Skeletal Elements Found

The distribution of skeletal elements per structure indicates that the sample is mainly front limb bones (18.6%) consisting of humeri, radii, and ulnae. Front limbs are followed by detached teeth (13.1%), vertebrae (12.9%) and crania bones (12%). Other elements account for less than 10% of the total sample (rear limb 9.9%, dentary 9.5%, metapodial/phalanges 9.4%, pelvis 5.3%, ribs 4.7%, scapulae 3.1%, calcaneous 1.0%, and astragalus 0.5%) (Table 3).

Templo Redondo (Q-152) is the structure where the most complete array of skeletal elements were found. This temple is also the structure with the highest NISP (n=511) and relative density of dog remains. Artisans’ houses in the vicinity of the monumental center -Q-176a (n=34) and Q-40a (n=25)- were next in terms of NISP and the density of skeletal remains, including cranial and postcranial bones. In contrast, Q-39 has a smaller NISP (n=12) and rear limb bones predominate in its assemblage. Given that it is one of the smallest structures in this study, a relatively large number of dog bones was recovered from Q-40a (Table 2, Figure 4). Structures H-15 (n=20), H-17 (n=9), and I-55a (n=8) in the Itzmal Ch’en outlying ceremonial center have consistently smaller NISP, quantities, and densities of skeletal remains. H-15 (the Hall) has the highest numbers in this sector, which is coherent with previous observations indicating the high number of dog remains in halls (^{Pollock and Ray 1957}; ^{Masson and Peraza 2008}, ²⁰¹³).

Figure 2 Skeletal element representation by structure. 

Figure 3 Correspondence analysis map of the distribution of skeletal elements by structure. 

(modified from ^{Masson et al. 2012: figures 9.3, 11.2, 12.1 and 13.5})

Figure 4 Distribution of skeletal elements in the structures I-55a, Q-40a, Q-39, and Q-176a 

Hall H-15 in the Itzmal Ch’en outlying ceremonial center is differentiated by higher concentrations of foot and toe bones. Crania elements are absent in both the Hall (H-15) and Temple (H-17). Most of the elements found at H-17 are limb bones, especially femurs. The wealthy commoner’s house (I-55a), also in the Itzmal Ch’en center, has a small number of elements but it is mostly characterized by the presence of detached teeth (Figure 2).

A series of Chi squares and Fisher exact tests were done to compare the pairs of structures analyzed (Table 4). The results indicate a significantly different distribution of skeletal elements in Templo Redondo (Q-152), an artisan’s house (Q-39), the Hall (H-15), and the Temple (H-17).

A correspondence analysis for the distribution of skeletal elements in each structure indicates the relationship between structures and the number of skeletal elements recovered from them (Figure 3). The structures with similar distributions are represented by their proximity in the graph. The proximity of the structures to specific elements illustrates the relative frequency of said element in the structure. Two artisans’ houses (Q-40a and Q-176a) have similar compositions to I-55a and are close to Q-152. As mentioned above, Templo Redondo (Q-152) is the structure with the highest number of elements. The Hall (H-15) shows a particularly different composition with a larger number of metapodials and phalanges. The Temple (H-17) is the only structure apart from Templo Redondo where astragali were recovered. Similarly, the artisan’s house Q-39 is the only structure apart from Templo Redondo where calcanei were found.

Distribution of Dog Remains per Structure

The spatial distribution of skeletal elements in and outside the structures was analyzed for residential houses (Figure 4) and temples and halls (Figure 5). House I-55a corresponds to a wealthy artisan’s house located in the vicinity of the Itzmal Ch’en group (^{Masson et al. 2016}). The majority of the dog remains found in this house were recovered from a midden outside the northeastern corner of the structure. A radius shaft, phalanges, metapodials, skull fragments, and detached teeth were deposited in this midden. The femur located at the southwestern corner of I-55a was in the structure’s construction fill.

(modified from ^{Masson et al. 2012: figures 6.260 and 7.194}, and from ^{Delgado Kú 2004: figura 22})

Figure 5 Distribution of skeletal elements in structures H-17, H-15, and Q-152. 

Most of the dog bones found at artisan’s house Q-39 were found just outside the western side of the structure and possibly belong to domestic refuse. One of the femur shafts was found in the fill below the burial cist. Only two femur shafts were found inside the structure. At artisan’s house Q-40a, the majority of dog remains were found just outside the southwestern structure boundary. Based on the discovery of two left ulnae, it may be assumed that at least two individuals were left there. Dog remains at artisan’s house Q-176a were recovered from middens located at the western and southern outside edges of the structure. Most of the dog remains from this structure were found during the excavation of an infant burial site six meters north of the house structure. At least two dog specimens were recovered from the same level as the burial. The MNI is calculated by the presence of two left humeri and two right femur shafts. The size of slipped serving vessels found at these three houses suggest that festive meals could have been served there (^{Peraza Lope and Masson 2014b: 122}). Dog remains were also used for ornamental purposes since two of the canine teeth from artisans’ houses Q-39 and Q-176a show holes drilled through the roots, suggesting their modification for use as personal ornaments that could have been used as status or lineage paraphernalia (^{Shaw 1991: 263}). The use of dog teeth as jewelry has been widely documented throughout the Maya area (^{Clutton-Brock and Hammond 1994: 825}; ^{Hamblin 1984: 114}; ^{Teeter 2004: 188}).

Most of the dog bones recovered from the colonnaded Hall H-15 were found at the west side of the structure. A staircase leading to the Itzmal Ch’en group platform was built at the northeast corner of H-15. Evidence of a mass human burial and a burning event were discovered at this locality. The human bones recovered from this area are disarticulated, highly fragmented, and present evidence of postmortem modifications such as direct exposure to fire, blows, and cut marks (^{Serafin et al. 2012: 242}; ^{Paris et al. 2017}).

The presence of ritual ceramics such as effigy censers along with these human remains suggests that the local elites of Itzmal Ch’en were the victims of this mass killing (^{Masson et al. 2012: 18-19}). Ritual pottery appeared in great quantities in the mass grave while very little was recovered from the hall, suggesting the removal of ritual pottery from the hall, and possibly other ceremonial buildings of Itzmal Ch’en, to be placed in the grave (^{Peraza Lope and Masson 2014b: 117}). Most disarticulated dog remains were recovered at the mass burial site together with other common species in Mayapán such as iguana, white-tail deer, brocket deer, peccary, turkeys, rabbits, turtles, and various marine fish species. Animal bones found in the mass grave suggest a mixture of midden debris with human remains (^{Paris et al. 2017: 80}). Metapodials and phalanges predominate in this sample, followed by dentary fragments, vertebrae and detached teeth. In addition to the two burned dog phalanges found in this sector, several human remains from the mass grave evidence burning (^{Paris et al. 2017: 73-74}), suggesting that burning events at the mass grave site may also have altered the faunal remains deposited nearby. The rest of the dog bones reported from H-15 were recovered from the front steps of the colonnaded hall (phalanges, a molar tooth, and a femur shaft), which means that they too were disarticulated.

Dog remains were recovered from the upper levels of the western and eastern sides of Temple H-17. On the eastern side, the skeletal elements were a femur shaft, a tibia shaft, and radius as well as metapodials and astragalus. On the western side, two dentaries were found along with another femur shaft and phalanges. Itzmal Ch’en temple H-17 was an important place for rituals and administrative activities as evidenced by the great abundance of ritual pottery recovered there (^{Peraza and Masson 2014b: 118}). Along with Temple Q-58 in the center of the city, H-17 is the second tallest temple in Mayapán (^{Masson et al. 2012: 10}). The presence of serpent imagery in Itzmal Ch’en temple H-17 suggests that ceremonies dedicated to Kukulcan were important in this group (^{Chowning 1956}; ^{Masson et al. 2012: 10}). Dog bones were found in association with other faunal remains and artifacts, including Ch’en Mul effigy censers (^{Cruz and Flores 2012}).

Templo Redondo (Q-152) is where the majority of dog remains has been found and it presents the largest number of skeletal elements identified in this study. The remains are concentrated in specific areas of the Templo Redondo platform (Figure 4). Most of the dog remains in this group were found at the edges of the Templo Redondo platform and behind the colonnaded halls (Q-151) or the Hall of Chaac Masks and Q-152c or Huhí Nah. The materials recovered by the Carnegie Project at the Hall of Chaac Masks suggests ritual activities that included vessels associated with copal burning (^{Shook and Irving 1955: 148}; ^{Milbrath and Peraza 2003: 11}). Some ceramic drums were recovered from Hall Q-152c but overall there are very few materials there (^{Peraza Lope and Masson 2014b: 113}). Human skulls and disarticulated bones between Q-152 and Q-151, and between Q-152 and Q-152c, were reported by previous excavations (^{Milbrath and Peraza 2003: 11, 15}). ^{Masson and Peraza (2013: 271)} also report large quantities of dog remains at Halls Q-87a and Q-87a/Q-88a, which are part of the Templo Redondo group. The concentration of dogs in this group, along with elite serving vessels, suggests that dogs were preferred and possibly included in activities such as sacrifices, feasts, and ceremonies held at Templo Redondo and its surrounding buildings (^{Masson and Peraza 2013: 271}; ^{Peraza and Masson 2014b: 111}).

The concentration located at the western side of the platform between Hall Q-151 and Sanctuary Q-152a represents 89.2% of the dog bones recovered from the Templo Redondo group. A total of 20 right dentaries were recovered from this same location. The dog remains from Hall Q-152c were found in much smaller quantities (7.8%), and the other two right dentaries came from this hall. Other elements dispersed in Templo Redondo, its platform, and at the base of Altar Q-152d make up the remaining 3% of the total bones. It is important to mention that dog phalanges are absent in this sample and foot bones make up a small percentage of the skeletal elements reported (Table 3).

Osteometric Analysis

Osteometric analysis included measurements of maximum length and breadth of the dogs’ proximal and distal long bone ends. Given the fragmentation of the sample and the large number of unfused bones, measurements were only recorded for a sample of six complete limb bones (see Table 5) recovered at Templo Redondo, which were fused at both proximal and distal ends.

Metric data can be compared to information reported for the five breeds identified in Mesoamerica: the Mesoamerican common dog, the Mexican hairless dog or xoloitzcuintle, the short-nosed or Mayan dog, the tlalchichi dog, and a wolf-dog hybrid. The Mesoamerican common dog is the most widely identified dog breed in Mexico. It is characterized by a medium-sized body and dolichocephalic skull (^{Valadez 2000: 195}; ^{Valadez et al. 2013: 572}). The Mexican hairless dog can be found in Central Mexican and Mayan settlements. Its most outstanding osteological characteristics are the absence of premolars, small teeth, and poorly formed incisors and molars (^{Valadez 2000: 197}, ^{Valadez et al. 2009}, ²⁰¹⁰). The short-nosed dog is limited to the Maya area. Its morphological characteristics are its complete dentition, comparatively short height, short nose, and brachycephalic skull (^{Valadez 2000: 200}). ^{Valadez (2000)} and ^{Valadez et al. (1999)} argue that this type of dog corresponds to ^{Allen’s (1920)} short-nosed dog as well as to the specimens recovered by ^{Hamblin (1984)} on Cozumel Island.¹ The tlalchichi, or short-legged dog, tends to be scarce in archaeological contexts but has been found in Central Mexico sites and at Chichén Itzá. It has full dentition and a dolicho-cephalic skull, but short limbs are this breed’s main characteristic (^{Valadez 2000: 198}, ^{Valadez et al. 2011}). The wolf-dog hybrid presents a large robust head, thin face, permanent deciduous incisors and canines together, and larger dentition. Specimens of this hybrid breed have been identified in Teotihuacán and Tenochtitlan in Central Mexico, Cerro de las Minas in Oaxaca, and Xcambó in Yucatán (^{Valadez et al. 2006}; ^{Valadez et al. 2014}).

Humerus, femur, radius and tibia length from the Mayapán specimens were compared to the measurements described in the literature as these are the most consistently reported measurements (Table 1). Figure 6 shows the comparison of humerus lengths with the only complete fused humerus from Mayapán. The Mayapán example falls below measurements reported for the Mesoamerican common dog, hairless dog, and wolf-dog hybrid and above those reported for the short-nosed and the tlalchichi dogs. Figure 7 indicates a similar comparison, as Mayapán radii measurements fall below those of the Mesoamerican common dog, hairless dog, and wolf-dog hybrid and above those of the tlalchichi dog. No measurements were available for the short-nosed or Mayan dog.

Figure 6 Comparison of humerus length between Mayapán dogs and Mesoamerican common dog (^{Blanco et al. 1999}; ^{Valadez 2000}; ^{Rodríguez et al. 2001}; ^{Valadez et al. 2004}; ^{Valadez et al. 2011}), hairless dog (^{Blanco et al. 1999}; ^{Valadez 2000}; ^{Rodríguez et al. 2001}; ^{Blanco et al. 2009}), short-nosed or Mayan dog (^{Blanco et al. 1999}; ^{Hamblin 1984}; ^{Valadez 2000}), Tlachichi (^{Valadez 2000}; ^{Valadez et al. 2011}), and wolf-dog hybrid (^{Valadez et al. 2001}). 

Figure 7 Comparison of radius length between Mayapán dogs and Mesoamerican common dog (^{Blanco et al. 1999}; ^{Rodríguez et al. 2001}; ^{Valadez et al. 2004}; ^{Valadez et al. 2011}), hairless dog (^{Blanco et al. 1999}; ^{Valadez 2000}; ^{Rodríguez et al. 2001}; ^{Blanco et al. 2009}), Tlachichi (^{Valadez 2000}; ^{Valadez et al. 2011}), and wolf-dog hybrid (^{Valadez et al. 2001}; ^{Valadez et al. 2006}). 

Figure 8 shows the comparison of Mayapán femurs with the aforementioned breeds. Although only one specimen of Mayapán dog is reported in this graph, this measurement is clearly below that of the Mesoamerican common dog, hairless dog, and wolf-dog hybrid and above that of the tlalchichi dog. Femur lengths from Mayapán are the same as those reported for the short-nosed dog femur. Tibia lengths are also very similar (Figure 9). The Mayapán dog’s tibia length falls below that of the Mesoamerican common dog, hairless dog, and wolf-dog hybrid and above that of the tlalchichi dog. This profile is a closer match to the short-nosed dog. It is important to mention that although the sample size is small, the comparison of different skeletal elements can still provide clues about the breeds of dogs from Mayapán. The sizes of the Mayapán’s dog limbs are uniformly smaller than those reported for the Mesoamerican common dog and Mexican hairless dog and seem to make a better match with the smaller breeds reported for Mesoamerica. A larger sample and skull measurements are needed for more conclusive data and to account for variations among Mayapán dogs.

Figure 8 Comparison of femur length between Mayapán dogs and Mesoamerican common dog (^{Blanco et al. 1999}; ^{Valadez 2000}; ^{Valadez et al. 2004}; ^{Blanco et al. 2009}; ^{Valadez et al. 2011}), hairless dog (^{Valadez 2000}; ^{Rodríguez et al. 2001}), short-nosed or Mayan dog (^{Blanco et al. 1999}), Tlachichi (^{Valadez 2000}; ^{Valadez et al. 2011}), and wolf-dog hybrid (^{Valadez et al. 2001}; ^{Valadez et al. 2006}). 

Figure 9 Comparison of tibia length between Mayapán dogs and Mesoamerican common dog (^{Blanco et al. 1999}; ^{Valadez 2000}; ^{Rodríguez et al. 2001}; ^{Valadez et al. 2004}; ^{Blanco et al. 2009}; ^{Valadez et al. 2011}), hairless dog (^{Blanco et al. 1999}; ^{Valadez 2000}; ^{Rodríguez et al. 2001}; ^{Blanco et al. 2009}), short-nosed or Mayan dog (^{Blanco et al. 1999}; ^{Hamblin 1984}; ^{Valadez 2000}), Tlachichi (^{Valadez 2000}; ^{Valadez et al. 2011}; ^{Valadez and Rodríguez 2012}), and wolf-dog hybrid (^{Valadez et al. 2001}; ^{Valadez et al. 2006}). 

Measurements of first molar tooth length and breadth were taken for 27 teeth, including detached teeth and teeth attached to maxillae and dentaries (Table 6). The measurements of the first lower and upper molar were compared to those given by ^{Rodríguez et al. (2001)}, ^{Valadez (2000)}, ^{Valadez et al. (2004}, ²⁰⁰⁹⁾ and ^{Blanco et al. (2009)} for the same teeth in the Mesoamerican common dog, hairless dog, short-nosed dog, and wolf-dog hybrid (Figure 10). Measurements of both breath and length are not provided in the literature fortlalchichi breeds. Ranges of length for lower first molar were obtained from ^{Allen (1920)}, ^{Rodríguez et al. (2001)}, ^{Valadez (2000)}, ^{Valadez et al. (2004}, ²⁰⁰⁹⁾ in examples of the Mesoamerican common dog, Mexican hairless dog, short-nosed dog, tlalchichi dog, and wolf-dog hybrid. The lower first molar teeth from Mayapán have lengths of 17-18 mm. These lengths clearly fall within the range of the short-nosed dog (17.5-18.7 mm) and the Mexican hairless dog (17.1-19.9 mm).² They are below the ranges reported for the Mesoamerican common dog (18.7-21.2 mm), tlalchichi (18.5 mm), and wolf-dog hybrid (25 mm). When measurements of both breath and length were given in the literature and compared to the Mayapán examples, it is clear that both lower and upper first molar dimensions fall below those given for the Mesoamerican common dog, hairless dog and wolf-dog hybrids. The reported measurement for the short-nosed or Mayan dog³ (^{Blanco et al. 2009: table 12-14}) falls within the range of measurements for upper and lower molars of Mayapán. Further analysis of dentition and additional cranial and tooth measurements are needed to make better comparisons. For the moment, the dental information suggests two possibilities: (1) that there are strong similarities with smaller breeds of dogs, or (2) that there are hairless dogs within the sample which usually have atypical teeth.

Figure 10 Comparison of upper and lower first molars between Mayapán dogs and Mesoamerican common dog (^{Rodríguez et al. 2001}; ^{Valadez et al. 2004}), hairless dog (^{Rodríguez et al. 2001}), the dog-wolf hybrid (^{Valadez et al. 2001}), and the short-nosed or Mayan dog (^{Blanco et al. 2009}). 

Measurements of mandibular height at the lower first molar were used to calculate the body weight of specimens.⁴ Mayapán body weights range from 3,408 g to 10,841 g. ^{Blanco et al. (1999: Figure 7 and Figure 9)} report body weight calculations for different breeds, including examples of two common Mesoamerican dogs found at Chac-Mool with weights of 9,480 g and 9,420 g and a short-nosed or Mayan dog weighing 8,610 g; a common Mesoamerican dog from Teotihuacán with a weight of 9,420 g; examples from Tula include two common Mesoamerican dogs with weights of 11,265 g and 9,885 g, a hairless dog weighing 11,475 g and a tlalchichi with a weight of 10,485 g. Moreover, ^{Valadez et al. (2001)} report the following range of weights: common Mesoamerican and hairless dogs from 9,000 to 10,000 g, the short-nosed or Mayan dog 5,500 to 8,700 g, and the tlalchichi 9,500 g.

Body weight estimates indicate that the smallest specimen is found in artisan’s house Q-40a, weighing an estimated 3,408 g, while the largest is at Templo Redondo, with an estimated weight of 10,841 g (Table 7). Weights below 9,000 g are more common in the Mayapán sample (n=14, 70%), suggesting the presence of smaller breeds such as the short-nosed dog. These were found in all the structures analyzed. On the other hand, the few examples (n=6, 30%) with weights above 9,000 g were recovered from Templo Redondo, which suggests the presence of larger breeds in this context.

^{Hamblin (1984: 100)} cautions that body weight variation could be related to breeds or age. In the Mayapán sample, almost all (n=18, 90%) of the mandibles measured exhibited erupted teeth, which occurs when the animal is about seven months old (^{Sutton et al. 2018: Appendix}). Two examples (10%) with unerupted third molars (with weights of 8,742 g and 9,333 g) in mandibles were recovered from Templo Redondo.

Age Distribution

Dog ages were calculated according to published epiphyseal fusion schedules (^{Sutton et al. 2018: table 11.3, table 11.7}). The fragmentation of the sample made it impossible to determine ages for all specimens. A total of 51.9% (n=125) of the sample of limb bones and foot bones were sufficiently preserved to document epiphyseal fusion stages. Limb bones with fused proximal and distal ends (or both) comprise 68% of the sample, while 32% present unfused epiphyses.

Table 8 presents late-fusing long bones along with metapodials and calcanea, which fuse early. The majority of elements from Mayapán are fused, indicating that the growth and ossification stage had been completed. Unfused elements were not recovered in all of the structures investigated. The majority of unfused bones were found in the Templo Redondo group, followed by House Q-40a and Hall H-15. The remaining structures present only fully fused bones (Table 9), suggesting the presence of adult dogs.

It is important to note that the sample from domestic contexts included in this analysis is very small and the majority of bones reported correspond to metapodials and calcanea, which are early-fusing elements.