Contribution to understanding Avar burials with equids in Croatia: detailed archaeozoological analysis

AVAR BURIALS WITH EQUIDS IN CROATIA 73 Contribution to understanding Avar burials with equids in Croatia: detailed archaeozoological analysis T. TRBOJEVIĆ VUKIČEVIĆ 1 *, A. RAPAN PAPEŠA 2, I. ALIĆ 1, A. EKERT KABALIN 3, M. OSTOVIĆ 4, S. KUŽIR 1 1 Department of Anatomy, Histology and Embryology, Faculty of Veterinary Medicine, University of Zagreb, Heinzelova 55, 10 000 Zagreb, Croatia 2 Vinkovci Municipal Museum, Trg bana J. Šokčevića 16, 32 100 Vinkovci, Croatia 3 Department of Animal Husbandry, Faculty of Veterinary Medicine, University of Zagreb, Heinzelova 55, 10 000 Zagreb, Croatia 4 Department of Animal Hygiene, Behaviour and Welfare, Faculty of Veterinary Medicine, University of Zagreb, Heinzelova 55, 10 000 Zagreb, Croatia *Corresponding author: tajana@vef.hr SUMMARY The analysis was carried out on five equid skeletons from Avar Period graves from the following Croatian sites: grave 88 at Stari Jankovci (AD 650-750), grave 4 at Otok - Gradina (AD 750-840), grave 16 from the same site (8 th century), graves 4 and 5 at Khuen-Belasi Castle in Nuštar (end of 8 th beginning of 9 th century). Based on morphological characteristics of some bones, the skeleton in Stari Jankovci grave 88 was identified to be a possible mule (average withers height 129 cm, age 3.5-4 years), while skeletal remains from the other four graves indicated that they were from horse. Horses from Otok - Gradina grave 4 (age 6 years) and grave 16 (10-year-old mare) as well as Nuštar grave 5 horse (age 7 years) were of medium withers height (139 cm). The tallest was the horse from Nuštar grave 4 with an average height of 143 cm (age 5.5-6 years). We found evidence of pathological changes in the horse spine in Nuštar grave: thoracolumbar transitional vertebrae, spondylosis ankylopoetica on the first two lumbar vertebrae and spondylosis chronica deformans on four thoracic vertebrae and one lumbar vertebra. This study provides the first comprehensive analysis of Avar Period equid skeletons from sites in present-day Croatia and shows that they are similar to other Avar Period horse burials in the region. Keywords: Avars, burials, horse, mule, archaeozoology, pathology RÉSUMÉ Contribution à la compréhension de sépultures Avar contenant des équidés en Croatie: Analyse archéozoologique détaillée Cette étude a été conduite en Croatie sur cinq squelettes appartenant à des équidés de la periode Avar sur les sites suivants: le tombeau 88 a Stari Jankovci (AD 650-750), le tombeau 4 a Otok - Gradina (AD 750-840), le tombeau 16 sur le mem site (8eme siecle), les tombeaux 4 et 5 au chateau Khuen-Belasi a Nuštar (fin du 8eme siecle debut du 9eme siecle). Sur la base des caractéristiques morphologiques de certains ossements, le squelette de Stari Jankovci, tombeau 88, pourrait être celui d une mule (hauteur du garrot 129 cm, âge 3.5-4 ans), tandis que les autres squelettes sur les quatre autres tombeaux indiquent qu ils appartiennent à des chevaux. Le plus petit cheval était celui provenant du tombeau 4 d Otok-Gradina (taille 138 cm, âge 6 ans). Le cheval provenant du tombeau 16 d Otok-Gradina était une jument de 10 ans de 139 cm. Le cheval (âge 7ans) découvert sur le tombeau 5 de Nustar avait la même taille de 139 cm tandis que le plus haut était le cheval du tombeau 4 de Nustar avec une taille de 142 cm (âge 5ans et demi- 6 ans). Nous avons trouvé des indices de changement pathologiques sur la colonne vertébrale du cheval du tombeau de Nustar: thoracolumbar sur la vertèbre transitionnelle, spondylosis ankylopoetica sur les deux premières vertèbres lombaires et spondylosis chronica deformans sur quatre vertèbres thoraciques et une vertèbre lombaire. Cette étude fournit la première analyse complète de squelettes de la Période Avar sur des squelettes d équidés sur des sites en Croatie et montre qu ils sont similaires à d autres sépultures de chevaux Période Avar dans la région. Mots-cles: Avar, sépulture, cheval, mule, archaéozoologie, pathologie Introduction The Avar Period (ca. AD 570 to ca. 800), tremendously important in the early medieval history of Europe, takes its name from the Avars, an ethnic group that migrated from the Eurasian steppes into the Middle Danube region in the second half of the 6th century [6], causing a turning point in the history of the Carpathian Basin [41]. The Avar qaganate was the creation of an elite group of nomadic horsemen of eastern origin [36]. The periphery of the Avar quaganate lay along the lowland forest-steppe along the Drava and Danube rivers (Southern Pannonia) in what is now part of Croatia. There the Avars lived from the end of the 6 th century to the beginning of the 9 th [10]. The fundamental importance of horses to Asian nomadic cultures like the Avars is reflected in horse burials, a custom that appeared first during the Iron Age in the Carpathian Basin and reemerges in the early Avar period (late 6 th -early 7 th century AD) [6]. The Avars and early Hungarians, as well as the most of their nomadic ancestors on the steppes buried the dead along with their weapons, jewelry and horses [33]. The tradition of sending the horse to death together with the fallen rider or warrior is associated with the Teutonic belief of an afterworld in which the horse continued to serve its rider [43]. This type of burial was relatively common during the Late Avar Period (ca AD 710-810), especially along the quaganate borders, where it was practised usually for fallen soldiers [11].

74 TRBOJEVIC VUKICEVIC (T.) AND COLLABORATORS The analises of the remains of horses from the Avar period were carried out in an Avar-Slavic cemetery in Nové Zámky, southern Slovakia [27]; Late Avar cemeteries in Slovakia, Southern Transdanubia, Hungarian Great Plain and Budakalász-Dunapart [1, 2]; a cemetery at the Drantumer Mühle in the Cloppenburg district in Germany [43]; a 10 th -century cemetery in Southeast Hungary [21]; and an 8th-century horse recovered from the Ušće necropolis near Obrenovac in Serbia [24]. The Avar contribution to horse burials occurred hand-in-hand with their innovations of new horseriding equipment, such as the wood-framed saddle with leather cover, stirrups and a powerful bow. These inventions were soon accepted in Byzantium and other neighbouring states [46]. What remains unclear is how horse burials in the Avar communities occupying presentday Croatia relate to horse burials described in other parts of central Europe. This is an important question because recent archaeological literature mentions horse skeletons superficially, without providing detailed morphometrical or other archaeozoological analysis. Of the approximately 600 Avar burials catalogued in Croatia, nine are horse burials. Most research on these burials has focused on the horseriding equipment, leaving the equine skeletons largely neglected. For the current study, five skeleton remains were available from three archaeological sites: 1. Avar-Slavic necropolis in the village of Stari Jankovci - Gatina in southeastern Slavonia dated from AD 650-750 [39]; 2. a larger Avar-Slavic cemetery from AD 750 840 situated on a prehistoric settlement on Otok - Gradina [40]; 3. cemetery near Khuen-Belasi Castle in the village of Nuštar, dated to the 8 th and early 9 th centuries in the Late Avar Period [34]. These studies of Avar horse burials in Croatia have raised far more questions than they have answered. One of these questions is the correlation of pathological changes in the horse bones with riding. Another unresolved question is whether mules are present at any burial sites, given that a poem from AD 796, cited in the Poetae latini aevi Carolini [9], describes how the Avar Khan mounted a she-mule to visit King Pippin of the Frankish Empire [4]. Data from Croatian sites may also help address the question of whether horseback riding by the Avars caused spine pathology, which has been suggested by numerous archaeozoological studies (e.g. [3, 16, 17, 23, 24, 28]. Material and methods Figure 1 show the Croatian sites from which five equid skeletons from Avar Period graves were analysed: grave 88 at Stari Jankovci (AD 650-750), grave 4 at Otok - Gradina (AD 750-840), grave 16 from the same site (8 th century), graves 4 and 5 at Khuen-Belasi Castle in Nuštar (end of 8 th beginning of 9 th century). Comparative morphological analysis of post-cranial equid bones with those of mules and donkeys was based on PETERS [30] and JOHNSTONE [18] including the following morphological characteristics: Figure 1: Map showing the locations of eguids skeletal findings Nuštar Stari Jankovci Otok Scapula: torsion of the collum scapulae and pronounced strengthening of the caudal border; Radius: sulcus on crista transversa, concavity/convexity on the palmar side above the distal articulation, rough area on the medial and palmar side of the shaft distal to the ulna; Metacarpal bone: depression on the distal palmar area of the diaphysis; Tibia: medio-plantar expansion of the medial half of the distal epiphysis (trapezium or rectangular shape of the distal articular surface); Phalanx proximalis: prominent muscle insertion triangle and its apex on the posterior surface, position of apex of that prominent triangle. The morphology of the premolars and molars and identification of the enamel fold pattern was analyzed by DAVIS et al. [8] and JOHNSTONE [18]. Animal age was estimated based on epiphysis and diaphysis fusion of long bones and vertebrae [20, 35], as well as incisor sequence wear [15, 22]. Animal sex was determined based on the presence of canines. Osteometrics was performed according to Von den DRIESCH [44] and included all complete long bones. The following measures were taken: GL = greatest length; GLl = greatest lateral lenght, Ll = lateral length ; Bp = greatest breadth of proximal end; SD = smallest breadth of diaphysis; and Bd = greatest breadth of distal end. Measurements were taken using a digital caliper. Withers height was calculated based on measurements of long bones using two methods. One method used corrected factors [25] to multiply lateral lengths [19] and the greatest lengths of long bones; those two resulting estimates were averaged to give the final withers height result. The second method used GL values and the corresponding withers height range by VITT [42].

AVAR BURIALS WITH EQUIDS IN CROATIA 75 The shape indices were calculated for the metapodial bones as a proportion of the greatest length using the formula: (SD/GL) x 100. Pathological changes was scored from 1 to 4 by BARTOSIEWICZ and BARTOSIEWICZ [3] system and from 1 to 4 by LEVINE et al. [23] system. Results The analysis included almost complete skeletons of five horses from five graves at three sites. Table I shows the number of bones of each skeleton and estimated age and sex. No skull were found at the Stari Jankovci grave, so sex could not be estimated, and age was estimated only on epiphyseal closure of the long bones and vertebre. The skeleton in Stari Jankovci grave 88 was identified to be a possible mule based on morphological characteristics of the scapula (pronounced strengthening of the caudal border), radius (deep sulcus on crista transversa and visible depression in the distal palmar area of the shaft), metacarpus (visible and palpable depression on the distal palmar area of the shaft), tibia (medio-plantar expansion of the medial half of the distal epiphysis, so viewed from the distal end the shape of articular surface is more trapezoidal shape) and all three proximal phalanges (expressed muscle insertion triangle and its apex on the posterior surface, position of apex of that prominent triangle located at a height of two-thirds of the shaft). Morphological analysis of postcranial skeletal remains, as well as identification of the premolars and molars enamel fold pattern (well developed Pli cabaline and elongated, asymmetric protocone at maxillary teeth; U-shaped and deep lingual fold, pointed metastylid and rounded metaconid at mandibular teeth) from the other four graves indicated that they were from horse. Pathological changes were found only on the skeletal remains of the horse from Nuštar grave 5 (table IV). Fragments for only 13 of the usual 18 thoracic vertebrae were identified, preventing the assignment of their specific position on the spine. At the four thoracic vertebrae (probable T11-T14 position), new bone spurs were evident on the ventral surfaces of the vertebral bodies adjacent to the intervertebral spaces (fig. 2), suggesting syndesmophyte formation. In addition, the first two lumbar vertebrae had grown together via the spinal processus, and the processus articularis caudalis had coalesced with the processus articularis cranialis with periarticular osteophytes. Closer inspection of these two vertebrae from the left side suggested that the more cranial bone was a thoracic vertebra with an articulated area on a small transversal processus (fovea costalis processus transversi). Inspection from the right side, suggested that this same vertebra resembled a usual lumbar vertebra (L1), with a complete, normally developed transversal processus (fig. 3). Figure 2: Ventral view of thoracic vertebrae (probable range T11-T14) from Nuštar grave 5 horse, showing new bone spurs on the ventral surfaces of the vertebral bodies (scale is 5 cm). Measurments of long bones and metapodial shape indices are shown in table II. The metacarpal shape indexes was slender (13.97 14.5) for both Nuštar horses, slightly slender (14.77) for Otok grave 16 horse and medium slender (16) for the Stari Jankovci equid. Mean for the metatarsal shape index was 11.5. Withers height was calculated based on two methods and shown in table III Stari Jankovci equid was smaller than medium (129 cm), horses from both Otok graves, as well as Nuštar grave 5 horse were medium (139 cm), while Nuštar grave 4 horse was larger than medium (143 cm). Figure 3: Horse first two lumbar vertebrae from Nuštar grave 5, showing fused spinal and articular processus. A) left side, showing cranially thoracic vertebra and caudally lumbar vertebra; B) right side, showing both lumbar vertebrae (scale is 5 cm). Site and grave Stari Jankovci grave 88 Otok grave 4 Otok grave 16 Nuštar grave 4 Nuštar grave 5 Age 3,5-4 years 6 years 10 years 5,5-6 years 7 years Sex? male female male male Number of skeletal element 80 16 76 132 150 Table I: Age, sex and number of skeletal elements of equids.

76 TRBOJEVIC VUKICEVIC (T.) AND COLLABORATORS Measure/Index N Mean±SD Min-Max Humerus G1 9 287.5 ±10.6 266.9-296.4 Humerus GL1 10 282.1±11.8 262.3-292.8 Humerus Bp 9 88.8±2.4 86.3-92.4 Humerus SD 10 34.9620 33.2-36.9 Humerus Bd 10 73.3±4.5 67.8-81.1 Radius GL 6 337.3±11.4 318.6-352.1 Radius L1 6 322.9±10.5 310.9-338.6 Radius Bp 6 77.7±2.3 73.2-79.3 Radius SD 7 37.1±1.5 34.4-38.9 Radius BD 6 71.5±2.7 68.8-74.7 MTC GL 5 221.9±9.4 206.0-228.1 MTC L1 5 215.5±10.3 198.4-224.5 MTC Bp 5 48.5±1.9 45.3-50.4 MTC SD 5 32.5±0.6 31.9-33.1 MTCBd 5 46.7±2.8 43.7-50.5 MTC Bp / GLx100 5 21.88±0.4 21.4-22.4 MTC SD / GLx100 5 14.7±0.8 14.0-16.0 MTC Bd / GLx100 5 21.1±0.8 20.2-22.1 Femur GL 4 380.3±13.9 365.6-397.6 Femur Bp 3 111.7±5.4 107.9-117.9 Femur SD 4 38.7±1.6 37.0-40.5 Femur Bd 5 85.7±4.1 81.3-92.4 Tibia GL 9 355.9±8.7 341.1-369.3 Tibia L1 9 330.6±6.2 321.0-338.8 Tibia Bp 8 90.4±2.4 87.9-95.5 Tibia SD 9 38.0±1.8 35.3-40.3 Tibia BD 9 66.2±1.4 63.9-68.4 MTT GL 8 265.8±13.8 244.3-278.4 MTT L1 8 258.1±13.2 238.0-273.1 MTT Bp 8 46.7±1.7 45.0-49.7 MTT SD 8 30.5±1.3 28.1-32.3 MTT Bd 8 46.4±1.7 44.3-48.7 MTT Bp / GLx100 8 17.6±0.9 16.5-19.0 MTT SD / GLx100 8 11.5±0.7 10.3-12.6 MTT Bd / GLx100 8 17.5±0.5 16.9-18.3 Table II: Equids forelimb and hindlimb bone measurements (N= number of measurements; SD= standard deviation; MTC Metacarpus, MTT Metatarsus. Measurements in mm. Site Lateral length Greatest length N Mean ± SD N Mean ± SD Average Vitt (1952) Stari Jankovci, grave 88 8 130.1±52.768 8 128.2±33.5 129 Smaller than medium (128-136) Otok - Gradina, grave 4 4 140.7±63.0 3 136.6±31.1 139 Medium (136-144) Otok - Gradina, grave 16 10 139.8±31.5 10 138.439±24.0 139 Medium (136-144) Nuštar, grave 4 9 143.88±38.9 9 141.7±45.5 143 Larger than medium (144-152) Nuštar, grave 5 11 139.5±31.5 11 138.6±39.8 139 Medium (136-144) Table III: Withers height in cm (N= number of measurements; SD= standard deviation). Measurements in mm.

AVAR BURIALS WITH EQUIDS IN CROATIA 77 Fragments for only four of the usual six lumbar vertebrae were identified, preventing the assignment of their specific position. At the probable fourth or fifth lumbar vertebra, osteophytes had grown beyond the cranial edge of the extremitas cranialis, and a large bone spur was visible laterally on the left side of the vertebral body (fig. 4). None of the bone fragments showed butchery marks. Figure 4: Horse lumbar vertebra (probable L4 or L5) from Nuštar grave 5. A) cranial view showing osteophytes grown beyond the cranial edge of the extremitas cranialis; B) left side showing large bone spur on the vertebral body (scale is 5 cm). Discussion This study provides, to our knowledge, the first comprehensive analysis of Avar Period equid skeletons from sites in present-day Croatia. BARTOSIEWICZ and GYÖNGYÖSSY [4] commented that the literature on hundreds of complete equid skeletons from Migration and Avar Period burials in Central Europe never considered the possibility that some might have been mules. Based on Poetae latini aevi Carolini [9], those authors proposed that the Avar Khan travelled to the Frankish Empire on a she-mule. Morphological analysis of post-cranial bones from our putative Stari Jankovci mule is consistent with the literature on mule bones [8, 18, 30]. Problem with the morphological identification is the absence of maxilla and mandibula of the Stari Jankovci equid, because the identification of the enamel fold pattern could not be perfomed and this is often much more useful for discrimination of horse and mule than the postcranial elements [18]. This animal was likely a medium-slender mule, based on data on metacarpal slenderness from Brauner (1916) published in BARTOSIEWICZ and BARTOSIEWICZ [3]. The withers height of the putative mule was 129 cm, smaller than that of the horses in our sample, and by VITT [42] smaller than the medium (128-136 cm). That mules would be buried together with their riders in the same way as horses is not surprising given that mules were the favoured pack animal for long-distance transport, combat and agriculture for at least 3000 years, until the invention of the steam engine [7]. Bone analysis of the human skeleton buried with our putative mule suggests that it belonged to a male aged 40-50 showing severe degenerative changes in his cervical vertebrae likely due to extensive occupational riding [32]. Since these degenerative changes probably caused pain, we speculate that this older man chose to ride a mule instead of a horse for greater comfort. As BARTOSIEWICZ and GYÖNGYÖSSY [4] concluded the gaited mule can be ridden farther without discomfort from the bouncing during the trot and at more manageable pace then a canter, and than mentioned Avar Khan must have been an elderly gentelman at the time of his travell. The age and withers height of the four horses at the Croatian graves is consistent with analyses of Avar horse remains discovered in other parts of Central Europe. The horse skeletons from the Budakalász-Dunapart Cemetery of the Late Avar Period, from the Hungarian Great Plain, from the Slovakia and from the Southern Transdanubia cemeteries were fully grown individuals and the mean estimated withers height was 136.4±0.829 cm for Budakalász horses and 136.4±0.292 cm for the horses from the other three areas [2]. The difference between the actual withers height and the estimated value has been estimated to be as much as 100 mm, with an average of 40 to 50 mm either side of the estimated value [25, 45]. FOREST [11] also states that the height of the shoulder always varies from 5 to 10% in a particular animal, as well as in the breeds. It should therefore be remembered that all reconstructed withers heights can only be estimates [18] and that is actually the approximate value due to several factors: multiple imprecision that makes the incorrect calculation, incorrect application of certain definitions, random selection coefficients, natural differences due to biological variability laws and mixture of the obtained data [11]. Based on Brauner s (1916) data [3], the horse from Otok - Gradina grave 16 had slightly slender metacarpals, while both Nuštar horses had medium slender metacarpals. Most Hungarian Avar Period horse metacarpals fall into these Pathological changes score probable T11-T14 Vertebrae position L1, L2 by BARTOSIEWICZ and BARTOSIEWICZ [3] 2 1 2 by LEVINE et al. [22] 1 4 1 probable L4 or L5 Table IV: Pathological changes at vertebras of the horse from Nuštar grave 5

78 TRBOJEVIC VUKICEVIC (T.) AND COLLABORATORS two categories [5]. All those data classified investigated Avar horses to Eastern Group, horses which have been reported to have an average withers height 135 cm (121-149 cm), metacarpal index of 15,24 and metatarsal index of 11,59 [5]. Compared with other periods of Central Europe, using periodic comparision of Yenikapi Byzantine Horse [29] (p. 41. figure 2), Croatian Avar horses had withers height higher than the horses of the Bronze and Iron ages, as well as Migration period, but somewhat lower than Roman and Yenikapi Byzantine Horse. We found evidence of pathological changes in the spine in Nuštar grave 5, though locating the damage precisely was impossible because we did not locate all the vertebrae. Although horses are commonly assumed to have five lumbar vertebrae, they may have six [13] or even seven [20]. The first lumbar vertebra, which has the characteristics of a thoracic vertebra on the left side but the characteristics of a lumbar vertebra on the right side, shows congenital distortion of transitional segments of the spine [37], which are similar in all species, but are described primarily in dogs. Transitional vertebrae are located between two adjacent vertebral regions and have the morphological characteristics of both regions, i.e. they are a hybrid vertebrae [14]. HAUSSLER et al. [13] showed that among thoroughbred racehorses, 22% of specimens had thoracolumbar transitional vertebrae We speculate that the pathological fusion of the spinal processus on the first two lumbar vertebrae of the Nuštar grave 5 horse may correspond to spondylosis ankylopoetica described by MORGAN [26], while the fusion of osteophytes to the ventral side of four thoracic vertebrae and one lumbar vertebra on the same horse may correspond to spondylosis chronica deformans. This conclusion should be considered preliminary because these two conditions cannot always be distinguished based purely on archaeozoological data in the absence of clinical diagnostic data [3], but some authors have suggested the most likely cause of spondylosis to be loadbearing as a result of riding [31]. The pathological changes in our horse could be also consistent with chronic riding and load-bearing. Evidence suggests that long-term saddle riding can cause chronic inflamation of soft tissue, which leads to an osseus tissue reaction and ultimately to such abnormalities as ankylosis of the zygapophyses and fusion of the vertebral bodies and arches [17]. As in the present study, MARKOVIĆ et al. [24] observed spondylotic changes on the spine of Late Avar horses, which they attributed to chronic inflammation of ligaments and vertebral discs caused by intensive riding over a long period of time. ONAR et al. [28] reported similar findings on the spine and limb bones of Yenikapi Byzantine horse skeletons, which those authors likewise attributed to riding and loading. In our case, the horse from the grave 5 from Nuštar could serve for riding, especially because the Avars had known cavalry which are usually used for the decisive part of the battle [46]. Analysis of the rider remains buried together with the horses at the Croatian sites showed that in Nuštar graves, femora from the horsemen at Stari Jankovci and Otok - Gradina grave 16 showed skeletal markers of occupational activity associated with riding, while the horseman from Stari Jankovci also showed severe degenerative changes in cervical vertebrae [30]. Based on this comparative analysis, we agree with BARTOSIEWICZ and BARTOSIEWICZ [3] that while riding stress may strongly contribute to fusion of vertebrae at the transition between the thoracic and lumbar regions in horses, it is not the only cause. Like those authors, we can conclude that old age, strain injury, genetic factors and additional exogenous factors may also contribute. In conclusion, analysis of Avar horse burials from the area of present-day Croatia shows that they are similar to other Avar Period horse burials in the region. In Croatia, as elsewhere, only a small number of Avars were buried with their horses, suggesting a hierarchical society in which certain individuals, including warriors, enjoyed high status [6]. Acknowledgements This paper is dedicated to the memory of Marija Šmalcelj, who brought the first Avar horse skeleton to the Department of Anatomy, Histology and Embryology, Faculty of Veterinary medicine University of Zagreb and introduced us to the world of Avars. We are also grateful to Dr Petra Rajić Šikanjić and Zrinka Premužić from Institute for Anthropological Research, Zagreb (Croatia) for provided access to the rest of the horse skeletons and helped us with anthropological data. References 1. - BARTOSIEWICZ L.: Avarkori lovak végtagarányai, Móra Ferenc Múzeum Évkönyve, Szeged, 1991, 301-310. 2. - BARTOSIEWICZ L.: Animal remains from the Avar period cemetery of Budakalász Dunapart. Acta Archaeologica Academiae Scientarum Hungaricae, 1995, 47, 241-255. 3. - BARTOSIEWICZ L., BARTOSIEWICZ G.: Bamboo spine in a Migration Period horse from Hungary. J. Archaeol. Sci., 2002, 29, 819 830. DOI: 10.1006/ jasc.2001.0715. 4. - BARTOSIEWICZ L., GYÖNGYÖSSY M.: The Khan s Mule: Attitudes toward a Forgotten Animal. In: S.L. OLSEN, S. GRANT, A. CHOYKE, L. BARTOSIEWICZ (Eds.): Horses and humans: the evolution of humanequine relationships, BAR International Series, vol. 1560, Archaeopress, Oxford, 2006, 289-302. 5. - BÖKÖNYI S.: History of Domestic mammals in Central and Eastern Europe. Akadémiai Kiadó, Budapest, 1974. 6. - BEDE I.: The status of horses in late Avar-period society in the Carpathian Basin. In: R. ANNAERT, K. De GROOTE, Y. HOLLEVOET, F. THEUWS, D. TYS, L. VERSLYPE (Eds): ACE Conference Brussels: The very beginning of Europe? Cultural and Social Dimensions of Early-Medieval Migration and Colonisation (5th-8th century), Flanders Heritage Agency, Brussels, 2012, 41-50. 7. - CLUTTON-BROCK J.: Horse Power. Natural History Museum Publications, London, 1992.

AVAR BURIALS WITH EQUIDS IN CROATIA 79 8. - DAVIS S. J. M., GONÇALVES M. J., GABRIEL S.: Animal remains from a Moslem period (12th/13th century AD) lixeira (garbage dump) in Silves, Algarve, Portugal. Revista Portuguesa de Arqueologia, 2008, 11, 183 258. 9. - DÜMMLER E.: Poetae latini aevi Carolini Vol. 1. Monumenta germaniae historica. Poetae Latini medii aevi 1. Wiedmann, Berlin, 1881. 10. - FILIPEC K.: Problem kronologije grobalja 9. i 10. stoljeća u sjevernoj Hrvatskoj. Starohrvatska prosvjeta, 2009, 36, 113-124. 11. - FOREST V.: De la hauteur au garrot des espèces domestiques en archéozoologie. Revue Med. Vet., 1998, 149, 55-60. 12. - GARAM É.: Das awarenzeitliche Gräberfeld von Tiszafüred. In: A. KISS and É. GARAM (Eds.): Cemeteries of the Avar Period (567-829) in Hungary, Akadémiai Kiadö, Budapest, 1995. 13. - HAUSSLER K. K., STOVER S. M., WILLITS N. H.: Developmental variation in lumbosacropelvic anatomy of Thoroughed racehorses. Am. J. Vet. Res., 1997, 58, 1083 1091. 14. - HENSON F. M. D.: Equine Back Pathology: Diagnosis and Treatment. Wiley-Blackwell, Oxford, 2009. 15. - HILLSON S.: Teeth. Cambridge University Press, Cambridge. 1986 16. - JANECZEK M., CHRÓSZCZ A., MIKLÍKOVÁ Z., FABIŠ M.: The pathological changes in the hindlimb of a horse from the Roman Period. Vet. Med-Czech., 2010, 55, 331 335. 17. - JANECZEK M., CHRÓSZCZ A., ONAR V., HENKLEWSKI R., PIEKALSKI J., DUMA P., CZERSKI A., CAŁKOSIŃSKI I.: Anatomical and biomechanical aspects of the horse spine: the interpretation of vertebral fusion in a medieval horse from Wroclaw (Poland). Int. J. Osteoarchaeol., 2014, 24, 623-633. doi/10.1002/ oa.2248 18. - JOHNSTONE C. J.: A Biometrical Study of Equids in the Roman World (PhD dissertation), University of York, York, 2004. 19. - KIESEWALTER L.: Skelettmessungen am Pferde als Beitrage zur theoretische Grundlage der Beurteilungslehre des Pferdes (Inaugural-Dissertation), Philosophischen Facultät der Universität Leipzig, Leipzig, 1888. 20. - KÖNIG H. E., LIEBICH H. G.: Veterinary Anatomy of Domestic Mammals. Textbook and Color Atlas, Schattauer, Stuttgart, 2009. 21. - LANGÓ P., RÉTI Z., TÜRK A. A.: Reconstruction and 3D-Modelling of a Unique Hungarian Conquest Period (10th Century AD) Horse Burial. In: E. JEREM, F. REDŐ, V. SZEVERÉNYI (Eds.): On the Road to Reconstructing the Past. Computer Applications and Quantitative Methods in Archaeology (CAA), Proceedings of the 36th International Conference, Budapest, April 2-6, 2008. Archeaeolingua, Budapest, 2011, 348-356. 22. - LEVINE M. A.: The use of crown height measurements and eruption-wear sequences to age horse teeth. In: B. WILSON, C. GRIGSON, S. PAYNE (Eds.): Ageing and Sexing Animal Bones from Archaeological Sites, BAR International Series, vol. 109, 1982, 223-250. 23. - LEVINE M.A., WHITWELL K. E., JEFFCOTT L. B.: Abnormal thoracic vertebrae and the evolution of horse husbandry. Archaeofauna, 2005, 14, 93 109. 24. - MARKOVIĆ N., JANECZEK M., MARINKOVIĆ D., STEVANOVIĆ O., KRSTIĆ N., PETRUJKIĆ K., TRAILOVIĆ R.: Paleopathological and radiological examination of the avar period horse bones from central Balkans (Serbia). Med. Weter., 2015, 71, 619-625. 25. - MAY E.: Wideristhöhe und Langknochenmaße bei Pferd ein immer noch aktuelles Problem. Zeitschrift für Säugertierkunde, 1985, 50, 368-382. 26. - MORGAN J. P.: Spondylosis deformans in the dog. Acta. Orthop. Scand. Suppl., 1967, 96, 1 88. 27. - MÜLLER H. H.: Die Pferdeskelettfunde des Slawisch- Awarischen Gräberfeldes von Nové Zámky. Slovenska Archeologia, 1966, XIV-1, 205-225. 28. - ONAR V., ALPAK H., PAZVANT G., ARMUTAK A., CHRÓSZCZ A.: Byzantine horse skeletons of Theodosius harbour: 1. Paleopathology. Revue Méd. Vét., 2012, 163, 139-146. 29. - ONAR V., PAZVANT G., PASICKA E., ARMUTAK A., ALPAK H.: Byzantine horse skeletons of Theodosius harbour: 2. Withers height estimation. Revue Méd. Vét., 2015, 166, 30-42. 30. - PETERS J.: Römische Tierhaltung und Tierzucht: eine Synthese aus archäozoologischer Untersuchung und schriftlich-bildlicher Überlieferung. Passauer Universitätsschriften zur Archäologie 5, Rahden/ Westfalen, Leidorf, 1988. 31. - PLUSKOWSKI A., SEETAH K., MALTBY M.: Potential osteoarchaeological evidence for riding and the military use of horses at Malbork Castle, Poland. Int. J. Osteoracheol., 2010, 20, 335 342. 32. - PREMUŽIĆ Z., RAJIĆ ŠIKANJIĆ P., TRBOJEVIĆ VUKIČEVIĆ T., RAPAN PAPEŠA A., (in press). Burials with Horses at the Avar Cemetery in Nuštar. In: Proceedings of International symposium Croatian archeology and the Treaty of Aachen AD 812 ; Zadar, Croatia (2012). 33. - PRISKIN K.: Kárpát-medence avar és honfoglalás kori lóállományának archaeogenetikai elemzése (PhD dissertation), University of Szeged, 2010. 34. - RAPAN PAPEŠA A. (in press): Rescue Excavations of Avar Time Cemetery Site in Nuštar. In: Proceedings of International symposium Croatian archeology and the Treaty of Aachen AD 812 ; Zadar, Croatia (2012). 35. - SILVER I. A.: The ageing of domestic mammals. In: D. BROTHWELL and E. S. HIGGS (Eds.): Science in Archaeology: A Survey of Progress and Research. Thames and Hudson, London, 1969, 283 302. 36. - STADLER P.: Avar chronology revisited, and the question of ethnicity in the Avar qaganate. In: F. CURTA

80 TRBOJEVIC VUKICEVIC (T.) AND COLLABORATORS and R. KOVALEV (Eds.): The other Europe in the Middle Ages: Avars, Bulgars, Khazars, and Cumans, Brill, Leiden, Boston, 2008, 47-82. 37. - ŠEHIĆ M.: Osteoartropatije u domaćih životinja: Klinička rentgenologija. Manualia Universitatis studiorum Zagrebiensis, Zagreb, 2000. 38. - ŠLAUS M.: Cranial Variation and Microevolution in Two Early Middle Ages Sites from Croatia: Privlaka and Stari Jankovci. Opuscula archaeologica, 1994, 17, 273-307. 39. - ŠMALCELJ M.: Stari Jankovci Gatina. In: V. JURKIĆ- GIRARDI (Ed.): Archeology and war: research undertaken by the Department of Archeology at the Faculty of Philosophy, University of Zagreb. Handbooks of cultural affairs, vol. 5, Ministry of Education and Culture Republic of Croatia, 1992, 48-49. 40. - TEŽAK-GREGL T., ŠMALCELJ M.: Otok Gradina. In: V. JURKIĆ-GIRARDI (Ed.): Archeology and war: research undertaken by the Department of Archeology at the Faculty of Philosophy, University of Zagreb. Handbooks of cultural affairs, vol. 5, Ministry of Education and Culture Republic of Croatia, 1992, 40-41. 41. - VIDA T.: Conflict and Coexistence: The Local Population of the Carpathian Basin Under Avar Rule (Sixth to Seventh Century). In: F. CURTA and R. KOVALEV (Eds.): The other Europe in the Middle Ages: Avars, Bulgars, Khazars, and Cumans. Brill, Leiden, Boston, 2008, 13-46. 42. - VITT V. O.: Loshadi Pazyrykskykh Kurganov. Sovetskaya Archeologia, 1952, 16, 163-205. 43. - VON BABO V. F.: Pferdebestattungen auf dem frühmittelalterlichen Gräberfeld Drantumer Mühle (Gemeinde Emstek, Kreis Cloppenburg, NIedersachsen) (PhD dissertation). Tierärztliche Hochschule Hannover, Hannover, 2004. 44. - VON DEN DRIESCH A.: A Guide to the Measurement of Animal Bones from Archaeological Sites. Peabody Museum Bulletin 1, Harvard University, Cambridge, 1976. 45. - VON DEN DRIESCH A., BOESSNECK J.: Kritische Anmerkungen zur Widerristhöhenberechnung auslängenmassen vor- und frühgeschichtlicher Tierknochen. Säugetierkundliche Mitteilungen, 1974, 22, 325-348. 46. - VUKSIC V., GRBASIC Z.: Cavalry, the history of a fighting elite 650 BC-AD 1914, Cassell, London, 1993.