An Examination of Nubian and Egyptian biological distances: Support for biological diffusion or in situ development?
Homo. 2009;60(5):389-404. Epub 2009 Sep 19.
Department of Anthropology, University of Tennessee, Knoxville, 250 South Stadium Hall, Knoxville, TN 37996, USA. firstname.lastname@example.org
Many authors have speculated on Nubian biological evolution. Because of the contact Nubians had with other peoples, migration and/or invasion (biological diffusion) were originally thought to be the biological mechanism for skeletal changes in Nubians. Later, a new hypothesis was put forth, the in situ hypothesis. The new hypothesis postulated that Nubians evolved in situ, without much genetic influence from foreign populations. This study examined 12 Egyptian and Nubian groups in an effort to explore the relationship between the two populations and to test the in situ hypothesis. Data from nine cranial nonmetric traits were assessed for an estimate of biological distance, using Mahalanobis D(2) with a tetrachoric matrix. The distance scores were then input into principal coordinates analysis (PCO) to depict the relationships between the two populations. PCO detected 60% of the variation in the first two principal coordinates. A plot of the distance scores revealed only one cluster; the Nubian and Egyptian groups clustered together. The grouping of the Nubians and Egyptians indicates there may have been some sort of gene flow between these groups of Nubians and Egyptians. However, common adaptation to similar environments may also be responsible for this pattern. Although the predominant results in this study appear to support the biological diffusion hypothesis, the in situ hypothesis was not completely negated.
The Mahalanobis D2 analysis uncovered close affinities between Nubians and Egyptians. Table 3 lists the Mahalanobis D2 distance matrix. As there is no significance testing that is available to be applied to this form of Mahalanobis distances, the biodistance scores must be interpreted in relation to one another, rather than on a general scale. In some cases, the statistics reveal that the Egyptian samples were more similar to Nubian samples than to other Egyptian samples (e.g. Gizeh and Hesa/Biga) and vice versa (e.g. Badari and Kerma, Naqada and Christian). These relationships are further depicted in the PCO plot (Fig. 2). Aside from these interpopulation relationships, some Nubian groups are still more similar to other Nubians and some Egyptians are more similar to other Egyptian samples. Moreover, although the Nubian and Egyptian samples formed one well-distributed group, the Egyptian samples clustered in the upper left region, while the Nubians concentrated in the lower right of the plot. One line can be drawn that would separate the closely dispersed Egyptians and Nubians. The predynastic Egyptian samples clustered together (Badari and Naqada), while Gizeh most closely groups with the Lisht sample. The first two principal coordinates from PCO account for 60% of the variation in the samples. The graph from PCO is basically a pictorial representation of the distance matrix and interpretations from the plot mirror the Mahalanobis D2 matrix.
The clustering of the Nubian and Egyptian samples together supports this paper’s hypothesis and demonstrates that there may be a close relationship between the two populations. This relationship is consistent with Berry and Berry (1972), among others, who noted a similarity between Nubians and Egyptians. If Nubians and Egyptians were not biologically similar, one would expect the scores to separately cluster by population (e.g. Nubians compared to Nubians would have small biological distances, and Nubians compared to Egyptians would have high biological distances). However, this was not the case in the current analysis and the results suggest homogeneity between the two populations. Many of the samples that are similar to one another, between the two populations, are separated by great amounts of time (e.g. Kerma and Badari). These similarities over time make sense because, as Konigsberg (1990) asserted, as time elapses, related groups become more genetically similar. In order to explicate the meaning behind all of these findings, the results here must be tempered by the DNA evidence. Both mtDNA (Krings et al., 1999) and Y-Chromosome data (Hassan et al., 2008; Keita, 2005; Lucotte and Mercier, 2003) indicate that migrations, usually bidirectional, occurred along the Nile. Thus, the osteological material used in this analysis also supports the DNA evidence.
Interpretation of the results framed by several of the groups’ histories helps to elucidate the subtle relationships depicted in the PCO scatter plot. The predynastic sample from Badari occupies a complex position in Egyptian history. The Badarians are Egypt’s oldest agriculturalists and produced some of the earliest known pottery (Hassan, 1986) that predated state formation in Egypt. Badarian crania, in comparison to dynastic groups, are slight and less robust than their later counterparts (Angel, 1972; Morant, 1935; Stoessiger, 1927). Stoessiger (1927) likened the gracile nature of the Badarians to the gracile nature of the people from Naqada, but she pointed out that the Badarians are more prognathic. On this basis, many have postulated that the Badarians are relatives to South African populations (Morant, 1935 G. Morant, A study of predynastic Egyptian skulls from Badari based on measurements taken by Miss BN Stoessiger and Professor DE Derry, Biometrika 27 (1935), pp. 293–309.Morant, 1935; Mukherjee et al., 1955; Irish and Konigsberg, 2007). The archaeological evidence points to this relationship as well. (Hassan, 1986) and (Hassan, 1988) noted similarities between Badarian pottery and the Neolithic Khartoum type, indicating an archaeological affinity among Badarians and Africans from more southern regions. Furthermore, like the Badarians, Naqada has also been classified with other African groups, namely the Teita (Crichton, 1996; Keita, 1990), while the Gizeh sample clustered with the Maghreb and Sedment (Dynasty IX Egyptians) (Keita, 1990).
Nutter (1958) noted affinities between the Badarian and Naqada samples, a feature that Strouhal (1971) attributed to their skulls possessing “Negroid” traits. Keita (1992), using craniometrics, discovered that the Badarian series is distinctly different from the later Egyptian series, a conclusion that is mostly confirmed here. In the current analysis, the Badari sample more closely clusters with the Naqada sample and the Kerma sample. However, it also groups with the later pooled sample from Dynasties XVIII–XXV. The unusual grouping of Badari, Naqada, Kerma, and the later Dynastic pooled sample may have been a product of the mixed nature of the pooled sample. The effects of pooled samples have been demonstrated in Nubians by obscuring relationships and creating a falsely close affinity between it and the samples it clusters with (Godde, 2009a). Moreover, affinities among the Badarian, Naqada, and Kerma samples have been revealed by other authors (Keita, 1990; Nutter, 1958) and it is no surprise that this relationship exists in the data here.
Relationships among Badari, Naqada, and Kerma have not always been overt in the skeletal data. Berry et al. (1967) concluded from their nonmetric analysis that their Badarian sample differed significantly from Naqada and Kerma, but was closely related to the Gizeh sample. Their study included the same samples as this analysis, but yielded results that are different from the current study and the craniometric research. Berry et al. (1967) employed a completely different range of statistics, which may account for the difference between the two conclusions. However, Berry and her coauthors also noted homogeneity across all the Egyptian groups, including Naqada and those that pre- and post-date the sample. This is indeed the case here, as is evidenced in the PCO plot; the Egyptians appear to be relatively homogeneously grouped. Some Badarian crania also classified well with the Gizeh sample (Keita, 1990).
The close clustering of Badari and Naqada with Kerma exemplifies the possible relationship of Nubians to Egyptians. Originally, the Nubian A-Group was thought to be Badarian in origin (Reisner, 1910). However, later work (Adams, 1977; Godde, 2009a) established that the A-Group were actually Nubian. Comparisons of C-Group and Pan-Grave Nubians to Badari and Hierakonpolis separate Badari from the other samples, indicating no biological affinities with these earlier Nubian groups (Godde, 2009b). The reoccurring notation of Kerma affinities with Egyptian groups is not entirely surprising. Kerma was an integral part of the trade between Egypt and Nubia. Collett (1933) concluded that Kerma was originally inhabited by Egyptians with neighboring Nubian settlements. Her investigation of the site pointed towards continuous Egyptian occupation of some sort at the site throughout the Kerma time period. This continued presence at Kerma is an optimal condition for gene flow to occur between the two populations.
Nubian groups have also been scrutinized as to their relationship with other Nubians. Both the Meroitic and X-Group were originally postulated to be foreign peoples migrating into Lower Nubia (Adams, 1968; Nielsen, 1970). These ideas were based on changes in pottery around the beginning of each of the respective time periods. However, the archaeological evidence actually showed slow change in form over time (Adams, 1977) and the biological evidence demonstrated a similar trend in the skeletal data (e.g. Godde, in press; Van Gerven et al., 1977). These conclusions negate the possibility of invasion or migration causing the shifts in time periods. The results in this study are consistent with prior work; the Meroites and X-Group cluster with the remaining Nubian population and are not differentiated.
Despite the biological similarities between the two populations, the Nubians appear relatively homogeneous. The homogeneity is consistent with Carlson and Van Gerven’s (1979) in situ hypothesis, but contradicts the findings of Buzon (2006). Buzon (2006) found a high level of heterogeneity in the Nubian samples she examined, including individuals from Kerma and the C-Group. Moreover, the Egyptian samples in her study were homogeneous overall, consistent with Berry et al. (1967) and the results in this paper. However, the levels of homogeneity appear to be similar within Nubians and within Egyptians in this study. The differences between this research and Buzon’s (2006) work may be related to the statistics used. Buzon’s (2006) goal was not to look at biological affinities; rather, she was trying to establish identity among her individuals by associating it with archaeological material. While this paper used a biological distance approach to investigate past population relationships, her paper used factor analysis, principal components, and a least squares regression. Although these (hers and those used here) statistics all have a solid methodological basis, they measure population relationships in two different manners and the results between them are not entirely comparable.
Gene flow may account for the homogeneity across these Nubian and Egyptian groups and is consistent with the biological diffusion precept. Small geographic distances between groups allow for the exchange of genes. One of the Nubian groups in this analysis is located in Upper Egypt (Hesa/Biga), near Egyptian occupation, and contact between the two populations may have been commonplace. Specifically, Nubians were often captured and enslaved by Egyptians to build pyramids, or employed by the Egyptian army (Trigger, 1976). Occasionally, Nubians were even directed to fight other Nubians as part of their duties as troops (Trigger, 1976). Moreover, some groups of Nubians allied with the Egyptians for the conquest of Nubian areas, primarily during Dynasty I (Trigger, 1976). Furthermore, as mentioned earlier, trade between Nubians and Egyptians flourished at Kerma and Meroe, during the time periods named after the sites, and enabled contact for potential gene flow. As a result of their respective histories, the multitude of interactions between them, geographic locations, and their biological composition, it appears that gene flow was possibly occurring between the two populations.
The similarities uncovered by this study may be explained by another force, adaptation. As stated above, the results appear to support the biological diffusion hypothesis because the Nubian and Egyptian groups are biologically similar. However, this resemblance may be indicative of a common adaptation to a similar geographic location, rather than gene flow. Carlson and Van Gerven (1979) stated this idea in reference to common adaptations of Nubian, Paleolithic, and aboriginal Australian populations. Additionally, Carlson (1976), Prowse and Lovell (1995), Van Gerven (1982), and Van Gerven et al., 1977 D. Van Gerven, G. Armelagos and A. Rohr, Continuity and change in cranial morphology of three Nubian archaeological populations, Man 2 (1977), pp. 270–277. View Record in Scopus | Cited By in Scopus (9)Van Gerven et al. (1977) also recognized this form of natural selection as a mechanism for in situ biological change; Egypt and Nubia have similar terrain and climate. Because of the similarity between and the overlapping of the two territories that would require similar adaptations to the environment, common adaptation cannot be discounted.
Sample size may have unduly influenced the results in this analysis. Four of the samples were represented by less than 30 individuals, while several of the remaining samples numbered close to 200 individuals. Moreover, only a small number of groups (six) from each population were examined in this study. Observations of more and larger population samples may produce different findings.
In summation, a portion of the in situ hypothesis in Nubians is supported in this paper, namely homogeneity. Gene flow appears likely between the Egyptians and Nubians, although common adaptations to a similar environment may have also been a factor in their cranial similarities. This study does not rule out the possibility that in situ biological evolution occurred at other times not represented by the samples in this analysis. Further research should incorporate more populations the Nubians were in contact with, to further shed light on Nubian population structure. Additionally, Konigsberg’s (1990) spatial–temporal isolation model should be applied to the dataset here to further explicate the results.
One for the records, as I can’t locate the full text at the moment. Not unsurprising to see some relationship between Nubians and Egyptians as they were right next door and the Badari and Nubians appeared to be closely related before state formation brought a lot of Lower Egyptians southwards. From other studies I’m guessing the relationship is due to the pre-dynastic Nubian/Badarian relationship. I’m amused to see this on Egyptsearch as ‘proof’ the Egyptians were black. I’m guessing they didn’t read it through- but thanks for posting it guys. As always forgetting that Nubians actually had about 60% Eurasian ancestry, same as the modern, so thinking this is proof of a ‘black’ Egypt is amusing. Still no response from them as to how Egyptians managed to swap race when their Y chromosomes show that only about 5% overall Eurasian historic immigration is possible. Hmm.
Still having so issues with my MS at the moment, but I’m slowly getting back to myself. I will respond to all comments eventually. Patience.