Journal of Undergraduate Research
Volume 6, Issue 9 - July 2002

On the Possible Prehistoric Domestication of the Caribbean hutia, Isolobodon Portoricensis

Brooke Garner

INTRODUCTION

Over the past forty years, researchers have debated whether or not the prehistoric Amerindians of the Caribbean islands domesticated the hutia, Isolobodon portoricensis. Bones of this large rodent are frequent constituents of archaeological sites in the region. Were hutias simply a hunted wild resource or were they undergoing domestication? The processes of domestication can be detected by changes in the skeletal morphology of a species over time. Yet, until now, no study has compared the hutia skeletal morphology of natural death assemblages to that of zooarchaeological ones for the purpose of addressing the issue. I. portoricensis has long been extinct but skeletal specimens of late Pleistocene-early Holocene age are available to serve as natural-death assemblages. Focusing on mandibular tooth-row length as an analog for hutia body size and femur epiphyseal fusion as an indicator for neotony, two paleontological assemblages from the Greater Antilles and two zooarchaeological assemblages from St. John, U.S. Virgin Islands were selected for comparative morphometric analysis.

Animal Domestication

A domestic animal is defined as one that has been bred in captivity for use by humans (Clutton-Brock 1981). Domestication can be identified in the zooarchaeological record from changes in morphology. One example is change in tooth-row length, as an analog for change in body size. Another is neotony, the selection for juvenile characteristics that select for animals that mature at a younger age (Clutton-Brock 1981). The presence of the epiphyseal fusion of the long bones has been used to identify such selection. Since fusion usually occurs around the time of matuaration, early epiphyseal fusion may indicate that individuals are maturing earlier in ontogeny (Reitz and Wing 1999:294).

Morphological changes associated with domestication can be difficult to distinguish from naturally occurring processes (Clutton-Brock 1981). For example, founder effect is a naturally occurring process involving the fission of a group of animals from a parent population (Mayr 1942). Genetic diversity is limited in such a fissioned population; therefore, variation becomes limited. Founder effect can occur through the natural isolation of a population or through human transport of animals to isolated locations. In either case, the results of founder effect in animal populations can be similar.

Isolobodon Portoricensis

There is evidence that Amerindians kept and transported hutias. Isolobodon portoricensis was native only to Hispaniola (Woods 1989:758) until Amerindians colonized the West Indies. Subsequently hutia remains became incorporated into archaeological middens throughout the West Indies (Flemming and MacPhee 1999:4; Woods 1989:758). Evidence for I. portoricensis being tended by humans is supported by the fact that relative to other hutia species of Hispaniola only I. portoricensis is found in archaeological deposits (Woods 1989:758). Furthermore, there is no evidence of other hutia species of Hispaniola dispersing to other islands, therefore, it can be argued that humans were instrumental in their dispersal.

MATERIALS AND METHODS

Figure 1. Maps of Hispaniola and St. John, U.S. Virgin Islands with sites (a) Rancho de la Guardia, (b) Trou Woch SaWo (c) Cinnamon Bay and (d) Calabash Boom.

Materials

Skeletal specimens of the extinct Isolobodon portoricensis were examined from the late Pleistocene-early Holocene cave sites of Rancho de la Guardia (Figure 1a) and Trou Woch SaWo (Figure 1b) and represent natural-death assemblages (Gary Morgan, personal communication 2001). Cinnamon Bay (Figure 1c) and Calabash Boom (Figure 1d), St. John, Virgin Islands, contained zooarchaeological hutias. Calabash Boom dates between A.D. 740 and 1460 and represents a refuse midden (Quitmyer 2001). Cinnamon Bay was a ceremonial site used from A.D. 1000 to 1490 (Wild 1999). The paleontological specimens are curated in the Vertebrate Paleontology division of the Florida Museum of Natural History (FLMNH). The Calabash Boom specimens are curated at the Environmental Archaeology laboratory at the FLMNH and the Cinnamon Bay specimens are curated by the National Park Service in St. John, Virgin Islands.

Methods

The The mandible tooth row length (TRL) was the only measurement considered for this study because it is the most abundant skeletal feature in all four sites (Figure 2).

Figure 2. (a) Mandibular tooth-row length (TRL). Actual toal length of mandible: 45.27 mm.

Left and right rami were identified to avoid measuring the same individual. The largest number of measurements from each site was compared among all sites. Each assemblage of measurements was divided into arbitrary size ranges of 8-10 mm, 11-13 mm, 14-16 mm, 17-19 mm, and 20-22 mm in order to identify the dominant size range of Isolobodon portoricensis at each assemblage. For each assemblage the mean and 95 percent confidence intervals around the mean were calculated. This statistical treatment is sufficiently robust to test the null hypothesis that all of these samples are form the same population (H_o: sample 1 = sample 2 = sample 3 = sample 4). The means of the samples that are statistically significant where the confidence intervals do not overlap (p≤.05).

To evaluate the number of ontogenetically younger animals in an assemblage, the femurs were separated into older and younger individuals based on their fusion of either proximal or distal ends (e.g., Figure 3). It is not known when fusion occurred during the lifetime of hutia individuals, only that it took place after growth abated. It is assumed that animals with unfused femurs are younger than animals with fused femurs. Only the zooarchaeological specimens were used for this evaluation because of the lack of paleontological long bones.

Figure 3. Two femurs from St. John, U.S. Virgin Islands. Notice that the proximal end of the right femur is unfused (a) and the distal end is beginning to fuse (b). The proximal end of the left femur is fused (c) but the distal end isnot fused (d). The right femur's actual length is 57.21 mm. The left femur's actual length is 60.33 mm.

RESULTS

The Rancho de la Guardia mandibular TRLs (n=142) range 11-19 mm with the majority (66.9%) being 14-16 mm (Figure 4a). The mean is 15.00 mm with a standard deviation of 1.55 mm and a 95 percent confidence interval of 0.26 mm. Trou Woch SaWo mandibular TRLs (n=97) range 8-20 mm with the majority (47.4%) being 14-16 mm (Figure 4b). The mean is 14.02 mm with a standard deviation of 2.25 mm and a 95 percent confidence interval of 0.45 mm.

 

Figure 4. The size ranges of mandible alveolar tooth row lengths at Rancho de la Guardia, Dominican Republic; Trou Woch SaWo, Haiti; Calabash Boom, St. John Virgin Island; and Cinnamon Bay, St. John Virgin Island respectively

The Calabash Boom mandibular TRLs range 14-22 mm with the majority (72.4%) being 17-19 mm (Figure 4c). The mean is 17.36 mm with a standard deviation of 1.42 mm and a 95 percent confidence interval of 0.54 mm. An almost equal number of fused femurs and unfused femurs are present in the assemblage. The Cinnamon Bay mandibular TRLs range 14-22 mm with the majority (80.4%) being 17-19 mm (Figure 4d). The mean is 17.47 mm with a standard deviation of 1.21 mm and a 95 percent confidence interval of 0.32 mm. An almost equal number of fused femurs and unfused femurs were present in the assemblage. A comparison of the 95 percent confidence intervals of the paleontological assemblages indicate that the means of the TRLs are not significantly different (p≤.05) (Figure 5) and could have been sampled from the same population. The mean TRLs from the zooarchaeological sites are also not significantly different from each other (Figure 5). When compared to each other, however, the paleontological assemblages are significantly different from the zooarchaeological ones (Figure 5). The standard deviations of the TRLs from the paleontological assemblages (1.55 and 2.25) (Figure 4a, b) are greater than those from the zooarchaeological assemblages (1.42 and 1.21) (Figure 4c, d).

 

 

Figure 5. Means (open circles) of mandible tooth-row length and 95% confidence intervals (bars) around the mean of hutia tooth-row lengths comparing paleontological and zooarchaeological populations

DISCUSSION

A lack of data has led previous researchers to doubt that Amerindians domesticated Isolobodon portoricensis. Fossil and zooarchaeological data, as used in the present study, are not perfect and limitations still exist. These new data do not definitively show that I. portoricensis was domesticated, but they do add significant evidence that humans influenced the size of I. portoricensis in a more direct way than simply by hunting them.

Based on mandible TRLs, the paleontological assemblages on average contained similar-sized Isolobodon portoricensis remains. But they also indicate a wide range of body sizes. Comparatively, both zooarchaeological assemblages contain remains from larger individuals and exhibit smaller ranges in body sizes. The smaller standard deviations for the zooarchaeological assemblages compared to the paleontological assemblages indicate human selection for larger individuals.

There are various possibilities for why larger Isolobodon portoricensis are represented at the St. John archaeological sites. The most likely are that the larger zooarchaeological animals were the subject of early domestication and/or founder effect took place. Human transport of I. portoricensis or arrival by natural means from Hispaniola would have isolated the population and allowed for the results of a founder effect (i.e. larger animals). The question then, is: was this a natural event or did it take place through human agency? Since I. portoricensis has never been found in paleontological sites of the Virgin Islands it seems likely that their arrival took place with humans. Their exclusive occurrence in zooarchaeological assemblages suggests a strong association with humans.

Furthermore, the differences in body sizes between the paleontological and zooarchaeological assemblages are suggestive of selective breeding (domestication) of Isolobodon portoricensis.

Proving any of these explanations is difficult, especially since distinguishing from a state of captivity, selective breeding, and founder effect is extremely hard. One way to distinguish domestication would be to look for skeletal evidence of neotony. The lack of neotony-influenced skeletal elements from all sites made this impossible. But since femurs were present in the zooarchaeological assemblages their epiphyseal features were analyzed, demonstrating that about half of the specimens were not fused. The remains of larger I. portoricensis that lack fused femurs implies that the zooarchaeological populations had young adults that were larger, a direct indication of selective breeding for size.

CONCLUSION

To definitively say that Isolobodon portoricensis was bred to a domesticated state before its extinction, one would have to disprove that a founder effect was the source of the size differences between the paleontological and zooarchaeological assemblages. Even though my results suggest this to be the case, they are not definitive. It is know that Amerindians specifically sought I. portoricensis as a food source and it is likely they had them in captivity. These conclusions are based on the lack of Virgin Island paleontological assemblages bearing I. portoricensis. The difference in sizes indicate that a genetic drift existed between the paleontological and archaeological assemblages. Due to limitations of this research, it is impossible to definitively determine if selective breeding took place. Further research is needed, using more and larger sample sizes, and perhaps techniques such as DNA analyzing.

ACKNOWLEDGEMENTS

My utmost gratitude goes to the University Scholars Program for the award that allowed me to accomplish this research. I thank the Environmental Archaeology, Mammalogy, and Vertebrate Paleontology divisions at the Florida Museum of Natural History and archaeologist Ken Wild for specimens used in this study. The Environmental Archaeology laboratory greatly supported this work with workspace and guidance. I thank Elizabeth Wing for sharing her knowledge of domestication and hutia, While Irvy Quitmyer helped guide this research. Karen Walker served as my advisor where she skillfully directed my efforts and provided editorial guidance.

REFERENCES

1. Clutton-Brock, J. 1981
   Domesticated Animals from Earlier Times. University of Texas Press, Austin.

2. Flemming C., and R. D. E. MacPhee 1999
   Redetermination of Holotype of Isolobodon portoricensis (Rodentia, Capromyidae), with Notes on Recent Mammalian Extinctions in Puerto Rico. American Museum Novitates 3278:11.

3. Mayr, Ernst 1942
   Systematics and the Origin of Species. Columbia University Press, New York.

4. Quitmyer, I. R. 2001
   The Zooarchaeology of Cinnamon Bay (12Vam-2-3), St. John, U.S. Virgin Islands: The Pre-Columbian Overexploitation of Animal Resources. Bulletin of FLMNH (Papers in Honor of E. S. Wing), in press.

5. Reitz, E. J., and E. S. Wing 1999
   Zooarchaeology. Cambridge University Press, Cambridge.

6. Wild, K. S. 1999
   Investigations of a "Caney" at Cinnamon Bay, St. John and Social Ideology in the Virgin Islands as Reflected in Pre-Columbian Ceramics. Paper presented at the XVIth International Congress for Caribbean Archaeology, Guadeloupe.

7. Woods, C. A. 1989
   The Biogeography of West Indian Rodents. In Biogeography of the West Indies: Past, Present and Future, edited by C. A. Woods, pp. 741-798. Sandhill Crane Press, Gainesville, Florida.

--top--

Back to the Journal of Undergraduate Research