Projects
The Value of Nonmetric Trait Analysis
Anthropology 490
April 2005
The field of physical anthropology is always looking at new techniques that allow researchers to examine the past life-ways of different peoples. With recent advancements in ancient DNA analysis, stable isotope analysis, and other forms of chemical testing on bone, older research methods are sometimes forgotten with the excitement surrounding new, novel ideas. Can older, traditional research methods still contribute scientifically significant information to physical anthropology? Or should researchers leave them behind in favor of these new techniques? While there are many different methods available for use by researchers, this essay will focus on a discussion of nonmetric trait analysis. It will examine what this type of analysis can tell researchers, and discuss the value of nonmetric analysis for study of past populations. Before investigating the significance of nonmetric trait analysis (also known as discrete trait analysis), I will begin by discussing different types of nonmetric traits. Briefly, nonmetric traits are variations of the skeleton that are generally minor in nature, scored as present or absent, sometimes measured in terms of the extent to which the trait is present, are non-pathological and do not interfere with the normal functions of the human body (Mays, 1998). There are many different nonmetric traits that have been identified. In numbers upwards of 400, nonmetric traits generally fall into two categories: those involving supernumerary bones and those involving a difference in the 'normal' shape or development of bones. There are many different ways of classifying nonmetric traits. For example, Shelley Saunders developed a system that separates eight categories of nonmetric traits. She lists these as, "hyperostotic, hypostotic, foramina/canals/grooves, supernumerary vault sutures, crainobasal structures, spinal structures, prominent bony processes, and facet variations," (Mays,
1998)1. However, for the purposes of this essay I will list several examples of the most common variations, divided into cranial nonmetric traits and post-cranial nonmetric traits.
As there are far too many known nonmetric traits to discuss them all here, some common traits on the cranium include supernumerary ossicles occurring on suture lines, inca bones, metopic suture retention, variations in the number and location of foramina, and bony tori occurring on the palatine suture or lingual aspects of the mandible (El-Najjar and McWilliams, 1978). Dental nonmetric traits are also common. These can include supernumerary teeth, missing teeth, variable numbers of cusps, shovel shaped incisors, and variations in the number of roots on molars (White, 2000). Dental traits are often studied, as teeth generally survive well in the archaeological record (White, 2000), compared with the softer bone of the skeleton that is more susceptible to diagenesis and degradation.
Nonmetric traits also occur in the post-cranial skeleton. Some common traits include supernumerary vertebrae, ribs, and bones in the hand and wrist, perforation of the olecranon fossa, accessory foramina in cervical vertebra, posterior bridging of the atlas, third femoral trochanters, sternal foramina, and variations of the surface of facets (White, 2000; Mays 1998; Haas, 1994).
There is much debate in the field of physical anthropology over whether nonmetric traits are genetically controlled, environmentally determined, or due to habitual activity. However, most research is performed on the assumption that traits are genetically determined, and therefore genetically inherited traits will be focused on in this essay. Simon Mays (1998) discusses the evidence for genetic determination. He examines the results from a study by a British researcher, Caroline Berry. Berry's study examined the frequency of the rate of occurrence of shovel shaped incisors, between children and their parents. They found that the frequencies do appear to rely on genetic inheritance, as the rate of occurrence in the children was 69% if two parents demonstrated the trait, and 57% if only one parent had it. For those children of whom neither parent demonstrated the trait, only 29% of children had shovel shaped incisors (Mays, 1998). Another example Mays (1998) uses to show that some nonmetric traits can be genetically inherited is of the palatine torus. In a Japanese study, 52% of offspring, for whom one parent showed the trait, also demonstrated it. The palatine torus only presented itself in 8% of children when neither parent had the trait. These studies indicate that there is at least some genetic determination of nonmetric trait expression.
In addition to genetic determination, the environmental conditions in which a population lives can affect the frequency of expression of nonmetric traits. L.A.P. Kohn (1991) writes, "Environmental deviations can be attributed to specific influences such as nutrition, climate, maternal care (both prenatal and postnatal), shared or common environment, or other factors." Studies have also been done on environmental determination of nonmetric traits. Jean-Pierre Bocquet-Appel examined the effects of wheat prices (and therefor the availability of a certain nutritional staple), on the expression of nonmetric traits, looking at an extremely large collection of 2000 nineteenth and early twentieth-century skulls from Portugal (Mays, 1998). He looked at five age groups, those born every 20 years from 1810 to 1909, and found that there was indeed a correlation between some nonmetric traits, such as suture ossicles and several foramina, and the price of wheat (Mays, 1998). Though environmental factors do affect nonmetric traits, and most studies mention this factor, there is much less research that has been done on environmental nonmetric trait expression then when compared with genetic studies.
Finally, habitual activity patterns also appear to affect nonmetric trait expression. These phenomena appear most often in relation to the bones present in the joints. A common example given to illustrate the effect of habitual activity are the so called "squatting facets". These facets are said to occur in populations that spend a lot of time squatting on their heels (Mays, 1998; Brothwell, 1965). Another type of habitual activity that has been correlated with nonmetric trait expression is the presence of bony tori in the external auditory canal as the result of repeated exposure to cold water (White, 2000; Mays, 1998).
Nonmetric trait analysis is most often performed to examine biological affinities. Biological affinities between different groups of people can be studied using the frequency of occurrence of nonmetric traits. However, White (2000) notes that this type of study does have an underlying assumption present. White (2000) writes, "Populations that display the most similarity are the most closely related". These studies also assume that nonmetric traits are genetically inherited, and while some studies have been done on environmental and habitual activities resulting in trait expression, the focus is generally on the genetic relatedness of certain groups. In addition, as with all studies performed on archaeological groups, there are some inherent difficulties that are encountered. It must be noted that when studying archaeological populations, the sample base that one studies is not necessarily an exact reflection of the true make-up of the entire group (White, 2000). There are many factors involved in this. For example, a cemetery might be made up of a certain social class that when excavated, is only representative of a small portion of the population on the whole. According to Douglas H. Ubelaker (as cited by White, 2000), to attempt to account for this problem, a sample size should be of 100 individuals at minimum.
There are other difficulties involved with nonmetric studies. Mays (1998) notes that one of the problems is that many traits cannot be studied on living individuals. In some cases, radiography can be used. For example, x-rays work for the study of genetic inheritance regarding metopic suture retention, missing third molars, and supernumerary bones of the post-cranial skeleton. However, many traits do not appear clearly on x-rays, and therefore, studies must be performed on dry bone (Mays 1998). This leads to a paradox when looking at skeletal samples. Most often, familial relationships are not identified in cemeteries, especially in those of non-western populations. It is in this type of situation when anthropologists study nonmetric traits as a way of assigning biological affinity. The paradox lies in that how can researchers be sure that the traits they are studying are genetically inherited, without doing studies to look at genetic trait inheritance itself? One way that researchers have examined this problem is to study a sample where the biological relationships of the skeletons were known. According to Mays (1998), in Europe, due to the limitations of space for burials, skeletons were exhumed from cemeteries when the cemetery ran out of room, and then were stored. During the eighteenth and nineteenth centuries, the exhumed skeletons were sometimes painted with the name of the individual before being stored in ossuaries. Some of these skeletal collections still remain available for study, and using historical records such as birth or marriage documents, researchers have been able to study patterns of inheritance, as the familial relationships are known. This is just one example of how researchers are working to resolve the paradox inherent in nonmetric analysis.
The next section of this essay will discuss three different case studies that deal with nonmetric trait analysis. The first study, by D.A. Donlon (2000), examined infracranial (postcranial) nonmetric traits in Australian populations. As the focus in nonmetric trait analysis tends to be on cranial traits (as oftentimes, many collections are made of solely of crania), the hypothesis in this study was that postcranial traits are still of value to research. In this study, the samples were chosen from those skeletons whose ancestors had come to Australia from five different areas, including Africa, East Asia, Europe, Polynesia, and Australia. This was to study the widest range of postcranial traits possible. Also used were two other groups from Australia, the main purpose being control groups to determine the extent of inter and intra population variation. Only adult skeletons were used in the study. A total of 20 traits were examined in the final stages of the study (some of the original traits being discarded due to non-variability or the likelihood that the trait was more determined environmentally, such as squatting facets). The results found that Australian aboriginals were most closely related to groups from South Asia. However, one problem that the study had was the amount of inter-sample variation as opposed to intra-sample variation. One suggestion the study makes is that while research on nonmetric postcranial traits is still of value, studies should be performed on groups within a certain region as opposed to unrelated groups. Donlon also noted that larger sample sizes would be of value.
The next case study, by Richard C. Sutter and Lisa Mertz (2004), examined nonmetric cranial traits in prehistoric Chile. Their hypothesis focused on determining the biological continuity or discontinuity in the people of the Azapa valley. They examined eight burial samples with 3 of the eight samples being combined due to previously determined evidence that the three burial sites actually represented the same group. Of the remaining samples, two dated from circa 1000 B.C.E. to 500 C.E., two samples dated from circa 750 C.E. to 1100 C.E., and one sample dated from circa 1100 C.E. to 1476 C.E. The researchers chose 37 cranial traits to study, including sutural ossicles, foramina, metopic suture retention, and auditory and palatine tori. The researchers statistically studied any traits that were significantly correlated between the sexes and those traits that were inter-correlated, and did not include these in the study. The original 37 traits were narrowed down to eight traits, as 27 failed to be significantly different, and two were correlated significantly between the sexes. The nonmetric traits studied were the parietal notch, the presence of multiple infra-orbital foramina, mylohyoid bridging, palatine tori, and the presence of coronal ossicles. Three of the eight traits were not found to be statistically significant between the studied samples. The study concluded that it was difficult to tell any clear trends between the groups, when based on chronology. Geographically, the coastal peoples tend to be more similar to each other, and the inland peoples also more similar to each other. However, the study determined that on the whole, the results were not statistically significant enough to determine biological distance, and concluded that the peoples of the Azapa valley were a relatively congruous breeding population. They also concluded the small amount of biological distance that was apparent was due to genetic drift and non-significant gene flow. They do however list several problematic areas that may have effected their results, such as whether or not cranial deformation affects nonmetric trait expression, and that eight nonmetric traits may not have been enough to compare between the samples. They suggest that studies over a larger geographical area might prove more significant.
A third study also examined biological affinities using cranial nonmetric traits, however they examined specimens from around the world and compared them to each other to find inter-regional correlation, support for previous studies of biological distance, and then apply the results to the study of human origins. The study, performed by Tsunehiko Hanihara, Hajime Ishida, and Yukio Dodo (2003), looked at 20 nonmetric traits, from 70 samples. The samples they used consisted of adult specimens (judged by sphenooccipital fusion and eruption of the third molars where applicable), with no cranial deformation to negate the possibility of cranial deformation affecting the instances of nonmetric trait expression. Twenty nonmetric traits were studied, from a total of 8000 skeletal specimens from the 70 groups. The scoring process was limited to the presence or absence of the trait, and if the trait only occurred unilaterally it was scored as present (where relevant). Due to a relatively similar ratio between males and females in all of the populations except the European ones (who were predominantly male), the different sexes were not examined independently of one another. The study found statistically significant results, and grouped the different populations into three overall categories. These were differentiated as the Subsaharan African group, the Southeast Asian group, and the Oceanian group, with the Subsaharan African group being much farther removed from the other two groups. The results they found do support the single-origin hypothesis, the hypothesis being that the Subsaharan African populations would be significantly different from the rest of the studied groups. However, they also note that the results do not disprove the multiregional hypothesis due to Subsaharan Africa having the greatest long-term population size. The researchers also acknowledged that some of the cranial nonmetric traits could be related to environmental adaptations, or different subsistence patterns. However, as one of the main research goals in the study was to find if they could replicate the results from previous genetic analyses, and analyses based on other morphological traits, the researchers state that the results they have compiled support this goal.
This study demonstrated the usefulness in looking at samples from a geographically large area on which to perform nonmetric trait analysis. Now obviously most studies will not focus on a world-wide scale, due to such issues like costs, time available, and the individual study's research questions, however all three studies confirm the need for large samples as proposed previously, by Ubelaker (White, 2000). Overall, nonmetric trait analysis can still provide useful information for physical anthropologists, however further study and refinement of the technique is required.
In the future, it might be useful to integrate the new scientific techniques with traditional methods of anthropological study. Due to the current popularity of DNA analysis, combining it with nonmetric trait analysis may be an area for further study. As previously mentioned, the cause of many nonmetric trait variations are unknown. Research on the genes responsible for nonmetric traits, perhaps in conjunction with the Human Genome Project2, could be a possible direction for anthropological study. DNA analysis could also be used to study the extent to which environmental or habitual activity affects nonmetric trait expression. For example, if one could find the gene responsible for a certain nonmetric trait, one could test people who displayed the trait for that gene, and conversely, look at people who tested positive for the gene who did not display the nonmetric trait. In a study of the development of a third molar in mice, researchers were attempting to examine the genetic control over the trait. Although they found that a genetic trait was responsible for third molar development, they also found that it was environmentally influenced (Mays, 1998). Further study could be done in an attempt to discover what traits are environmentally affected, and what traits (if any) are solely determined genetically.
With the current interest in new and exciting fields of scientific research, traditional techniques, such as nonmetric analysis may be overlooked. However, these new analyses are sometimes extremely expensive, and bring up other problems such as the necessity to destroy part of one's sample, which may not be an option. Older techniques can still provide researchers with useful information, and still have room to be refined. Perhaps by combining the new techniques with the traditional ones, researchers will be able to further validate and perfect methods popular in the past.
1 Hyperostotic traits are those identified as an abnormal bony formation, and hypostotic traits are a bone deficiency such as a concavity in the bone or a perforation (Mays, 1998).
2 See www.genome.gov for more information on the Human Genome Project.
Brothwell, Don R. 1965. Digging up bones: The excavation, treatment and study of human skeletal remains. London: Trustees of the British Museum (Natural history).
Donlon, D.A. The value of infracranial nonmetric variation in studies of modern Homo sapiens: An Australian focus. 2000. Am J Phys Anthropol 113:349-368.
El-Najjar, Mahmoud, and McWilliams, Richard K. 1978. Forensic anthropology: The structure, morphology, and variation of human bone and dentition. Springfield: Charles C. Tomas.
Haas, Jonathan. 1994. Standards: For data collection from human skeletal remains, proceedings of a seminar at the Field Museum of Natural History. Jane E. Buikstra and Douglas H. Ubelaker, eds. Arkansas Archaeological Survey Research Series No. 44. Fayetteville Ark: Arkansas Archaeological Survey.
Hanihara, Tsunehik, Ishida, Hajime, and Dodo, Yukio. 2003. Characterization of biological diversity through analysis of discrete cranial traits. Am J Phys Anthropol 121:241-251. http://www3.interscience.wiley.com/cgi-bin/fulltext/ 104081817/HTMLSTART. (accessed April 14, 2005).
Kohn, L.A.P. 1991. The role of genetics in craniofacial morphology and growth. Annu. Rev. Anthropol 20:261-278. http://links.jstor.org/sici?sici=0084-6570%281991%292%3 A20%3C261%3ATROGIC%3E2.0.CO%3B2-O (accessed April 14, 2005).
Mays, Simon. 1998. The archaeology of human bones. London: Routledge.
Sutter, Richard C. and Mertz, Lisa. 2004. Nonmetric cranial trait variation and prehistoric biocultural change in the Azapa valley, Chile. Am J Phys Anthropol 123:130-145. http://www3.interscience.wiley.com/cgi-bin/fulltext/ 104537948/HTMLSTART (accessed April 14, 2005).
White, Tim D. 2000. Human osteology. 2nd edition. San Diego: Academic Press.
© 2005 Rosanne Lester