Variation in porotic hyperostosis in the Royal Cemetery complex at Abydos, Upper Egypt: a social interpretation.
This paper presents the results of a palaeopathological study of skeletal remains from the high status cemetery of Abydos in Upper Egypt associated with the rulers of the First Dynasty (Petrie 1900). The crania from two groups of spatially-distinct subsidiary burials were examined, associated with the First Dynasty kings Djer and Diet (Uadji), the second and third, or third and fourth kings of the dynasty, depending on who is regarded as the first ruler: Narmer or Aha (Wilkinson 1999). These burials are thought to have taken place near the time of the kings' interments. The investigation focused on porous defects, broadly called porotic hyperostosis, which were observed in the crania of the buried individuals, and the variations in this lesion were used to explore possible social differences. There is a relationship between social status and health that is readily appreciated, since the former determines access to nutrients, the overall environmental quality, and the ability to provide and receive medical care (Crooks 1995).
The sample population
All the rulers of Dynasty I (Kemp 1966, 1967) as well as predynastic kings and elites were interred at Abydos, a holy place perhaps regarded as an ancestral home, as opposed to the capital at Memphis, in northern Egypt (Spencer 1993; Wilkinson 1999). The main part of the cemetery, called the Umm El Qaab, contains the actual tombs of the rulers, although none of their physical remains have survived. In the same place, however, are subsidiary burials, frequently marked by stelae, which are regarded as those of court functionaries and members of the rulers' personal entourage (Petrie 1900; Emery 1961; Kemp 1966; Trigger 1983; Spencer 1993). The idea that they were simply servants or slaves has been contested (Thomson & Randall-MacIver 1905, Petrie 1925, Bestock 2002). Approximately one mile away in north Abydos are another set of subsidiary burials arranged roughly in squares (Petrie 1925) which outline the remains of structures variously called 'funerary palaces' or funerary enclosures (Kemp 1967; O'Connor 1989). These graves can be identified with the reigns of particular rulers by associated artefacts (Petrie 1925). The associated grave occupants were apparently a mixture of artisans and lower court officials such as a 'seal bearer'. Examples taken from skeletal material of both these groups of burials formed the subject of this investigation.
'Porotic hyperostosis' is generally used to describe defects in the outer table of the cranial vault of the skull, associated with a widened diploic space, and has varying degrees of severity (Angel 1964; Ortner & Putschar 1981; Goodman et al. 1984; Larsen 1997). The standard depictions of porotic hyperostosis show gross large bilateral vault porous defects sometimes associated with new bone formation in notably thickened parietal bones, and usually in children, but this is only the most severe manifestation. Lesions in the orbital roof(s) are usually called cribra orbitalia. The classic defects, whether in vaults or orbits, are hypothesised to be caused by a compensatory expansion of the bone marrow with resultant pressure induced alterations in the outer skull bones (Larsen 1997). Other bones can be affected. The expansion of bone marrow reflects the increased production of red blood cells in response to physiological states that have caused either decreased longevity of red blood cells and/or altered iron metabolism and deficiency. Clear examples of such states are thalassemia major and sickle cell anaemia where clinical data support the correlation of bony pathology with anaemia. The marrow response helps to counter the effects of anaemia. Sickle cell anaemia has been firmly diagnosed by bio-molecular means in predynastic sub-adult Egyptians whose remains exhibited skeletal changes known to be associated with haemolytic anaemias (Marin et al. 1999; Cerutti pers. comm.). Unfortunately it is rare to get this kind of laboratory confirmation of a skeletal diagnosis.
The cause of the lesions has usually been thought to be due to dietary iron deficiency (Sanford et al. 1983; Stuart-MacAdam 1985, 1992; Larsen 1997). Experimental data provide evidence for a more important role for whole blood loss. Chronically bled rats show a much greater marrow response than those given an iron-deficient diet (Burkhard et al. 2001). Extrapolating to the human case, this suggests that the chronic blood loss frequently associated with parasitic diseases such as schistosomiasis and hookworm (Stephenson & Holland 1987; Tanaka 1989), could cause compensatory marrow expansion and resultant porotic hyperostosis. Theoretically, iron loss from diarrhoea, also found associated with various infectious diseases, could trigger a marrow response. Reflex mechanisms that reduce serum iron in some bacterial infections could also be a factor (see Kent & Weinburg 1989; Stuart-MacAdam 1992). The clinical severity of anaemia has to be considered in the contexts of crowding, poor general nutrition, weaning stress and overall disease burden, including chronic infections (Kent 1986; Palkovich 1987) and the handling of sanitation (see e.g. Dixon 1972).
Other defects and causes have to be considered in the evaluation of vault porosis. Localised lesions confined to the periosteum may indicate primary periostitis, and are usually related to trauma or proximal scalp infections. If symmetrically widespread and restricted to the periosteum, lesions may indicate vitamin C deficiency or secondary periostitis, the latter being understood to reflect a general inflammatory response to a non-local process (Ortner & Putschar 1981); however, poor vitamin C intake is less well documented as a probable cause. In the Nile Valley, secondary periostitis could be a part of the response to parasitic infection, given the prevalence of schistosomiasis and hookworm (Stephenson & Holland 1987; Tanaka 1989), even if the infection is not severe enough to provoke grossly detectable diploic expansion.
Vault porosis thought to be related to iron deficiency including blood loss is hypothesised to manifest only in childhood and not to occur in individuals who first become anaemic as adults (Stuart-MacAdam 1985), based on cross-sectional studies. It is known that skeletal lesions do not occur or are not severe in all anaemic children, even those with sickle cell disease who have haemolytic crises. It is assumed that quantitative and qualitative variations in adult skulls reflect the relative differences that would have been found in childhood; the simplifying assumption is that there are no great differences in remodelling rates between surviving groups. However, there is a caveat: the percentage of skulls with lesions could also be affected by group differences in pre-adult mortality, assuming that the samples under consideration are from distinct and 'real' populations (or strata). Such demographic information is not known for these crania, which in part are likely to consist of people from northern Egypt (see Keita 1992).
Lesions in early Egypt would most likely have been related to anaemia, either from parasitic, genetic and/or dietary related causes in the context of a challenging socio-ecological environment with a consequent biological challenge. The remains of adult individuals with clinically severe genetic anaemias are not likely to be recovered from ancient cemeteries in great numbers, because of general and specific mortality. Vitamin D deficiency can reasonably be excluded as a frequent cause because of the high solar radiation in the area, even if only to the face and hands. It is likely that blood loss from schistosomiasis and hookworm are primarily responsible. Given the concern here for examining the pattern of pathology in relationship to social position, a childhood aetiology of the lesions would be advantageous because it may reflect aspects of social origins. Early childhood lifestyle and life-history are more likely to reflect aspects of social life such as rank or status because of the vulnerability of children (Nestel 1990), and ties to their parents' social position. Social status in adulthood may not be the same as that held in early life, even in ancient monarchical societies.
This study thus offers the opportunity to examine lesions believed to reflect a health setback that occurred during childhood, in two groups of adults with perhaps broadly different roles at the royal court.
The material studied consists of all existing adult crania from the burials around the tomb of King Djer primarily, and from the lower cemeteries (funerary enclosure sites, Abydos north) of Kings Djer and Diet as far as can be ascertained. Djer's tomb was flanked by 318 subsidiary graves, and his funerary enclosure by 269. King D jet's retainer burials numbered 174 and 161 respectively. Sex was determined by standard anatomical criteria (Brothwell 1981), and the consideration of the names of individuals when known; no pelves were available. It is suggested that the royal tomb sample has 27 males and 17 females (44 total), and the funerary enclosure tombs have 38 males and 10 females (48 total). These are the only known extant remains available for study and they represent only a fraction of court functionaries, the 'populations' actually interred (see Petrie 1925). Analysis of variance (ANOVA) revealed no difference between the sexes. However, sex sub-samples would have been combined in any case to obtain a group impression, because the lesions do not connote diseases that are genetically sex-linked. The royal tomb material is stored at the Natural History Museum in London. Cambridge University houses the funerary enclosure/palace sample.
Due to various kinds of damage not all structures of interest were present or complete in each cranium. The sample sizes by structure are shown in Table 1. The second grouping of vaults refers to those having all three bones complete or nearly so.
The standard method of approach is based on macroscopic observation with the naked eye. For this study poroses were assessed in two ways. Vault, but not orbital, poroses were graded from 0-6 following Hillson (1978) who used the term porosities to describe the defects. The grade is used as a variable called the vault porosity or porosis score (VPS) and noted as below:
1--scattered fine porosis,
3--some linked porosis,
4--'canal-like' linked porosis,
5--small trabecular outgrowths from outer table,
6--marked trabecular structure on the outer table.
Lesions were mapped for each vault onto a scheme that included the coronal, sagittal and bregma sutures. VPS was recorded for the most anatomically severe lesion wherever found, which was usually the parietals. The number of kinds of superior vault bones affected was recorded as the extent score (ES), range 0-3. (Parietal involvement was only counted once.) Two dichotomous variables were devised. The first designates vaults with any lesion porosis scores of one or more and is called VPP1. The second enumerates vaults with lesions of porosis scores of two and greater (VPP2), and is of more interest in this study, because previous experience has shown that this variable was more often associated with some parietal thickening. VPP2 serves as a kind of screen and facilitates the recording of lesions indicative of more than periostitis. Cribra orbitalia lesions were scored as present or absent.
Parametric ANOVA, Mann-Whitney and contingency table analyses were carried out. The five per cent probability level was chosen for significance.
A range of lesion quality was observed. Noticeable, although not necessarily severe, parietal thickening was usually observed with the higher grade lesions; thus the second variable (VPP2) tends to restrict the count to those vaults having lesions closer to 'porotic hyperostosis' as originally described (see Angel 1964; Ormer & Putschar 1981), although mild. Crania with the extreme classic lesions were not found, an observation made about other Egyptian material by Hillson (personal communication). Some crania had what might be described as pitting, not porosis, perhaps representing remodelling or other tissue activity. The royal tomb and funerary palace samples have lesion frequencies above ten per cent in all dichotomous variable categories (Table 2). Lesions with a porosis score of one were not generally associated with visible thickening and could be due to periostitis. The level of porosis-defined lesions is noteworthy in the two samples. The frequency of individuals having any degree or kind of vault lesion is greater than 45 per cent.
Contingency table analyses indicate significant inter-group frequency differences (Table 2). The royal tomb group is observed to have three times more affected individuals than the funerary palace sample for cribra and higher grade vault lesions (VPP2). Individuals recovered from around the royal tombs have one and a half times the observed lesions (VPP1), found in the funerary palace folk. The percentage of crania having a vault porosis score of one, obtained by subtracting VPP2 from VPP1, is nearly the same for both groups, approximately 30 per cent. This suggests a general background of biological challenge resulting in lesions largely restricted to the periosteum. The average porosis score and extent of porosis lesions per individual for the royal tomb vaults, is twice that of the funerary palaces (Table 3). Although the standard deviations indicate high within-sample variability, the central tendency differences are statistically significant.
The results of the analysis are consistent in suggesting that the individuals recovered from around the funerary palaces--as opposed to those from near the royal tombs--experienced less physiological challenge in childhood. Fewer individuals were affected, and those who were had less anatomically extensive lesions. Although both series evince noteworthy levels of pathology, the royal tomb sample has more severe lesions and a higher frequency of affected individuals with a vault porosis score of two or more. Strictly speaking it can only be said that a group difference in vault porosis frequency and severity exists in the recovered material. Therefore any more interpretive scenarios are speculative.
As a starting point, and tentatively accepting Stuart-MacAdam's (1985) and others' interpretation of pre-adult genesis, it is worth noting that there is a difference in the two adult groups for a lesion hypothesised to occur in childhood. In a hypothetical society with high social mobility, one could reasonably expect no difference between adult groups for a childhood lesion. In one without such equality such differences might be expected to 'cluster' in adulthood groups, assuming that the social position of children and adults was somehow connected. If poroses associated with diploic expansion do represent a childhood lesion then they can be seen as an osteobiographical tool that allows the social 'tracking' of individuals or groups in some selected circumstances.
Does the inter-group difference itself imply that those from the two sub-sites are true samples from distinct 'populations' within which individuals are connected by a 'principle' like class, caste or ethnicity? Or is it the result of sampling accident? This is a relevant theoretical and statistical question, with implications for other research in similar circumstances. Stated another way, do the recovered site-differentiated individuals represent true samples, either random or non-random, of actually existing entities ('populations'), or are they non-statistical fragments of the general population? There is no evidence that the crania preserved by the excavators represent a specially selected subset of the burials, so the surviving groups can reasonably be treated as representative.
The reason for the disparity between the 'groups' may reflect institutionalised social structure, in which those buried in the funerary enclosure had greater environmental advantages than those buried with the king. However, there is little corroboration of such a model from current documentation. There seem to be no later Egyptian references to internal social ranking in this early period, or to the custom of having current court functionaries buried at the time of the rulers (Baines personal communication). The results of studies of later dynastic Egyptian society are consistent with there being four broad strata (Trigger 1983). By extrapolation the Early Dynastic individuals interred around the royal tombs and funerary palaces would have ranked below the ruler and royal family, nobles and high officials, and just above the 'peasants'. Grave goods and stelae have been interpreted as marking the royal tomb occupants as having primarily been the rulers' personal retinue, and largely of servile status. The funerary enclosure group was composed apparently of minor officials and artisans as indicated by titles, and the presence of fine copper tools (Petrie 1925). However, the interpretation of social status from funerary remains is always difficult (Ucko 1969).
Several biosocial models are consistent with the findings. If the funerary enclosure group consists of individuals from a strata or 'population' with higher ascribed status, then the results fit current normative expectations, namely that better health corresponds with higher social status, all other things being equal. These individuals in childhood would have theoretically been at less risk from disease, especially for chronic conditions related to nutrition. By contrast, the results from the royal tomb group correlate with their interpretation as servants, which would in turn imply that court 'servants', companions of the king, were recruited from the 'poor'. A 'class' explanation works no matter when the lesions or their aetiology(ies) appear in life, and only requires that the defects represent evidence of a pathological condition. For example if the lesions were acquired in adult life, then it would suggest that those buried around the king did work, or were fed in a manner, which placed them at greater risk.
If the origins of the royal tomb individuals were from a group of ascribed higher status then the results would require a more complex explanation. One of these would be that the higher status permitted the parents to invest more in sick children; thus more of them survived to adulthood, if the diseases in question were life-threatening. A higher lesion frequency in an adult cohort in this instance would indicate a higher survival of challenged individuals who are missing from the other group, in which 'better health' is actually an illusion. This is a version of what has been called the osteological paradox (Wood et al. 1992).
The concept of status may need rethinking for Early Dynastic Egypt. For example, another interpretation would employ the concept of phyles described from ancient Egypt (Roth 1991), and hereditary occupational castes linked to kings or elites, the latter even known from more recent societies (see Levtzion 1973, and Tamari 1991). The evidence suggests that phyles were clan-like hereditary-based work associations associated with kingship (but not exclusively), and may have developed from predynastic totemic clans. Symbolically, in the Early Dynastic world it is conceivable that phyles or phyle-associated groups would have been expected to assist the king in the afterlife. The royal tomb and funerary enclosure groups may be of the same social rank, but horizontally differentiated by the professions of the households into which they were born or recruited as young children, thereby participating in the work associated with these domiciles or homesteads.
Young artisans (and other non-farmers) would have been less exposed to parasitic infections and their nutritional sequellae than agriculturalists or shore fishermen exposed to Nile alluvium. There is positive evidence for childhood occupation differences; the footprints of children approximately six years of age have been preserved in plaster at some elite Early Dynastic building sites (Baines personal communication), and children are seen in paintings of farming activities. An occupational model would explain the group difference if the royal tomb folk were drawn from farmers, and the funerary enclosure sample connoted artisans or at least non-farmers, both perhaps connected to the kingship, assuming that the lesions are of childhood origin. Differences would stem from occupation irrespective of class status. It is not unreasonable to postulate that there may have been 'royal farmers' (as there were artisans), and that as children, some of the adults in the household helped in food production, thus sustaining a higher exposure to parasites, than the children of non-farmers. While there is no described direct evidence that the samples in this study were from phyles, the individuals represented did perform broadly different kinds of work. Although it is known that phyles in the Old Kingdom rotated through different kinds of work, they may have been initially more confined to one occupation.
Another possibility is that the royal tomb people comprised a hereditary companion servant caste, fed from infancy a ritually prescribed, but nutritionally deficient diet, and perhaps secluded from the sun, in a kind of purdah. However, there is no known documentary evidence for such a practice, or reference to it in later texts (Baines & Roth personal communication). Also, as mentioned, there are no long bones to assess for rickets.
The occupational model receives additional but indirect support from an analysis of linear hypoplasias on the first and second molars. One would expect markers of severe physiological stress to track with each other. However, there is no statistical difference between the royal tomb and funerary enclosure samples for this lesion at any level. The frequency of individuals having first molar defects was 35.0 and 36.4 per cent respectively [n = 22 (RT) and 40 (FE); Chi square = 0.012; p = 0.914); and for the second molar 42.1 and 41.7 per cent [n = 24 (RT) and 38 (FE); Chi square = 0.001; p = 0.973]. The average number of lesions for each tooth in the two groups was also statistically insignificant (p > 0.05). These findings would seem to show that the group contrast in porotic hyperostosis frequencies does not reflect a difference in overall (general) health. Otherwise it could reasonably be expected that the royal tomb sample would have had more linear hypoplasias; in fact if true this would lend somewhat more support to an explanation of a status difference in origins, since a general and specific indicator of childhood stress would both have high values. Instead a difference is only seen for the variable that can most easily be associated with a particular lifestyle/ecology and disease risk.
The recovery of more data from predynastic and dynastic contexts may provide information with which to make better social inferences about these findings. However, based on a synthesis of the available evidence, some variant of the model incorporating horizontal occupational differentiation, with or without the concept of hereditary caste, would seem to be more convincing in explaining the differences than a simple hierarchical class model, accepting the hypothesis that the lesions are of childhood origin.
We thank Dr C. Stringer, Dr R. Foley and M. Bellati, for permission to examine collections in their care. Professor J. Baines, Professor G. Armelagos, Dr A. Roth and Dr S. Hillson made comments in various discussions useful to this research. This work was made possible in part by grants from the Griffiths and Boise funds of Oxford University. This piece is dedicated to John Baines, Professor of Egyptology, Oxford University, and the late Mamdouh Sharara, University of Maryland Statistical Laboratory.
Received: 2 February 2004: Accepted: 10 November 2004: Revised: 11 November 2004
ANGEL, J.L. 1964. Osteoporosis: thalassemia. American Journal of Physical Anthropology 22: 369-74.
BESTOCK, L. 2002. The ideological significance and societal complexity of the subsidiary graves for royal and elite monuments of the Early Dynastic period. Paper presented at the Annual Meeting of the American Research Center in Egypt, April 26-28, John Hopkins University, Baltimore (MD).
BROTHWELL, D.L. 1981. Digging up bones. Ithaca, NY: Cornell University Press.
BURKHARD, M.J., D.E. BROWN, J.P. MCGRATH, V.P. MEADER, D.A. MAYLE, M.J. KEATON, W.P. HOFFMAN, J.L. ZIMMERMAN, D.L. ABBOTT & S. SUN. 2001. Evaluation of erythroid regenerative response in two different models of experimentally induced iron deficiency anaemia. Veterinary Clinical Pathology 30 (2): 76-85.
CROOKS, D. 1995. American children at risk: poverty and its consequences for children's health, growth and school achievement. Yearbook of Physical Anthropology 38: 57-86.
DIXON, D.M. 1972. The disposal of certain personal, household and town waste in ancient Egypt, in P.J. Ucko, R. Tringham & G.W. Dimbleby (cd.) Man, settlement and urbanism: 647-50. London: Duckworth.
EMERY, W.B. 1961. Archaic Egypt. London: Penguin.
GOODMAN, A., D.L. MARTIN, G. ARMELAGOS & G. CLARK. 1984. Indications of stress from bone and teeth, in G. Armelagos & R. Cohen (cd.) Palaeopathology at the origins of agriculture. 13-50. New York: Academic.
HILLSON, S. 1978. Human biological variation in the Nile valley, in relation to environmental factors. PhD Thesis, University of London.
KEITA, S.O.Y. 1992. Further studies of crania from ancient northern Africa: an analysis of crania from first dynasty Egyptian tombs, using discriminant functions. American Journal of Physical Anthropology 87: 245-54.
KEMP, B.J. 1966. Abydos and the royal tombs of the First Dynasty. Journal of Egyptian Archaeology 52: 13-22.
--1967. The Egyptian 1st Dynasty royal cemetery. Antiquity 41: 22-31.
KENT, S. 1986. The influence of sedentism and aggregation on porotic hyperostosis and anemia. Man 21: 605-36.
KENT, S. & E. WEINBURG. 1989. Hypoferraemia: adaptation to disease. New England Journal of Medicine 320: 672.
LARSEN, C.S. 1997. Bioarchaeology: interpreting behavior from the human skeleton. New York: Cambridge University Press.
LEVTZION, N. 1973. Ancient Ghana and Mali. New York: Methuen.
MARIN, A., N. CERUTTI & E.R. MASSA. 1999. Use of the amplification refractory mutation system (ARMS) in the study of HbS in predynastic Egyptian remains. Bollettino della Societa Italiana di Biologia Sperimentale 75 (5-6): 27-30.
NESTEL, P.S. 1990. Nutritional vulnerability of the child, in G.A. Harrison & J.C. Waterlow (cd.) Diet and disease in traditional and developing societies: 185-208. Cambridge: Cambridge University Press.
O'CONNOR, D. 1989. New funerary enclosures (Talbezirke) of the Early Dynastic Period at Abydos. Journal of the American Research Centre in Egypt 25: 51-86.
ORTNER, D. & W. PUTSCHAR. 1981. Identification of pathological conditions in human skeletal remains. Washington, DC: Smithsonian Institute.
PALKOVICH, A.M. 1987. Endemic disease patterns in palaeopathology: porotic hyperostosis. American Journal of Physical Anthropology 74: 527-37.
PETRIE, W.M.F. 1900. The royal tombs of the First Dynasty. Part I. Eighteenth Memoir. London: Egypt Exploration Fund.
--1925. Tombs of the Courtiers and Oxyrhynkhos. London: Egypt Exploration Fund.
ROTH, A.M. 1991. Phyles in the Old Kingdom: the evolution of a system of social organization. Chicago: Oriental Institute, University of Chicago.
SANFORD, M.K., D.P. VAN GERVEN & R.R. MEGLEN. 1983. Elemental hair analysis: new evidence on the etiology of cribra orbitalia in Sudanese Nubia. Human Biology 55: 831-44.
SPENCER, A.M. 1993. Early Egypt. London: British Museum Press.
STEPHENSON, L.S. & C. HOLLAND. 1987. The impact of Helminth infections on human nutrition: 57-152. New York: Academic.
STUART-MACADAM, P. 1985. Porotic hyperostosis: representative of a childhood condition. American Journal of Physical Anthropology 66: 391-98.
--1992. Porotic hyperostosis: a new perspective. American Journal of Physical Anthropology 87: 47.
TAMARI, T. 1991. The development of caste systems in West Africa. Journal of African History 32: 221-50.
TANAKA, H. 1989. Hookworm disease, in R. Goldsmith & D. Heyneman (ed.) Tropical medicine and parasitology: 361-68 Norwalk: Appleton and Lange.
THOMSON, A. & D. RANDALL-MACIVER. 1905. Ancient races of the Thebaid. Oxford: Oxford University Press.
TRIGGER, B. 1983. The rise of Egyptian civilization, in B. Trigger et al. (ed.) Ancient Egypt, a social history: 1-70. Cambridge: Cambridge University Press.
UCKO, P. 1969. Ethnography and archaeological interpretation of funerary remains. World Archaeology 1: 262-80.
WILKINSON, T. 1999. Early Dynastic Egypt. New York: Routledge.
WOOD, J.W., G.R. MILNER, H.C. HARPENDING & K.M. WEISS. 1992. The osteological paradox: problems of inferring prehistoric health from skeletal samples. Current Anthropology 33 (4): 343-70.
S.O.Y. Keita (1) & A.J. Boyce (2)
(1) National Human Genome Center, Howard University and Department of Anthropology, Smithsonian Institution, USA
(2) Institute of Biological Anthropology and St Johns College, University of Oxford, UK
Table 1. Sample sizes by structure Royal tomb Funerary enclosure/ Palace Orbits (roofs) 37 45 Vaults (all) 44 48 Vaults (most complete) 37 42 Table 2. Dichotomous variables, comparisons Frequencies (%) Series N VPP1 VPP2 Cribra Royal tomb (44, 37) 77.3 47.7 35.1 Funerary enclosure (48, 45) 45.8 14.6 11.1 [chi square] 9.53 11.91 6.84 [rho] 0.002 <0.001 0.008 Table 3. Means and standard deviations, comparisons VpS ES N Mean SD N Mean SD Royal tomb 44 1.74 1.18 37 1.97 0.93 Funerary enclosure 48 0.74 0.76 42 0.98 0.92 Comparisons ANOVA [rho] <0.001 (F = 23.64) <0.001 (F = 22.82) Mann Whitney [rho] <0.0001 (Z = -3.85) <0.0001 (Z = -4.20)
|Printer friendly Cite/link Email Feedback|
|Author:||Keita, S.O.Y.; Boyce, A.J.|
|Date:||Mar 1, 2006|
|Previous Article:||Boat remains and maritime trade in the Persian Gulf during the sixth and fifth millennia BC.|
|Next Article:||Colonials, merchants and alabaster vases: the western Phoenician aristocracy.|