisors—501 and 601—the dental germs of the 1st permanent incisors—101 and 201—were shown, with large unerupted crowns occupying most of the space (Figure 6, 1P and dg). The length of the unerupted crowns of 1st permanent incisors—101 and 201—were apparently larger than in any of the previously reported groups. The infundibulum (Figure 6B, INF) of the unerupted incisors were quite large and completely formed by calcified tissue, from its occlusal part to the apical point. In sagittal scans at midlevel of the 2nd deciduous incisors—502 and 602—(not shown in the Figure 6) and in transverse scans (Figure 6C) two small hypodense areas were identified (Figure 6C, dg2) laterally to the dental germs of the 1st permanent incisors—101 and 201—and caudally to the roots of the 2nd deciduous incisors—502 and 602. They corresponded with the dental germs of the 2nd permanent incisors—102 and 202.

**Figure 6.** Intraoral radiographic images (**A**) and sagittal and transverse scans (**B**,**C**) of maxillary incisor arcade of one skull representative of group 1d. **1D**, left 1st deciduous incisor—601; **2D**, left 2nd deciduous incisor—602; **3D**, left 3rd deciduous incisor—603; **1P**, unerupted left 1st permanent incisor—201; **dg**, dental germ of left 1st permanent incisor—201; **dg2**, dental germ of left 2nd permanent incisor—202; **Gc,** gubernacular canal of left 1st permanent incisor—201; **Ic**, incisive canal; **INF**, infundibulum of the unerupted left 1st permanent incisor—201.

In radiographs of skulls of *group 2* (Figure 7A) the 1st permanent incisors—101 and 201—appeared quite large. Most of the apical portion of these teeth was occupied by a wide radiolucent zone that ended rostrally in two horns. This zone corresponded with the pulp cavity (Figure 7, Pc). Collaterally they were limited by two radiopaque bands. Rostrally, a radiopaque triangular zone, between the horns of the pulp cavity, corresponded with the infundibulum, which in the more occlusal part of the tooth ended in an oval and slightly radiolucent image. In some of the radiographs of skulls of this group, remains of the 1st deciduous incisors—501 and 601—could be identified related with the labial side of the occlusal surface of the deciduous incisors. Laterally, with respect to the apical part of the 1st permanent incisors—101 and 201—shorter and conic crowns of the unerupted 2nd permanent incisors—102 and 202—were identified, in an oblique position. Surrounding the apex of the 1st and 2nd permanent incisors—101, 201, 102, and 202—there were semicircular radiopaque lines, which represented the lamina dura of the alveoli. In some skulls, small radiolucent areas could be identified at the level of the labial side of the occlusal surfaces of the unerupted 2nd permanent incisors—102 and 202. Sagittal scans at these levels confirmed that these images corresponded to the gubernacular canals of the 2nd permanent incisors—102 and 202. Sagittal CT scans at the level of the median plane of the 1st permanent incisors—101 and 201—(Figure 7B) showed most of these teeth enclosed by the corresponding alveoli on the dorsal and palatal sides, where a slight hyperdense line, the lamina dura, was clearly identified. A complete view of the internal morphology and structure of these teeth could be assessed. The hypodense pulp cavity occupied most of the apical part of the tooth, and it also extended within the occlusal part of the tooth, rostrally to the infundibulum. As stated for deciduous incisors, the enamel cover extended more apical in the labial side than in the palatal side of the permanent incisors. All the teeth, the surrounding soft tissues, and the lamina dura of the corresponding alveoli were clearly identified. The different grades of density allowed us to recognize of the structural components of the teeth. Hence, the 1st permanent incisors—101 and 201—of the skulls of this group showed the apical part of the infundibulum filled with cementum. The rostral

surface of the incisive bone, covering the labial side of the incisors, was quite thin and remains of the 1st deciduous incisors—501 and 601—might still be attached to a part of the skulls.

**Figure 7.** Intraoral radiographic images (**A**) and sagittal and transverse scans (**B**,**C**) of maxillary incisor arcade of one skull representative of group 2. **1D**, remain of left 1st deciduous incisor—601; **2D**, left 2nd deciduous incisor—602; **1P**, left 1st permanent incisor—201; **2P**, unerupted left 2nd permanent incisor—202; **Pc**, pulp cavity of the left 1st permanent incisor—201; **Gc2**, gubernacular canal of 2nd permanent incisor—202; **Ic**, incisive canal; **INF**, infundibulum of the left 1st permanent incisor—201; **Ld**, lamina dura of the alveolus of the left 1st permanent incisor—201. A small remain of right 1st deciduous incisor—501—is also present on the labial side of the corresponding permanent tooth.

#### *3.3. Length Measurements and Proportions*

Due to the lack of statistical analysis, the results of the present section must be taken with caution. Results of length measurements for 1st deciduous incisors—501 and 601—and 1st permanent incisors—101 and 201—in sagittal scans at midlevel of the corresponding 1st incisors are shown in Table 2. Since in group 1c and 1d the enamel seemed to cover the whole surface of the unerupted 1st permanent incisors—101 and 201—the length of the tooth and length of the crown of these teeth were considered to be the same in these two groups.

**Table 2.** Results (mean ± standard deviation, mm) of lengths of the tooth, crown, and infundibulum (LTOOTH, LCR, and LINF, respectively) of 1st deciduous—501 and 601—and permanent—101 and 201—incisors of horses with an estimated age of between 12 and less than 30 months (groups 1a–1d) and of between 30 and less than 42 months (group 2). A study of 25 horse skulls. Since in group 1c and 1d the enamel seemed to cover the whole surface of the unerupted 1st permanent incisor, length of the tooth and length of the crown of these teeth were considered to be the same in both of these two groups.


LTOOTH: Length of the tooth, LCR: Length of the crown, LINF: Length of the infundibulum.

The lengths of the crown and infundibulum of the 1st deciduous incisors—501 and 601—decreased from groups 1a and 1b to groups 1c and 1d. In the 1st permanent incisors—101 and 201—the lengths of the tooth, crown, and infundibulum increased from the unerupted teeth (groups 1c and 1d) to the erupted teeth (group 2). The lengths of the tooth, crown, and infundibulum of the 1st incisors were notably greater in permanent teeth—101 and 201—(i.e., group 2) than in deciduous teeth—501 and 601—(i.e., group 1).

Results of relative lengths of the crown and infundibulum as a percentage of the total length of the tooth for the 1st deciduous incisors—501 and 601—and 1st permanent incisors—101 and 201—are presented in Table 3. In groups 1c and 1d, the enamel seemed to cover the whole surface of the unerupted permanent teeth. Hence, the relative length of the crown (% CR) was considered 100% of the length of the tooth.

**Table 3.** Results (mean ± standard deviation) of relative lengths of the crown and infundibulum (% CR and % INF, respectively) of 1st deciduous incisors—501 and 601—and/or 1st permanent incisors—101 and 201—with respect to the total length of the tooth. Study of 25 horse skulls.


Relative length of the crown (%CR): percent of length of the crown with respect to the total length of the same tooth.Relative length of the infundibulum (%INF): percent of length of the infundibulum with respect to the length of the same tooth.

In the 1st deciduous incisors—501 and 601—the relative lengths of crown and infundibulum decreased from groups 1a and 1b to groups 1c and 1d. The relative lengths of the crown and infundibulum of the 1st incisors were notably greater in the permanent teeth (i.e., group 2) than in the deciduous teeth (i.e., groups 1).

#### **4. Discussion**

As stated by some authors [12,15,18] when determining a horse's age by its incisors, the eruption dates and changes in appearance of the occlusal surfaces represent the main criteria. The chronology and sequence of eruption of the permanent cheek teeth are of great assistance for this purpose. Based on the published aging guidelines [11,15,17] and the expertise of one of the authors in equine dentistry, in the present study, the skulls were firstly classified as having between 12 and less than 30 months of age (group 1) or between 30 and less than 42 months of age (group 2). To avoid inaccuracies in teeth eruption times resulting from differences between breeds and types of horses, neither ponies nor breeds that differed from the general dental aging system have been employed in the study. Radiography has been widely used in horses for diagnoses and treatment purposes [2,4,21,22], and it can also provide valuable information from fossil sites about the life history, the age at death of the individuals and the evolution of past species [20]. In the present study, intraoral radiographs of the maxillary incisor arcades were used for selection purposes. In our opinion, the horses' skulls in group 1 have a relatively wide range of ages and were divided into four different groups according to the development of their dental germs, as appeared in their respective radiographs. The complex and overlapping arrangement of the incisors within the maxillary arcade hampers the description of the three-dimensional positions of the individual teeth, especially while evaluating their 2D radiographs [4,14]. The CT scans obtained in the present study helped us better understand the internal structure of the erupted incisors, the dental germs of the unerupted incisors, and also the periodontal structures. They were also suitable to obtain

some length measurements and proportions, which complemented the information supplied by the visual inspection and the radiographic study. In young horses before shedding, we can use radiographs to determine the degree of development of their corresponding deciduous incisors as well as the dental germs of their permanent incisors before shedding. Without adequate sedation, live horses would not tolerate the placement of the imaging plate in the mouth without chewing [3]. Hence, the authors of the present study are aware of the added requirements of repeating the study in live animals.

#### *4.1. Deciduous Incisors: Radiographic and CT Studies*

The horses' incisor teeth are connected together to form a continuous arch in each arcade and are thus implanted so that their roots converge [28]. Although all six maxillary incisors could be analyzed by intraoral radiographs, the 1st deciduous and permanent incisors—501,601,101, and 201—could be studied better, because they were positioned in the middle of the images and with less superimposition. Equine incisors feature a single Y-shaped pulp cavity made up of two pulp horns, labially to the infundibulum, and a single oval cavity more apically [14,21]. The referred configuration is present in both arcades but appears more pronounced in the maxillary incisors [14]. By means of intraoral radiographs, the referred Y-shaped configuration could be identified in the 1st permanent incisors—101 and 201—of group 2, and sagittal scans of these teeth showed the pulp horns to be located labially to the infundibulum. In the radiographs of all the skulls of group 1, the pulp cavity appeared as a radiolucent area located in the embedded parts of the teeth. Sagittal scans determined the extension and position of the occlusal and apical parts of the pulp cavity within the tooth. Sagittal scans were also of great use in assessing other aspects of the internal morphology and structure of the incisors. It has been studied that, in the horse's incisors, the enamel cover, i.e., crown, extends more apically in the labial side than in the palatal side [14]. Sagittal CT scans obtained in the present study confirmed this non-uniform extension of the enamel cover in deciduous incisors. According to some authors, in horses, a large portion of the dental crown (the enamel-containing part of the tooth) of incisors lies within the dental alveolus [10,13,29]. Although not mentioned in the latest articles, it is supposed that the above statement was made in reference to permanent incisors. The same was found for the 1st permanent incisors—101 and 201—of the skulls of group 2 of the present study, since most of them were enclosed by their corresponding alveoli. It has been described that, after eruption, the large crowns of the horse's teeth are only gradually extruded, and the delayed development of the roots allows growth to continue for some years after the teeth have come into wear [10,13,28]. The unerupted crown of an equine's teeth functions as a root to counterbalance their abrasive diet and to support the strong stresses generated during mastication [29]. In the present study, it was observed that the majority of the crown of the 1st deciduous incisors—501 and 601—of skulls of groups 1a to 1c were erupted, contrary to what was observed for permanent incisors—101 and 201. In the groups 1a to 1c, the position of the crowns with respect to the alveoli was more similar to that of brachyodont teeth, such as primates or dogs, which have the entire crown exposed in the oral cavity [10,28,29]. In skulls of group 1d, supposedly the oldest of estimated ages between 12 and less than 30 months, a great part of the root of the deciduous incisors were not embedded within the alveolus, probably due to the impending dentition change.

#### *4.2. Permanent Incisors and Alveoli: Radiographic and CT Studies*

Intraoral radiographs and CT scans were used in the present study to analyze the permanent incisors, from the state of the dental germ to the erupted teeth. Odontogenesis in the horse follows the same principles and involves the same sequential processes as seen in other mammals [10]. The first period comprises the formation of the gingival epithelium of the fetal oral cavity of the so-called "dental lamina" and, later, its subsequent separation into cyst-like structures known as "tooth buds" [10,11]. Although both the deciduous and permanent tooth buds form within a short time of each other, the tooth buds for permanent teeth remain dormant until the mandible and maxilla obtain the sufficient length to provide space for their development [11]. In the second period, cellular differentiation and

interaction with the surrounding mesenchyma leads to the formation the tooth germs [10]. After that, a complicated process of growth and remodeling of the tooth germs, and production of the tooth substances leads to the fully developed teeth [10,13], deep into the roots of the temporary set of equivalent teeth [28]. After completing the development of the permanent incisors, the deciduous incisors are displaced labially by the erupting permanent incisors [10], which usually erupt on the lingual aspect of the corresponding deciduous teeth [18]. The latest process involves continuous adjustment and remodeling of the alveoli and surrounding bone [28]. Intraoral radiographs of skulls obtained in the present study were suitable to identify the dental germs of the 1st permanent incisors—101 and 201—and 2nd the permanent incisors—102 and 202—when they were present. CT scans confirmed the position and morphology of these dental germs. Some authors have shown in equines that tooth germs of some permanent molar teeth that do not have predecessor deciduous teeth (4th cheek teeth) appear radiographically at 6 to 7 weeks of age and early crown mineralization of them is detected at 7 weeks of age [19]. The same authors have showed radiographic evidence of tooth germs in premolar teeth that have predecessor deciduous teeth (1st to 3rd cheek teeth) at 10 to 12 months of age and early signs of mineralization at 11 to 13 months of age. The dental germs of the 1st permanent incisors—101 and 201—in the present study were identified radiographically in the skulls of the horses of groups 1a (supposedly the youngest of the study), and the first sign of crown mineralization appeared in group 1b, supposedly being a few months older than the previous ones. In skulls of the horses of groups 1c and 1d (supposedly older than horses of groups 1a and 1b, but also with an estimated age of between 12 and less than 30 months), the crowns of the 1st permanent incisors—101 and 201—were mineralized. CT scans of these last skulls (groups 1c and 1d) showed the crowns of the 1st permanent incisors—101 and 201—are clearly hyperdense due to their grade of calcification. Intraoral radiographs of a yearling (between one and two years old) showed the dental germs of the 1st permanent incisors—101 and 201—to have a similar grade of development and mineralization as the skulls of group 1c of the present study [22]. All the above may indicate that the dental germs of the 1st permanent incisors—101 and 201—may appear radiographically shortly before 12 months of age, and that the first sign of mineralization would appear some months later.

It has been described that at the periphery of the dental germ, cell differentiation, and calcification organize the individual constituents of the periodontium, i.e., the cementum, the periodontal ligament, and the alveolar bone [13]. Periodontal research has demonstrated that the periodontium, in addition to fixing the tooth within the alveolus, it also possesses a shock-absorbing and reparatory function and also contains multiple elements, such as mesenchymal cells, capable of transforming into any of the periodontal tissues when repair is needed [9]. Alveolar bone is very flexible and constantly remodels to accommodate the changing shape and size of the dental structures it contains [10,18]. The intraosseous stage of tooth eruption needs resorption of bone in the direction of eruption and formation of bone on the opposite side [10]. These activities depend upon the adjacent parts of the true dental germ [30]. Sagittal CT scans at midplane of the 1st deciduous incisors—501 and 601—of the skulls of group 1d (Figure 6A) showed a substantial part of the root was not included within the corresponding alveoli. On the contrary, CT scans of the 1st permanent incisors—101 and 201—of group 2 (Figure 7A) showed these teeth enclosed by the corresponding alveoli on the dorsal and palatal sides. The above may demonstrate in equines that bone resorption and bone formation are also polarized around erupting incisor teeth. The inner wall of the alveolar bone, a thin layer of compact bone, has been shown to appear radiographically as a delicate radiopaque line called the lamina dura [10,11,13,18]. It has been stated that the lamina dura is not detectable on computed tomography [11]. We disagree with this statement, because in the present study, the lamina dura of the alveoli of the 1st permanent incisors—101 and 201—was detected radiographically in some of the skulls of groups 1c and 1d and in all the skulls of group 2. Confirmation of the presence of this osseous part of the alveoli was confirmed by the hyperdensity shown in the structure in sagittal CT scans.

By visual inspection of skulls of the present study, small oval openings were detected in the palatal region of the alveolar bone, close to the deciduous incisors. They were identified as the openings of the so-called gubernacular canals. In humans, the gubernacular canals have been defined as small channels that connect the bony crypt of the permanent unerupted teeth to the oral surface of the alveoli [30,31]. In the skulls of horses in the present study, they seemed to appear gradually from medial to lateral from group 1a to group 2, related to the grade of development and calcification of the corresponding dental germs found in radiographs. In one of the skulls of group 2, there was also a gubernacular canal close to the corresponding already erupted 1st permanent incisors—101 and 201. The gubernacular canal contains the gubernacular cord, a structure composed of conjunctive tissue, which links the tooth follicle (part of the dental germ) to the overlying gingiva [31]. Both the gubernacular canal and the cord were supposed to help guiding and directing the course of tooth eruption [31]. However, the existence and functions of the gubernacular canal and the cord in humans are still controversial and questioned [31]. To the authors' knowledge there are no references in the literature describing the presence and the function of the gubernacular canal in horses. Therefore, studies on this issue are needed to deepen the knowledge of the role of the gubernacular canal and the cord in the eruption of the incisors in equines. Anyway, this structure is present in the change of dentition of equines, and it may appear on radiographs (and CT scans) of the incisor region, as was the case for group 2 (Figure 7, Gc2). Differential diagnosis using a possible pathological image must be employed in the corresponding cases.

#### *4.3. Deciduous and Permanent Incisors: Lengths and Proportions*

To the authors' knowledge, there is no available data concerning tooth length of equine deciduous incisors. Hence, the length measurements and proportions obtained in the present study may complement the morphological data studied by radiographs and CT scans. However, due to the lack of statistical analysis, these results should be considered with caution. It has been studied that formation of permanent equine teeth is not finished once the teeth erupt [13]. Abrasion and mastication wear down the functional crown at a rate of 2 to 3 mm per year, but the reserve crown erupts continuously so that approximately 2 cm of exposed crown is maintained [11]. In the 1st deciduous incisors—501 and 601—of the present study, the relative lengths of the crown and the infundibulum decreased from groups 1a and 1b to groups 1c and 1d (i.e., the relative length of the root increased in the same manner), coincident with an increase of the total length of the tooth. These results may suggest the leading role of root elongation to outbalance the grade of occlusal wearing in deciduous incisors, as it had already been suggested for permanent incisors from two to four years post eruption [14]. The infundibulum of the equine incisors is an enamel infolding inside the occlusal surface [15] that, in horses, has a depth of 10–30 mm [10]. In the present study, the length of the infundibulum ranged between 4.5 ± 1.4 mm of group 1d and 35.9 ± 3.2 mm of group 2. Our results agree with what has been stated by some authors who affirmed that deciduous incisors contain shallower infundibula than their permanent successors [10]. Once formed, the length of the infundibulum cannot be shortened, except by being worn from the occlusal side [14]. Hence, it seems logical that, in the present study, the minimum length of the infundibulum of the deciduous incisors was found to exist in the supposedly older animals (group 1d). It has been affirmed that during eruption, hypsodont teeth have no true roots, i.e., with an apical enamel-free area [10]. We suppose that the above statement referred to permanent incisors, since in the present study the length of the root in the erupted 1st permanent incisors—101 and 201—(group 2) represented less than 13 % of the tooth length (relative length of the crown 87.4 ± 5.6 mm). However, the length of the root in the 1st deciduous incisors—501 and 601—of the supposedly youngest animals (group 1a) represented more than 50% of the length of the tooth (relative length of the crown 48.5 ± 4.1 mm).

From this study, veterinary clinicians might improve their knowledge of the radiographic appearance of deciduous incisors and dental germs in the permanent incisors of horses in the period prior to teeth shedding. These radiographic data may be considered when injuries or disorders affect the maxillary incisor arcade. The main limitations of the present study were the lack of knowledge of the precise age of the horses to which the skulls used to belong and the small number of skulls in each group to perform a statistical analysis of the dental measurements and proportions. Further radiographic and CT investigations on a representative number of horses of known ages would provide a more comprehensive understanding of the development of dental germs of permanent incisors.

#### **5. Conclusions**

Radiographic intraoral images are suitable to identify the grade of development of the dental germs of the maxillary permanent incisors. In horses conforming to the general dental aging system, the dental germs of 1st permanent incisors—101 and 201—(identified radiographically as rounded radiolucent areas located at the level of the apical parts of the 1st deciduous incisors—501 and 601—appear at the age of twelve months. The first sign of crown mineralization of these dental germs appears a few months later. Before the teeth shedding, the unerupted mineralized crowns of the 1st permanent incisor can be identified radiographically as being relatively caudal to the corresponding 1st deciduous incisors.

**Author Contributions:** Conceptualization, F.M. and M.N.; methodology, F.M., C.M., A.D. and M.N.; formal analysis, F.M., C.M. and M.N.; writing—original draft preparation, F.M. and M.N.; writing—review and editing, F.M. and M.N.; supervision, F.M., C.M., A.D. and M.N. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*
