Complexity in burial practices implies social differentiation, signalled to an archaeologist by variations in funerary structures or grave goods (e.g. Gualtieri and Jackes, 1993; Rothschild, 1979). To a skeletal biologist complexity may imply differentiation by age, sex, state of health, family relationships. If a mortuary practice allows us to distinguish groupings among skeletons on these bases, then the burial practices are complex, even though grave goods or burial type may not be unequivocally differentiated.

The next step in interpretation might be that a society with differentiated burial practices is more "complex" than a society with simple undifferentiated burials. On the evidence to be presented, this step is unwarranted. Two very similar societies may have very different burial practices, one apparently simple, one differentiated. Although the two societies may apparently have different cultural emphases expressed through their burial practices, the two social structures may be similar.

Such a circumstance occurs in southern Ontario with two Iroquoian societies, the Huron and the Neutral, during the first half of the seventeenth century. The two occupied contiguous areas and their cultures were broadly similar, despite environmental and economic differences which must have become more acute as the Huron gathered into Simcoe County to take advantage of trade with the French, while Neutral trade to the south remained important (Jackes, 1988:109-110, 117-118,143; Jamieson, 1992:79). Although there have been suggestions that important differences in government had appeared by 1626 (Noble, 1984, 1985 but cf. Trigger, 1985: 221-224) , the French missionaries laid no emphasis on any but superficial differences. In 1626 Daillon reported of the Huron and the Neutral that "their manners and customs are quite the same" (LeClercq, 1881:271) and the Jesuits, writing of the period 1640-41, also regarded the Neutral as very similar to the Huron (Charlevoix, 1900 2:152). Both societies were destroyed by Iroquois raids from the south in 1649-50.

Sixteenth and seventeenth century Ontario is not generally considered to be an area of complex burial patterns because major emphasis has been given to the ossuary burials of the Huron. Ramsden (1990:381) has recently suggested that there has been "an over-emphasis on this burial mode" and others have drawn attention to extra-ossuary burial for individuals other than infants and captives (e.g. Knight and Melbye, 1983; Sutton, 1988). Nevertheless, it must be acknowledged that ossuary burial is the principal mode of interment for Huron of the the sixteenth and seventeenth centuries and, based on the Maurice (Jerkic, 1975:48-55), Kleinburg (Jackes, 1977:8-10) and Uxbridge (Pfeiffer, 1983) Ossuaries, we have a clear idea of this predominant Huron burial custom, as well as a partial record of the variations in practice which occurred at each site. Moreover, we have ethnohistorical evidence, and Huron Feasts of the Dead are known in some detail, since one such burial ceremony was attended by Jesuit Fathers in May 1636, at Ossossané, the capital of the Attignawantan Nation of the Huron. The Jesuit description, together with excavations of a number of comparable burial sites dating from AD 1500 to the 1630-40s, provide us with relatively full knowledge of the large burial pit into which the defleshed and bundled bones of the dead were dropped from a scaffold and subsequently stirred around (see Kidd, 1953).

An ossuary is no more than a repository for bone, any place where the bones of the dead are kept. Masset (1993:102) mentions "un 'rite-ossuaire', dont l'expression archéologique aurait été l'amas d'ossements disloqués issu d'inhumations secondaires", and goes on to discuss the possibility that the "disorder" of the bones is in the mind of the archaeologist because of techniques of excavation or lack of knowledge of human osteology. European ossuary caves may, indeed, have very careful bone placement, though without the retention of individual identities (Zilhão, 1984). "Ossuary" has a definite meaning for North American anthropologists, also relating to multiple secondary burial, but the word is not well defined. I propose that ossuary be defined as a multiple burial in which most individuals are interred after natural or artifical disarticulation. Bones may be arranged by skeletal element, but are rarely retained in bundles containing recognizable individuals. In the case of sedentary or semi-sedentary populations having a well-delimited area in which the majority of individuals are buried, I would suggest that fewer than 25% of the calculated minimum number of individuals (MNI) buried within the cemetery area should be recognizable as individuals rather than simply from counts of skeletal elements.

The definition of a "recognizable individual" cannot be given in terms of the number of bones or articulations present: a number of major long bones in association would probably be the best indication of a placed bundle. Clearly associated innominates and long bones, or skull and long bones, or long bones and articulated vertebrae might be quite sufficient to constitute an individual. A number of long bones alone would be less certainly "an individual", but attribution to a single individual might be possible after an examination of indicators of , for example, age, sex., robusticity and proportions.

The need for a definition of ossuary burial arises because the Neutral Nation, while broadly similar to the Huron, apparently had different burial practices which could justifiably be called complex because of individuation and differentiation. Although large multiple burial pits which have been called ossuaries occur among the Neutral, the Neutral did not practice ossuary burial as defined above.

Huron ossuaries included all the dead (except infants and certain individuals who had died from violence, freezing or drowning) from a number of villages over a period which may have been up to 15 years, when a village was moved as soil fertility and convenient firewood decreased. Feasts of the Dead thus occurred rarely. The ceremonies lasted some days, accompanied by gift giving and dancing, and culminated in the deliberate mixing of the bones of the dead, possibly numbering up to 700 (Patterson [1986] gives an estimate above this for the Kleinburg Ossuary dated around AD 1580-1600; Katzenberg and White [1979] give an estimate close to this for Ossossané, dated AD 1636). We have no evidence for similarly large burial pits among the Neutral (Stothers, 1970).

Nearly thirty years ago, Marian White pleaded for a clearer understanding of the complexity of burials in southern Ontario and northern New York state: "...ossuary must be separated from other burial types like multiple bundles or multiple articulated individuals placed in a single pit" (1966:16). She mentions that the words used by early investigators ("bone pit", "ossuary", etc.) in fact influence the identification of a site as Neutral or non-Neutral. G.K.Wright (1963:66) implies that if a site on the Niagara Frontier has an ossuary then it is Neutral; no ossuary, not a Neutral site. White, on the contrary, reassessing the Orchid Site, after stressing the inadequacy of the data, goes on to distinguish four types of burial, two involving ossuary burial. It seems that by "ossuary" White meant "disarticulated individuals" and "thrown or placed bundles" (1966:19). However, earlier (1966:17) she mentions sites which have "true ossuaries because of the confusion of thrown bundles (my emphasis)". In order to clear up the contradiction here, a distinction should be made between: (1) burial pits (ossuaries) containing the bones of many individuals in which very few articulations or placed bundles are present; and (2) burial pits containing the bones of many individuals in which the majority retain articulations or the possibility of recognizing placed bundles which contain one or several recognizable individuals (multiple bundle burials).

Why bother about the distinction? I will argue that the cultural and social meaning of the two forms of burial is distinct.

Neutral burials

When Neutral burial complexity is mentioned in the literature, it is generally described in terms such as "ossuaries in which contents are vertically segregated by a clay partition or cap" (Jamieson, 1992:79). Noble (1968) described Neutral burials as double layered ossuaries with peripheral burials. This description derived partly from ten days of salvage work at the pre-Contact Orchid Site (White, 1966) and partly from Ridley (1961) who gathered together scattered information on seven Ontario Neutral sites. In only three cases is a false floor mentioned: in the case of the Hosken Site, by the farmer's son who destroyed the site (Ridley 1961:33); in the case of the Dwyer Site by the original excavator who mistook as exotic the drumlin clay into which the pit was dug (Ridley [1961:29,30] re-excavated and found no sign of a false floor); and the Walker Site where, in 1944, amateur excavators noted that a pit containing 11 adults and four children underlay a larger pit (Ridley 1961: 16,18). In 1974, further excavations at Walker showed no evidence of false floors (M.J.Wright, 1981:119), although a subfloor burial pit was also exposed. The 1974 sub-ossuary pit skeletons (six adults and six children) were described by Crerar (1983). It is likely that multiple and overlapping burial pits constitute the basis for the vertical segregation described. But it seems very possible that Neutral burial complexity goes far beyond a multiplicity of pits.

The Grimsby Neutral cemetery (AHGv-1) (Figure 1) provides a unique opportunity to understand Neutral burial practices: it was an almost undisturbed site dug in the late 1970s under difficult salvage conditions but with efforts at archaeological precision (W.Kenyon, 1982; Jackes, 1988). Many other Neutral sites have been unsystematically ravaged by "pot-hunters" since the 1820s.

Evidence for disruption of burial patterns following epidemics

However, the possibility that Grimsby is an atypical site, the result of the epidemics recorded by the Jesuit fathers, must be considered (Noble, 1985). Smallpox could have reached Neutral country by several routes in 1633-34, and a possible case of smallpox has been identified which might mark an even earlier epidemic (Jackes, 1983). Smallpox was certainly in Ontario by 1639. If Noble is correct in his opinion, we cannot use Grimsby as an exemplar of Neutral burial practices. The Grimsby cemetery is dated between AD 1615 and 1650 (I.Kenyon and Fox, 1982; Fitzgerald, 1983), and certainly encompasses the period of famine, warfare and epidemics which accompanied the introduction of intense trading with the French into Ontario and the introduction of firearms into New York State.

The 373 individuals identified in the Grimsby cemetery were distributed among 58 burial pits, containing from 1 to 103 individuals. The demographic breakdown of these burial features is diverse (Figure 2). Features 9, 26 and 1 were unusual, and these three burial features had more disarticulation and mixing than most others (Table 1). Feature 28 has a demographic profile similar to Feature 1 and probably comes from the same period. This burial was destroyed by vandals before it could be mapped or photographed (W.Kenyon, 1982:119). Could these atypical features reflect the famine and disease which struck the Neutral after 1630?

In Feature 9, 47% of the individuals were between the ages of 5 and 15 (cf. 16% for the cemetery as a whole). The smallpox most prevalent in that age group is the least lethal type (Dixon 1962:325) and that age group is the least likely to succumb to the after effects of epidemic disease. Feature 9 shares its general pattern of age and sex distribution with Features 11 and 45, although in a more marked form. Feature 45 was partially disturbed, but it seems to have had extreme disarticulation.

In the analyses of glass beads undertaken by I.Kenyon and Fox (1982) which may allow us to add time depth to our understanding of the Grimsby Cemetery, Features 9 and 11 cluster together (in loose association), quite separately from Feature 45. None of these three falls within the same bead cluster as Features 1 and 28.

Feature 26 also has much disarticulation, and 40% of the individuals were over 35 years, congruent with what we know of smallpox and measles epidemics in previously unexposed populations (see Jackes 1986:39; 1988:138), but it contained only shell beads and thus was not analyzed for period by I.Kenyon and Fox. Feature 1, on the other hand, contained six shell beads and 123 glass beads. We have no basis, therefore, for considering all the burial features with unusual demographic characteristics or a large amount of disarticulation and mingling of bones to be indicative of the late and hurried burial of victims of the final disintegration of the Neutral Nation. Feature 1 remains the only burial pit which is a good contender for this title (Jackes, 1983).

Feature 1 probably does reflect the disruption at the end of the final period before the last Iroquois assault on the Neutral. Here there is apparently less disarticulation than in Features 9 and 26, but the bundles are unusual. Some bundles were neatly placed, suggesting individuals: however, these could contain, for example, the bones of two left legs, but none from a right leg. W.Kenyon (1982:10) has described the confusing picture presented to the excavators. Feature 1 first seemed to be an ossuary, but below the scatter of disarticulated bones lay a series of tightly flexed bundles, none of which, however, contained a complete skeleton (see Jackes, 1983:76 for a description of one of the bundles). In fact, laboratory analysis showed that bone fragments from the bundles and from the overlying scatter could be reconstructed into the one skeletal element.

Feature 1 has a demographic pattern that is "normal", explicable in biological terms, and equivalent to the pattern for the cemetery as a whole (all features combined). This is an important reason for suggesting that Feature 1 may be the result of disruption. Feature 9, with its high proportion of juveniles all almost completely disarticulated, may on the other hand illustrate that disarticulation is characteristic of the burial pits in which many juveniles were buried (Features 9, 11 and 45: see Table 1 and Table 2).

Disarticulation may thus be characteristic of one sort of Neutral burial feature, but extreme care in the placement of bundles containing semi-articulated skeletons is also quite characteristic. This is exemplified especially by Feature 62 which is dated to the same, latest, bead period as Feature 1.

Feature 62 contained 103 people. Source population size cannot be calculated accurately for a cemetery, since that requires knowledge of life expectancy at birth (a value which cannot be derived from most archaeological cemeteries, [Jackes, 1992]); but estimates of deaths per year in a stationary or declining population would be under 50 per year for the Grimsby source village(s) (Jackes, 1986; 1988:128). On this estimate, both Feature 62 and Feature 9 would be comprised of: 1) the dead of several years with delayed burial, or 2) some of the dead of a number of villages, or 3) people who died of an epidemic disease. Features 62 and 9 present completely divergent demographic characteristics, and in each case cultural, not biological, factors seem to have determined interment. Thus we must exclude disease as the major determinant of burial patterns for any Grimsby burial features other than Feature 1 (and Feature 28?), and consider the first two possibilities above. In fact, it would be impossible to distinguish these two, unless the second possibility encompassed only a selected group of individuals, e.g. elderly males.

Arrangements of bones

Feature 62 is a magnificent grave, suggestive not of the disruption following upon an epidemic (accounts of epidemics in North America describe how the dead went unburied and eaten by dogs [e.g. Thwaites, 1898 16:219]) but of great attention to the bones of the dead and of delayed burial. This is in accord with the Jesuit Relations:

Our Hurons immediately after death carry the bodies to the burying ground and take them away from it only for the feast of the Dead. Those of the Neutral Nation carry the bodies to the burying ground only at the latest moment possible when decomposition has rendered them insupportable: for this reason, the dead bodies often remain during the entire winter in their cabins: and, having once put them outside upon a scaffold that they may decay, they take away the bones as soon as is possible, and expose them to view, arranged here and there in their cabins until the Feast of the Dead. These objects which they have before their eyes, renewing continually the feeling of their losses, cause them frequently to cry out and to make most lugubrious lamentations, the whole in song. But this is done only by the women (Thwaites, 1898 21:199).

The Jesuits saw the Huron Feast of the Dead at Ossossané in 1636, but they saw no Neutral feast. The Neutral feast cannot have been like that of the Huron: even in the quite large Neutral burial pits, the numbers of dead do not approach the numbers found in Huron ossuaries. The fifteenth century Glen Williams site (Neutral of the Milton Cluster) had a burial pit of 290 individuals with partial articulations and bundles, as well as smaller multiple and single burial features (Hartney, 1978). The largest known seventeenth century Neutral burial pit is one feature at Shaver Hill, containing a minimum of 163 people based on proximal femora (Stothers, 1971). A slightly larger oval pit at the Walker site (Ossuary C: M.J. Wright, 1981; Crerar, 1982) may well have contained more individuals, but nothing approaching the 450 to 680 known to be the minimum number of skeletons (MNI) for single sixteenth and seventeenth century ossuary pits at Uxbridge and Ossossané.

The burial pattern within Feature 62 at Grimsby (Figure 3) is very clear. The bones were laid more or less in an oval, "two curved and sloping banks" (Kenyon, 1982:193) of long bones, with crania generally placed along the central east-west passage. Stothers (1971) describes the main burial feature at Shaver Hill as having the skulls and long bones placed on one side, other adult skeletal elements on the other side, and most subadult postcranials in one discrete area. The oval burial arrangement may well find echoes in other sites: Shaver Hill, Walker and Dwyer - the early descriptions of the post-Contact Dwyer Site speak of the dead in circles facing each other (Ridley, 1961:26). Within Grimsby, Feature 9 also exhibited an oval form with some vertebral columns still in articulation, but skeletons generally bundled, skulls to the outer part of the oval. Feature 9/1 (a woman of mixed French/Indian ancestry; Jackes, 1988:29-32) lay extended at the western end, her head to the north, buried with tubes made from the humeri of trumpeter swans.

Everyone in Feature 62, except one older male laid out at the east end (Fe62/58), was buried in more or less the same state of partial or complete disarticulation. The Neutral practice therefore seems to differ from that of the Huron by which very large numbers of people were buried at the same time. The articulated and partially articulated skeletons, present in a Huron village just before the Feast of the Dead, must have undergone a process of dismemberment so that they were ready for bundling in the same way as the disarticulated skeletons. The basis for this statement is the presence of numerous cut marks on Huron bones, especially at the hip joint and neck, indicative of purposeful dismemberment. It is particularly telling that there are no cut marks in the Grimsby cemetery, a point of sharp contrast with Huron ossuaries (Jackes, 1988:137). Furthermore, the frequent articulation of the spinal columns, very evident from the published Grimsby grave plots (W.Kenyon, 1982), contrasts markedly with Huron material from the Kleinburg Ossuary. Jackes (1977) examined over 12,000 vertebrae which had been randomly spread throughout a pit about 4 metres across: no more than a handful of vertebral articulations were present in the Kleinburg Ossuary. It seems most likely that the Neutral were buried in various states of articulation or disarticulation (W.Kenyon 1982:230-231), without efforts made to dismember bodies. Huron skeletal articulations were, by contrast, purposefully disrupted.

The Neutral may well have had quite frequent burial ceremonies, after the spring thaw and before the winter freeze, for example, during which any number of people could be buried. W.Kenyon (1977:11) considered that a summer death lead to a single burial and that those who died in the winter were given spring burials in multiple graves. While this may explain some of the multiple burial pits, it cannot be the explanation for the largest of the features. Burial in the larger pits was delayed until natural disarticulation was almost complete, surely beyond one Ontario winter. Burial was also delayed to some extent to allow burial by social category, very evident from Feature 62 at Grimsby.

Grouping of individuals in Feature 62

Figure 3 shows the distribution of bodies in Feature 62 as determined by skulls, except for the areas marked x, y and z. The arrangement of skulls in a rough oval is evident, the major deviation being the group of subadults in the SW portion of the burial next to the area marked x, y, z.

Individuals 103 and 104 (marked as "x") were two tall children, buried with mandibles, but not skulls. They were buried so close together that their bones were mixed by the excavators. Their permanent dentitions both displayed an unusual eruption sequence (Jackes 1988: 21,25,132).

The area marked "y" and "z" contained the bodies of two males (without skulls) and one female who shared an abnormality of familial origin (Legge-Calvé-Perthès' Disease; see Jackes, 1988:179 index references to individuals Fe62/N, Fe62/O, Fe62/B).

Further evidence of familial relationships comes from the aggregation of minor inherited traits. At the east end of the pit lay three individuals with semisacralized fifth lumbar vertebrae. Seven cases of neural arch separation of the fifth lumbar vertebra occurred, four of them in close proximity in the southern sector of the oval. Altogether, 63% of the fifth lumbar vertebral neural arch anomalies in the entire site occurred within a 1.5m² sector of Feature 62 (Jackes, 1988:60).

Two tall women buried together at the eastern end of Feature 62, marked by "c", both had inca bones (divided occipital bones in the skull; these two cases constitute 50% of those present in the cemetery [Jackes, 1988:12,40,132]).

Two people buried together, marked by "d", display 67% of the cases of agenesis of the left lower third molars in the site (Jackes, 1988:40).

The evidence is very strong then, that familial relationships pattern the burial of individuals within Feature 62.

But sex and age clearly determine the burial pattern also. Males lie at the eastern end of Feature 62. Only two adult males were buried at the west end: Fe62/4, who may have had ankylosing spondylitis (Jackes 1988:73-75), and Fe62/2, who had a fracture of the left ulna with misaligned healing and several ruptured intervertebral discs. The other adult males (those without disabilities) lie in the eastern half, and at the far eastern point of the oval seven old males were buried together, along with one old female and the two tall young females with inca bones mentioned above. Since 25% of the older males buried in the Grimsby cemetery lie within the half metre or so at the eastern end of Feature 62, it is justifiable to consider that age and sex are determinants of place of burial.

Disposition at burial of skeletons in Feature 62

Within Feature 62, variations in the burial mode were recorded by the excavation team (led by K. Mills). It has been possible to correlate burial mode with age and sex in 67 cases (Table 3). In 57% of these individuals, the burials were described as "accordian" by the excavators: 69% of males vs. 54% of females (a significant difference, P = .002). The accordian burial mode is no doubt that described by Kenyon (1982:231) as "semiarticulated bundle".

The majority of the individuals were therefore buried in a state of semi-disarticulation. Forty-two percent of individuals lay "on their backs" with regard to their skulls and spinal columns. Thirty-three percent lay "face-down", 11% on their right sides and 13% on their left sides. There is a difference between the sexes: males were significantly more likely to be buried lying on their backs than females (P = .000).

Males in the accordian mode were equally likely to be prone or supine, whereas females were most likely to be buried face down, or on their left side (Table 4). Apart from the tendency for children 1 to 4 years of age to be buried in multiple person bundles, and for children under 15 to be more likely to be "jumbled", no clear trend by age at death can be observed. At all ages, the "accordian" mode is most common. Adolescent and young adult males, as well as old females, are sometimes given "flexed" burial, implying a lesser state of disarticulation.

In sum, although there are differences between the sexes, Feature 62 does not give us definite information on a correlation between age, sex and the mode or posture at burial, except, perhaps in the case of Fe62/58, the old male buried extended across the eastern apex of the burial oval. There is variability in the burial postures, but the complexity to be discerned in Feature 62 is not based on burial mode. Rather the complexity derives from the clear patterning of burial placement, indicating both genetic relationships and status. The cemetery as a whole, in fact, supports the contention that age, sex and status are important determinants of place of burial.

Burial by sex and age in Grimsby Cemetery

Multiple burial features from Grimsby can be clustered in an analysis of the cemetery as a whole (Figure 4).

Group B are the features with many children: Features 9 and 11 (and perhaps the disturbed Feature 45) constitute Group B features. Together with Features 17, 18, 51, 56 and 59, these features have more females and children than adult males.

Group C are those features with many adult males: in Features 20, 26, 30, 36 and 62, up to 80% of individuals are adult males.

Thus, with the exception of Features 1 (and 28?), all of the undisturbed larger burial features in the Grimsby cemetery can be grouped into two contrasting types of burial. This confirms the apparent tendency to separate males from children in burial.

In general, it is the older males who are separated from women and children: the old males are grouped together (Features 26, 30, 36 and 62), buried by themselves (Features 10, 60 and 61), or with older females (Features 23, 33 and 50). There is a tendency to group old males together and to separate them from the young.

Infants are generally buried separately. Young children are most likely of all age groups to be buried in single graves, otherwise they are buried with old females. The excess of 5 to 15 year olds in Feature 9 is a clear indication of cultural selection at burial. Late adolescents are nearly always buried in large features and their burial pattern is different from that of infants and of children from 5 to 15. The separation of the group of features with children (Group B) and those with many males (Group C) is underlined by the fact that all older males are buried in Group C features apart from one possible individual in Feature 11.

Status

If older males were of highest status in Neutral society, together with certain females (for example, Fe9/1 and the two women buried at the eastern end of Feature 62), it is clear that burial in the Grimsby Cemetery did differentiate between higher and lower status individuals. However, Fox considers that the Grimsby Cemetery is unique -- apart perhaps from the extensively disturbed and incompletely excavated Walker Site -- possibly a spiritual centre with burials of high status individuals (I.Kenyon and Fox, 1982:13; Fox, in litt. 21/1/83). It is completely justifiable to question whether the demographic profile presented by the Grimsby skeletons has biological meaning, or whether the aggregation of skeletons in the cemetery has cultural, rather than biological meaning . In so far as it is possible for osteological evidence to stand alone, without its archaeological context, the Grimsby skeletons suggest that, although burial is by cultural category, the total sample indicates a biological population.

The evidence which comes from an analysis of the age at death distribution suggests that the Grimsby Cemetery as a whole could represent a single community (a town or an intermarrying group of villages). The sample gives no indication of bias such as would be expected were high-status individuals assumed to be older males and females. In total, the Grimsby sample gives strong evidence of high mortality virtually unparalleled in the demographic literature, on a par with that in the Arikara post-Contact sites (Jackes, 1986), because more subadults aged 5 to 20 are buried than would be expected except under very severe mortality conditions.

It is extremely difficult, perhaps impossible, to provide accurate ages at death. For this reason, we must look for other evidence to sustain the hypothesis that Grimsby people died at an earlier age than those in other Ontario burial sites. In fact, the suggestion of a low mean age at death is strongly supported by examination of dental pathology. Dental pathology is an age-dependent characteristic, so that early age at death would lower rates of dental pathology very strongly. Grimsby has reduced dental pathology, relative to other Ontario populations, although stable isotope and trace element analyses do not suggest that the diet was less cariogenic because of reduced dependence on maize (Jackes, 1988:118,125). The hypothesis of high mortality is thus supported by independent evidence from dental pathology.

Burial by family relationship at Grimsby

It has long been considered that family relationships may determine placement within the smaller burial units in Neutral cemeteries. Crerar's finding (1983) of a woman and child with bipartite patellae buried together at Walker, the burial together of two slightly anomalous individuals in Feature 17 at Grimsby (Jackes, 1988:13), as well as the overwhelmingly strong evidence provided by Feature 62, all point to the burial together of closely related individuals.

Burial by physical condition

The extent to which individuality is maintained in the Grimsby cemetery is noteworthy. People are more or less people. There are a few cases where people were, it seems on purpose, buried without their heads. These have been discussed above in the context of a detailed description of Feature 62: the adults were people with marked disabilities. No skull was associated with Feature 1/33, an individual who may have suffered from smallpox osteomyelitis (Jackes, 1983), but Feature 1 cannot be regarded as typical of Grimsby burials.

Given the recorded emphasis on resurrection noted for the Neutral (G.K.Wright, 1963:19-20), one might suspect that individuality was retained for a purpose:

The Attiuoindarons enact Resurrections of the dead, chiefly of those who have deserved well of their country by remarkable services, to the end that the memory of illustrious and valorous men may in some manner come to life again in the person of others. So they call meetings with that object, and hold councils, at which they choose some one among them who possesses the same virtues and characteristics, if that is possible, as he whom they wish to resuscitate, or at least one whose manner of life is irreproachable among savages. When ready to proceed to the Resurrection, they all rise except the one to be resuscitated, on whom they bestow the name of the dead man, and all putting their hands far down pretend to lift him from the ground, meaning thereby that they draw out of the tomb that great man who was dead and restore him to life in the person of the other, who rises to his feet, and after great applause by the people receives the presents offered him by those taking part. They also congratulate him with several feasts and henceforth treat him as the dead man whom he represents, and thus the memory of worthy people and excellent brave chiefs never dies among them.

While we find no clear distinction in the ethnohistorical accounts of the two societies, there is every reason to consider that the Neutral ceremonial treatment of the dead was not of the Huron type. The emphasis was on community among the Huron (Tooker 1967:139-140), and the Huron Feast of the Dead was the occasion for stressing corporate relations: the dead were brought and their bones mingled. The life of the dead was thought of as a reflection of the world of the living (Tooker 1967:139-149). Their living relatives feasted together and indulged in gift exchanges in a classical anthropological rite of community intensification. (see also Hickerson, 1960, 1963).

In contrast, the Neutral burials seem to emphasise distinctions among the dead. The reasons for this may lie in differentiation in the fate of the dead in the afterlife. We know from the Jesuit accounts of the Huron that the souls of children, the elderly, perhaps the disabled, those who died in battle, suicides (Tooker, 1967:140-141; Ramsden, 1990:380) might not end up in the general village of the dead.

While the Neutral may have laid greater emphasis on these distinctions than the Huron, it is more likely that the explanation for the difference lies in the nature and function of Neutral burial areas. It seems very likely that the cemetery area associated with a Neutral population centre was used more often and over a longer period of time than was the case for Huron ossuaries. A single Huron burial pit contains the bodies of those who died during the period of the occupation of one village site, all of whom were buried in one long and ornate ceremony. The Grimsby cemetery probably had an extended period of use and was the scene of numerous burial ceremonies (see Jackes, 1986:40-41).

Bead clusters and demography

Kenyon and Fox (1982) suggest that the Grimsby Cemetery may cover as much as 35 years and they distinguish three phases based on glass beads included in 29 features, each phase covering 10 years or more. Fox has kindly provided his assesment of the temporal sequence of all possible Grimsby features, although "these temporal identifications stretch intuitive assessment to the limits, as they are often based on limited artifact samples" (in litt. 21/1/83).

The first phase (II) is represented by only six features and 27 individuals, too few to make any determination of demographic trends. Feature 45 is included in this phase. The second phase (IIIA) comprises 114 individuals in nine features, including Features 9, 11 and 36. The third phase (IIIB) includes Features 1, 28 and 62, and has in it 146 individuals in 15 features.

The groupings deduced from bead types and from the sex and age make-up of the features cross-cut each other; but Jackes (1986:45) has shown that the sex and age breakdown for the two later bead phases (IIIA and IIIB) are significantly different. The earlier bead phase has significantly more children, while the later phase has significantly more males and significantly more older people of both sexes.

It is therefore essential that we consider the possibility that the feature groups B and C and the bead phases reflect, not cultural selection at burial, but changes in demography over time.

Demographic analysis undertaken here is of the limited type designed to assess bias or trends in populations undergoing increase or decline (Jackes, 1986;1992). Analysis of the last bead period (Phase IIIB) provides a scenario of fairly high mortality in a population which is not in active decline (mean childhood mortality [MCM] = .126, juvenile:adult ratio [J:A] = .310). Analysis of the penultimate bead phase (Phase IIIA), however, provides a picture of extreme bias (MCM = .219, J:A = .766), or else population increase (MCM = .19, J:A = .59 at r = +.01) beyond the levels imaginable within the context of fertility in Ontario in the seventeenth century (Jackes, 1994). Given this, we must consider the possibility that feature groupings and bead phases are broadly contemporaneous aspects of the same burial mode in which adult males and young children are most likely to be given separate burial (see Jackes, 1988:141).

Such an interpretation is in line with the hypothesis of Fitzgerald (1983:18): "It would seem rather than being a distinct bead period ... Period 3a is an assemblage that would seem to span the latter years of Period 2 and the early part of ... Period 3b."

Conclusions

While Figure 5 illustrates how disparate the features are in their demographic makeup, it cannot encompass the extreme complexity of single and multiple graves and old and young adults, infants, children and adolescents. Explanations in terms of high status burials, or in terms of smallpox epidemics, while not irrelevant, cannot embrace all the complexity of Grimsby Cemetery. The cemetery as a whole appears to represent a community, most probably a community which was suffering an extreme rate of mortality. Mortality among subadults was excessively high, and the community could not have sustained itself for long (in this context, it is relevant to note that the Grimsby skeletons do not provide us with any evidence of an influx of refugees from Huronia, or from other Neutral villages: distance studies of traits of the teeth, crania and post-cranials indicate that the Neutral at Grimsby constituted a gene pool, separable from those represented by other Neutral skeletal collections [Jackes, 1988]).

It is unfortunate that we have little to compare Grimsby with, although Grimsby is not an isolated example. Shaver Hill (AD 1600-1620) seems to have had careful bone arrangements and burial by age categories (Stothers, 1971, 1972), but the information is sketchy. White (1966:6) mentions arranged bones in the pre-Contact Orchid Site, but admits that, under the conditions of excavation, it was impossible to confirm the observation that the burial pit was outlined by skulls and that skulls and pelves were often associated. The Walker Site had a cemetery like that of Grimsby, and is at least partly contemporaneous. Descriptions of other Neutral burial sites provide little firm information: some sites have been "potted" continuously since the last century, some sites have now disappeared and all that is left are vague references to "ossuaries".

The evidence points to Grimsby being neither unique nor the result of epidemics, but the common pattern for Neutral cemeteries in the seventeenth century. That being the case, Neutral burial is nothing like that of the Huron who intermingled the dead in stressing wider corporate relationship and perhaps trading relationships. The Neutral were concerned to emphasize relationships within the smaller corporate units important to them, with much stress upon the categorization of individuals by age and by sex and ultimately by social status.

Ossuary burial seems not to have been consonant with Neutral beliefs, and we have no evidence that the Neutral practised ossuary burial as defined. Rather the Neutral sites, in general, probably had a mixture of single and multiple burials, the latter varying in size and often having arrangements of placed bundles. These bundles of semi-articulated or disarticulated individuals were placed with attention to age, sex, status and family relationships. Burials of the various types may overlie and disturb each other.

It is not possible here to explain the differences between Huron and Neutral burial practices, apart from suggesting that Neutral villages were occupied for longer periods than those of the Huron, since there is no good ethnohistorical evidence of marked differences in social structures. Therefore, it is not always warranted to make an interpretative leap and conclude that complexity in burial practices based on differentiation and individuation reflects greater complexity in social structure .

Acknowledgments

Working with Walter Kenyon was an experience I would not have wanted to miss. I wish that circumstances had been different and that I could have continued our discussions on the details of Grimsby once the osteological analyses were complete. I am grateful to Bill Fox for information on his work on bead phases at Grimsby and to Milt Wright for discussions on Neutral archaeology.

References

• Dixon, C.W. (1962). Smallpox. London: Churchill.
• Charlevoix, P.F.X. (1900). History and General Description of New France translated and published by J.G. Shea. N.Y.
• Crerar, J.M. (1982). The historic Neutral burial grounds at the Walker Site. MS on file with Dr. S. Saunders, Department of Anthropology, McMaster University, Hamilton.
• Crerar, J.M. (1983). An unusual sub-ossuary pit in the Walker Site burial grounds. Paper presented to the Canadian Archeological Association. Halifax, Nova Scotia. April, 1983.
• Fitzgerald, W.R. (1983). Further comments on the Neutral glass bead sequence. Arch Notes 83-1:17-25.
• Gualtieri, M. and M. Jackes (1993). The cemetery areas: material culture and social organization. Pp. 140-226 in M. Gualtieri (ed.) Fourth Century B.C. Magna Graecia: a case study. Paul Åströms förlag, Partille, Sweden.
• Hartney, P.C. (1978). Palaeopathology of Archaeological Aboriginal Populations from Southern Ontario and Adjacent Region. Ph.D. Thesis, Department of Anthropology, University of Toronto.
• Hickerson, H. (1960). The feast of the dead among the seventeenth century Algonkians of the Upper Great Lakes. American Anthropologist 62(1): 81-107.
• Hickerson, H. (1963). The sociohistorical significance of two Chippewa ceremonials. American Anthropologist 65(1): 67-85.
• Jackes, M.K. (1977). The Huron Spine. Ph.D. Thesis, Department of Anthropology, University of Toronto .
• Jackes, M.K. (1982). Historic Neutral Burial Practices. Paper presented to the 10th Annual Meeting of the Canadian Association for Physical Anthropology, Guelph, November, 1982.
• Jackes, M.K. (1983). Osteological evidence for smallpox: a possible case from seventeenth century Ontario. American Journal of Physical Anthropology 60:75-81.
• Jackes, M.K. (1986). Mortality of Ontario Archaeological Populations. Canadian Journal of Anthropology 5:33-48.
• Jackes, M.K. (1988). The Osteology of the Grimsby Site. Edmonton: Department of Anthropology, University of Alberta.
• Jackes, M.K. (1992). Paleodemography: Problems and Techniques. In S.R. Saunders and M.A. Katzenberg (eds.) Skeletal Biology of Past Peoples: Research Methods. New York: Wiley-Liss, pp. 189-224.
• Jackes, M.K. (1994). Birth Rates and Bones. In A. Herring and L. Chan (eds.) Strength in Diversity: A Reader in Physical Anthropology. Canadian Scholars' Press Inc.
• Jamieson, S.M. (1992). Regional interaction and Ontario Iroquois evolution. Canadian Journal of Archaeology 16:70-88.
• Jerkic, S. (1975). An Analysis of Huron Skeletal Biology and Mortuary Practices: The Maurice Ossuary. Ph.D. Thesis, Department of Anthropology, University of Toronto .
• Katzenberg, M.A. and R. White (1979). A paleodemographic analysis of the os coxae from Ossossané ossuary. Canadian Review of Physical Anthropology 1:10-28.
• Kenyon, I. and W. Fox (1982). The Grimsby Cemetery - a second look. KEWA (Newsletter of the London Chapter, Ontario Archaeology Society) 82-9: 3-16.
• Kenyon, W.A. (1977). Some bones of contention. Rotunda 10:4-13.
• Kenyon, W.A. (1982). The Grimsby Site, an Historic Neutral Cemetery. Toronto: Royal Ontario Museum.
• Kidd, K.E. (1953). The excavation and historical identification of a Huron ossuary. American Antiquity 18:359-379.
• Knight, D. and J. Melbye (1983). Burial patterns at the Ball Site. Ontario Archaeology 40:37-48.
• LeClercq, C. (1881). First Establishment of the Faith in New France. Translated and published by J.G. Shea. N.Y.
• Masset, C. (1993). Les Dolmens: Sociétés néolithiques et pratiques funéraires. Editions Errance, Paris.
• Noble, W.C. (1968). Iroquois archaeology and the development of Iroquois social organization (1000-1650 A.D.). Ph.D. dissertation, Department of Archaeology, University of Calgary.
• Noble, W.C. (1984). Historic Neutral Iroquois settlement patterns. Canadian Journal of Archaeology 8:3-37.
• Noble, W.C. (1985). Tsouharissen's chiefdom: an early historic 17th century Neutral Iroquoian ranked society. Canadian Journal of Archaeology 9:133-146.
• Patterson, D.K. (1986). Changes in oral health among prehistoric Ontario populations. Canadian Journal of Anthropology 5:3-13.
• Pfeiffer, S. (1983). Demographic parameters of the Uxbridge ossuary population. Ontario Archaeology 40: 9-14.
• Ramsden, P.G. (1990). The Hurons: archaeology and culture history. Pp. 361-384 in C.J. Ellis and N. Ferris The Archaeology of Southern Ontario to A.D. 1650. Occasional Publications of the London Chapter, Ontario Archaeological Society Inc., Publication Number 5. London, Ontario.
• Ridley, F. (1961). Archaeology of the Neutral Indians. Port Credit, Ontario: Etobicoke Historical Society.
• Rothschild, N.A. (1979). Mortuary behavior and social organization at Indian Knoll and Dickson Mounds. American Antiquity 44:658-675.
• Stothers, D.M. (1970). A comparative analysis of Huron and Neutral ossuary burial information. Toledo: The Toledo Area Aboriginal Research Society.
• Stothers, D.M. (1971). The Shaver Hill burial complex: reflections of a Neutral Indian Population. Unpublished Ph.D. dissertation, Department of Anthropology, Case Western Reserve University.
• Stothers, D.M. (1972). The Shaver Hill burial complex - reflections of a Neutral Indian population: a preliminary statement. Toledo Area Aboriginal Research Bulletin 2(1): 28-37.
• Sutton, R.E. (1988). Palaeodemography and late Iroquoian ossuary samples. Ontario Archaeology 48:42-50.
• Thwaites, R. (1898). The Jesuit Relations and Allied Documents. 73 Vols. Cleveland: Burrows.
• Tooker, E. (1967). An Ethnography of the Huron Indians 1615-1649. Huronia Historical Development Council, Sainte-Marie among the Hurons, Midland, Ontario.
• Trigger, B.G. (1985). Natives and Newcomers: Canada's "Heroic Age" Reconsidered. Kingston and Montreal: McGill-Queen's University Press.
• White, M.E. (1966). The Orchid Site ossuary, Fort Erie, Ontario. Bulletin of the New York State Archaeological Association 38: 1-24.
• Wright, G.K. (1963). The Neutral Indians, a source book. New York State Archaeological Association, Occasional Paper No. 4.
• Wright, M.J. (1981). The Walker Site. Archaeological Survey of Canada Paper No. 103. National Museum of Man Mercury Series. Ottawa.
• Zilhão, J. (1984). A Gruta da Feteira (Lourinhã): Escavação de salvamento de uma necrópole neolítica. Trabalhos de Arqueologia 1. Lisbon: Instituto Português do Património Cultural.

	Figure 1.  Major sites mentioned in  the text: 1- Glen Williams;  2 - Kleinburg;  3 - Uxbridge;  4 - Maurice;  5 - Ossossané;  6 - Shaver Hill; 7 - Walker;  8 - Orchid).  back to text
	Figure 2. Percentage of dead by age (J - juveniles)  and sex (M, F - males and females over 20 years of age)  for some features in the Grimsby Cemetery. back to text
dark screen adult males; light screen adult females; stipple under 20 years of age	Figure 3.  Distribution of skulls in Feature 62 at Grimsby.  back to first mention in text Key:  a. semisacralized L.5;  b. L.5 neural arch separated; c. Inca bone;  d. M3 agenesis; x. area of burial of Fe62/103 and Fe62/104;  y. area of burial of Fe62/N and Fe62/O;  z. area of burial of Fe62/B. back to detailed discussion
Key: red - Group C, many adult males; blue - Group B, many females and juveniles;  yellow - adult female, single burial; grey - adult male, single burial; green - infants or children only;  hatched - vandalized or indeterminate multiple burials.	Figure 4. Distribution of features within the Grimsby Cemetery.  back to text
	Figure 5. Composition of larger burial features by age and sex. back to text