Content uploaded by Alex Bayliss
Author content
All content in this area was uploaded by Alex Bayliss
Content may be subject to copyright.
85
Date of the West Kennet Long Barrow
Talking About My Generation:
the Date of the West Kennet Long Barrow
and contains transepted orthostatic chambers at its
eastern end. These consist of an end chamber, two
pairs of opposed chambers, and a passage, principally
formed by sarsen orthostats (Fig. 1). The passage led
in from what was originally a concave orthostatic
investigated, revealing a core of small sarsen stones
Thirty-one radiocarbon results are now available from the West Kennet long barrow, and are
presented within an interpretive Bayesian statistical framework. Two alternative archaeo-
logical interpretations of the sequence are given, each with a separate Bayesian model. In
our preferred interpretation, the barrow is seen as a unitary construction (given the lack of
dating samples from the old ground surface, ditches or constructional features themselves),
with a series of deposits of human remains made in the chambers following construction.
Primary deposition in the chambers is followed by further secondary deposition of some
of the monument at West Kennet, as dated from the primary mortuary deposits, occurred
in , probably in the middle decades of the thirty-seventh century cal. .
The last interments of this initial use of the chambers probably occurred in
cal. -
ary activity probably continued for only
and
continued into the second half of the third millennium cal. . In an alternative interpreta-
tion, we do not assume that all the people dated from the primary mortuary deposits were
they could therefore
have died before the monument was built, although they must have died before the end of
the formation of the mortuary deposit. In the Bayesian model for this interpretation, the
monument appears to belong either to the thirty-seventh century cal. or the mid-thirty-
sixth century cal. , and deposition again appears short-lived, but the model is unstable.
The West Kennet long barrow is one of the best
recognized from at least the time of John Aubrey in
by John Thurnam in the nineteenth century and more
above the upper Kennet valley in the region around
Cambridge Archaeological Journal
et al.
this sarsen core could suggest that this feature was
part of an earlier, perhaps free-standing version
ditches about a third of the way from the western end,
suggested that the mound may be of more than one
-
from a small part of the buried soil,
well west of the façade, was carried
Primary deposits
The monument is thus notable as an
imposing chambered long barrow,
generally thought of as part of the
wider Cotswold transepted type, but
of its chambers. Thurnam’s investi-
gation came down through the end
chamber, revealing inhumations, and
went some way along the passage,
encountering substantial secondary
-
the side chambers, covering further
primary inhumations. These pri-
mary mortuary deposits consisted of
articulation and completeness, as
was originally estimated that at least
represented in
later suggested that these remains
could be seen as arranged by some
sort of categorization on the basis of
-
ber, predominantly adults (men and
women) in the inner side chambers,
-
original estimate of numbers was too
Figure 1. The West Kennet long barrow.
Date of the West Kennet Long Barrow
selective abstraction.
Both the style of the monument and artefacts
associated with the primary interments suggested a
-
carbon dates then recently obtained from Windmill
Secondary deposits
Following the active mortuary use of the tomb and
the partial collapse of sections of internal dry stone
walling, the contents were sealed beneath deliber-
to the roof.
of the passage untouched by Thurnam, contained
accumulated over a very long period of time elsewhere,
monument, in a single act, at a time given by the lat-
argued that West Kennet could have been in use as a
mortuary facility for several hundred years before being
remained open for use over a span of centuries became
-
much more gradual process of deposition over a long
of radiocarbon dates from Hazleton.
-
ably more adult and immature human material depos-
in loose groups or sub-assemblages and have been
Objectives of this study
Further dating of the West Kennet long barrow was
advances in radiocarbon dating and the interpretation
of radiocarbon dates which have been made in the last
Figure 2. Revised demographic distribution for the primary deposits.
W Chamber
(1 Child)
SW Chamber NW Chamber
5 Adult
and cranial fragment only)
SE Chamber NE Chamber
1 Adult 1 Adult
1 Adult 1 Adult
1 Adult ?
88
et al.
decade or so (Bayliss et al. this issue). These develop-
ments provide the potential to produce much more
et al
understanding of the chronology of the Neolithic
monuments within the Avebury area has also recently
contribute to public understanding and appreciation
• to date the primary construction of the monument
• to determine the dates of the mortuary deposits
• to determine whether there was spatial variation
in the deposition of human remains within a burial
• to determine whether the mortuary deposits in the
• to determine whether the articulated burials in the
•
• to establish the relative position of West Kennet
-
and mortuary processes enabled the strategic selection
-
including of its failure to distinguish
in some cases between material from
details will be provided elsewhere
This re-
cent study also suggests that there are
no direct taphonomic traces evident
elsewhere for any of the material
a number of individuals were clearly
deposited in an articulated state as
can be seen from the original plans
seen in partially articulated individuals do not con-
-
corpses.
Sampling
-
ment was constructed to assess the number of samples
a resolution which would be archaeologically useful
et al
error terms for the radiocarbon measurements based
on the available samples.
Certain types of sample were targeted for dat-
• articulated bone groups which could not have
the death of the individual concerned without be-
• bone groups where articulated deposition could be
• disarticulated human remains from individuals
-
logical duplications.
All specimens were selected to ensure that each dated
sample was from a separate individual. This allows
measurements on the same body to be combined
before calibration, so that all dates included in the
Figure 3. Radiocarbon dates for the long barrow obtained in 1984, calibrated
using the maximum intercept method (Stuiver & Reimer 1986) and data
from Pearson et al.
Date of the West Kennet Long Barrow
models are statistically independent
-
tion, sampling locations on individual
bones were chosen to avoid any areas
showing previous use of consolidant
or adhesives.
The first series of samples
-
and disuse of the monument. Once
these results were received, and a
preliminary model constructed, fur-
ther samples were selected to resolve
results and address more detailed
disarticulated human bone from the
primary mortuary deposits were
selected to see whether they were of
second series of measurements had
been completed, a technical problem
-
-
et al
Bayliss et al. this issue). The resolution
of this problem necessitated a third
series of replicate samples.
Results
Thirty-one radiocarbon results are
now available from West Kennet
-
individuals come from the primary
-
man samples and the goat come from
the secondary deposits.
The results are conventional
-
vided in Table 1 have been calculated
et al
et alet al
samples were processed and measured according to
Figure 4. Probability distributions of simulated dates from West Kennet.
Each distribution represents the relative probability that an event occurs at
a particular time. For each radiocarbon date, two distributions have been
with, for example, YY, is the growth of the person whose bones were dated.
The other distributions correspond to aspects of the model. For example,
the distribution ‘build’
et al.
Table 1.
et al. this issue).
Laboratory
no.
Sample no. and material Radiocarbon
age (bp)
δ13C (‰) δ15N
(‰)
C:N
ratio
Weighted
mean (bp)
Calibrated date
range (95%
condence)
Posterior density
estimate (probability)
c.
strongly suggest that it was articulated
when deposited.
T'
T'
= 1
3650–3620 cal. (79%) or
3565–3530 cal. (16%)
c.
deposit
to strongly suggest that it was
articulated when deposited.
T'
T'
= 1
3655–3620 cal. (80%) or
3560–3535 cal. (15%)
8.8
c.
deposit
to strongly suggest that it was
articulated when deposited.
3655–3620 cal. (79%) or
3565–3530 cal. (16%)
deposit
the individual was at least partially
articulated on deposition.
3655–3625 cal. (81%) or
3560–3535 cal. (14%)
deposit
the individual was at least partially
articulated on deposition.
3660–3625 cal. (81%) or
3555–3535 cal. (14%)
3650–3610 cal. (79%) or
3570–3530 cal. (16%)
NW chamber primary deposit
11.8
T'
T'
= 1
3655–3625 cal. (81%) or
3555–3535 cal. (14%)
1
(assumed)
3655–3615 cal. (79%) or
3570–3530 cal. (16%)
NW chamber primary deposit
(assumed)
3660–3615 cal. (79%) or
3565–3530 cal. (16%)
humerus from NW chamber primary
deposit
3655–3625 cal. (79%) or
3560–3535 cal. (16%)
humerus from NW chamber primary
deposit
3650–3615 cal. (79%) or
3570–3530 cal. (16%)
c.
deposit
(assumed)
3655–3615 cal. (79%) or
3570–3530 cal. (16%)
c.
elements survive to suggest that
the individual was at least partially
articulated on deposition.
3655–3620 cal. (80%) or
3560–3535 cal. (15%)
Date of the West Kennet Long Barrow
Laboratory
no.
Sample no. and material Radiocarbon
age (bp)
δ13C (‰) δ15N
(‰)
C:N
ratio
Weighted
mean (bp)
Calibrated date
range (95%
condence)
Posterior density
estimate (probability)
deposit
3650–3620 cal. (79%) or
3565–3530 cal. (16%)
deposit
3655–3615 cal. (79%) or
3565–3530 cal. (16%)
deposit
11.8 3645–3605 cal. (80%) or
3570–3525 cal. (15%)
T'
T'
3655–3630 cal. (81%) or
3560–3540 cal. (14%)
primary deposit
(assumed)
T'
T'(5%) =
= 1
3650–3620 cal. (79%) or
3560–3530 cal. (16%)
WK 11, right femur from partially
3655–3625 cal. (80%) or
3560–3535 cal. (15%)
of post-cranial remains, suggesting
articulation or partial articulation at
the time of deposition
3345–3205 cal. (88%) or
3195–3150 cal. (7%)
that it was articulated on deposition
11.5 3300–3085 cal. (91%) or
3060–3030 cal. (4%)
c.
suggesting articulation on deposition
3305–3320 cal. (95%)
WK 8, humerus from an infant,
suggesting that it was articulated on
deposition
2875–2800 cal. (41%) or
2780–2595 cal. (54%)
WK 5, rib fragment of infant,
c.
secondary deposits. From largely
suggests articulation on deposition.
2840–2805 cal. (4%) or
2760–2565 cal. (87%) or
2525–2495 cal. (4%)
infant, c.
secondary deposits
indicating articulation at the time of
deposition.
3495–3455 cal. (4%) or
3380–3260 cal. (34%) or
3250–3095 cal. (57%)
Capra sp., male) from
chamber secondary deposits
2570–2515 cal. (20%) or
2500–2335 cal. (75%)
Table 1. (cont.)
et al
et al
processed using the gelatinization protocol described
et al-
ery in the laboratory of a contamination problem
associated with this method, in eighteen cases the
contaminated material was re-processed, graphitized
et al
-
et al.
et al
et al.
-
sey et al
et alet al.
Interpretations
Two alternative chronological models for the West
Primary deposits
A schematic representation of the composition of the
primary deposits of human bone in each chamber is
deposits in the southeast chamber. Two were dated
case producing statistically consistent radiocarbon
-
abundant ribs, vertebrae, hand and foot bones assign-
that they were probably in an articulated or partially
is represented by a disarticulated adult femur, and so
we have no evidence whether this body was originally
deposited articulated. From the published site pho-
tographs, it seems that this material, much of which
appears to have been originally deposited in at least
no certain stratigraphic relationships between the
dated individuals in this chamber although some of
the adult bones were recorded overlying some of the
Five individuals have been dated from the
primary deposits in the northwest chamber. Two
statistically inconsistent radiocarbon results (Table 1)
have been obtained on the scapula from the partially
-
-
tistical outlier. Therefore the measurements have been
combined before calibration. Four other disarticulated
bones were dated. Two right humeri, which on mor-
-
however, might be from one of the individuals whose
right humerus has been dated.
Five individuals have also been dated from the
-
mur was dated from a partially articulated sub-adult
that it was at least partially articulated on deposition,
although the remains were recovered disarticulated.
adults. These bones were recovered disarticulated,
in articulation.
Five radiocarbon results are available from three
chamber. Two were dated in replicate, in each case
producing statistically consistent radiocarbon meas-
Figure 5. Summary of prior information incorporated in
the chronological model shown in Figures 6 and 7. The
stratigraphic relationships between samples are shown
the right-hand side represent uniformly distributed
phases of activity.
Date of the West Kennet Long Barrow
incorporated in the chronological model, because
No samples were dated from the west chamber.
-
tions of this chamber does survive, having been re-
this history and its scarcity, it was decided not to use
it in this programme. Four male crania (recently re-
radiocarbon dating.
Secondary deposits
Numerous sub-assemblages containing immature hu-
-
cretions were found in the archive labelled as com-
where labelling was absent or uninformative, such
sub-assemblages still contained fragmentary animal,
ceramic and carbonized material, again
-
or poorly labelled sub-assemblages to either primary
material from both primary and secondary deposits
-
came from the primary deposits).
-
is ‘from above the dry stone walling of the northwest
corner of the southeast chamber’. The precise locations
this relationship is not included in the chronological
samples in this chamber, coming ‘from the surface
A largely complete, and partially articulated,
Results from chronological modelling
the construction of the transepted monument at West
Kennet, as dated from the primary mortuary depos-
its, occurred in 3670–3635 cal. (81% probability) or
3575–3545 cal. (14% probabilitystart primary). The
last interments of this initial use of the burial cham-
bers occurred in 3640–3610 cal. (77% probability) or
3550–3520 cal. (18% probability end primary). The
that this primary mortuary activity continued for
Table 2. Posterior density estimates for the dates of archaeological
events and the duration of activities at West Kennet, derived from the
model described in Figures 5–7.
Model 1 (Figs. 5–7)
Distribution Posterior density
estimate (68%
probability)
Posterior density estimate
(95% probability)
start primary 3655–3635 cal. 3670–3635 cal. (81%) or
3575–3545 cal. (14%)
end primary 3635–3615 cal. 3640–3610 cal. (77%) or
3550–3520 cal. (18%)
3510–3295 cal. 3620–3240 cal.
2475–2225 cal. 2545–2065 cal.
primary_use 10–30 years 1–55 years (94%) or
115–140 years (1%)
abandonment 95–320 years 1–375 years
895–1235 years 775–1420 years
et al.
only 10–30 years (68% probability),
1–55 years (94% probability) or 115–140
years (1% probability primary use) (see
The short duration of the pri-
mary mortuary activity at West
radiocarbon dating, it is not possible
to distinguish any chronological
variation in the spatial distribution of
corpses within chambers, or between
chambers, when the actual duration
of the activity was so short.
The radiocarbon determinations
primary mortuary deposit are statis-
tically consistent (T'T'(5%) =
This means that all these individuals
could have died at the same time,
although they do not have to be pre-
-
ever, that they are all close in date.
The chronological model shown
agreement (Aoverall
radiocarbon dates are consistent with
the interpretation, included in the
model, that none of the disarticulated
material was residual or ancestral.
-
ments from these deposits form such
a coherent group, if the disarticulated
samples were ancestral by even one
or two generations, the model would
probably show poor agreement (see
below for further discussion of this
point). The model shown also has
et al. this is-
sue).
Once the primary mortuary
to have been a hiatus before the
secondary accumulation of deposits
within the chambers began. This
hiatus probably lasted for rather more
Figure 6. Probability distributions of dates from West Kennet, with
all burials in the primary mortuary deposits interpreted as freshly dead
individuals. The format is identical to that in Figure 4. The large square
model exactly.
Figure 7. Probability distributions
of the number of years during which
various activities occurred in the West
Kennet long barrow, derived from the
model shown in Figure 5.
Date of the West Kennet Long Barrow
abandonment). According to
chambers began in 3620–3240 cal. (95% probability
), and continued into the second half of the
third millennium cal. ). Overall
). We discuss the archaeological
An alternative model for the chronology of the
have not assumed that all the people dated from the
primary mortuary deposits were placed in the monu-
They could therefore have died before the monument
was built, although they must have died before the end
of the formation of the mortuary deposit (Fig. 8). This
interpretation is included in the second model (Figs.
(Aoverall
is unstable and does not produce consistent results.
This can be illustrated by more detailed consid-
eration of the posterior density estimate for the start
of deposition of the primary burials in the chambers
(start primary
because of the strong ‘wiggle’ in the calibration curve
et al
Bayliss et al
the possibility that the actual date of the distribution
where the sampler has become trapped in the earlier
part of the thirty-seventh century cal. , even though
start primary really dates to the
-
terior density estimates derived from this alternative
start primary
model. Consistently, however, this interpretation of
the primary deposits at West Kennet favours the later
preferred.
We believe on archaeological and osteological
grounds (discussed above with reference to the prob-
able absence of secondary rites) that, of these two
and disarticulation are the only reasons for supposing
that there could have been secondary burial at West
Kennet, but it is clear that disarticulation is the result
at least in part of movement of material around the
chambers, as seen in the rearrangement of material
in the northwest chamber (e.g. caches of
Discussion
The dating programme at West Kennet was slightly
more limited than those for Hazleton long cairn and
series, for which samples constraining the construc-
tions were available. An archaeologically more reliable
underlying the barrow, the primary sarsen core and
Figure 8. Summary of prior information incorporated
in the chronological model shown in Figures 9 and 10.
The stratigraphic relationships between samples are
down the right-hand side represent uniformly distributed
phases of activity.
et al.
-
tended at some point in the future.
-
fore be stressed that the dates ob-
tained here are termini ante quos for
the construction of the monument,
in the form that encompassed the
chambers, and that we are assuming
that bodies were deposited in the
chambers as soon as or very soon
cannot discount the possibility that
the monument began with a smaller
as already noted above, nor entirely
disprove the possibility that other
bodies or human remains had earlier
been deposited in the chambers but
However, it is demonstrated above
that there is no support for the dis-
articulated remains here being any
older than the articulated, or partially
articulated, remains.
On the basis of the main model
presented above, we can now date
the construction of the monument
to the middle decades of the thirty-
seventh century cal. (3670–3635
cal. at 81% probability). There is,
however, some possibility that the
monument was actually built in the
century cal. (3575–3745 cal. at
14% probability). We believe that the
alternative model, which perhaps
favours this later construction date, is
less plausible on archaeological and
osteological grounds. Compared with
the literature, the results suggest a
surprisingly short span of primary
use (10–30 years at 68% probability or
Figure 9. Probability distributions of dates from West Kennet, following
an alternative model, with disarticulated single bones from the primary
mortuary deposits interpreted as potential older, secondary depositions. The
format is identical to that in Figure 4, with the convergence for each posterior
density estimate given in square brackets. The large square brackets down the
Figure 10. Probability distributions
of the number of years during which
various activities occurred in the West
Kennet long barrow, derived from the
model shown in Figure 8.
Date of the West Kennet Long Barrow
). That our
-
sets of particular people in particular places and times
is reinforced by the succeeding gap of rather more
than a century (abandonment
the inactivity — at least in this domain — of their suc-
cessors. This model contrasts now rather strongly with
monument lasted some centuries, and was followed
secondary deposition. One further clue to the real-
ity of an abandonment may be found in the signs of
decay, represented by the partial collapse of portions
secondary deposits. By this stage, this was a monu-
ment whose fabric and contents were no longer at the
forefront of people’s minds, or no longer accessible.
-
can note also that, in terms of the total duration of
was surprisingly good in an era when radiocarbon
dating had scarcely been established, since he ended
his report by stating that ‘it seems unavoidable that
the use of the West Kennet tomb, from building to
a long history is incorrect. The dates now available
indicate gradual deposition, over a period of centu-
ries (
elsewhere (even if only in the forecourt) over a long
-
-
pointing out again that an archaeologically more
further investigation. Future research here, however,
organic residues and of calcined bone. The claim for
the secondary deposits recalls the suggestion by Hum-
This is further supported by the observation that
chamber had collapsed on to the primary deposits
and were overlain by the earliest layers of secondary
also give us cause to consider the nature of the hiatus
the southwest chamber suggests the passage of some
-
ment in the sense of an unplanned retreat in the face of
contrary circumstances, there may have been a more
formal or deliberate closure, perhaps also involving
-
-
during their development. The latest date achieved in
the current programme at West Kennet long barrow is
, although this sample may
well not represent the latest activity in the process of
2570–2515 cal. (20% probability) or 2500–
2335 cal.
-
cal. . The date from the goat is not inconsistent
with the currency of
et al.
remains the source of the main material constituting
-
aged as coming from any number of constructions, but
there are few if any signs of activity at monuments in
this material could have been drawn.
to reconsider the chronology of the early part of the
, c.
, c.
cal.
, c.)
came more evidence for clearance and occupation in
and more elaborate barrows, and the appearance of
for the Avebury area now seem too
based on too few dates from too few
sites. Other reviews of the evidence
broad and imprecise timescales (e.g.
trend may have been detected, it
now seems more useful to set named
phases aside and to concentrate on
developments in terms of their abso-
-
that can be placed before c.
. West Kennet long barrow can now
strongly be suggested as belonging
to the thirty-seventh century cal. .
Other potentially earlier long barrows
et al
by far fewer determinations, and so
far, their results have not been treated
within a Bayesian statistical frame-
-
bon dates, could be seen as later than West Kennet, in
the ‘mid to later fourth millennium
was suggested as belonging to the ‘middle of the
fourth millennium
and a visual (non-Bayesian) inspection of the results
-
ible with a period of use for the enclosure from before
. Further results can
dating and Bayesian analysis covering causewayed
-
in detail to West Kennet long barrow. The results from
West Kennet in a real sense change everything for this
region, and we will need to construct other detailed
want to grasp more nuanced histories.
-
miliar, and much discussed, including an interest in
the human dead and their arrangement, categorization
Figure 11.
single sample. This is only a small section of the total sampling run but
allows one to see whether the model is concentrating on particular parts of
the distribution.
Date of the West Kennet Long Barrow
West Kennet long barrow presented here add, using
activity over a short period of time, and an interest
not in timeless, generalized or anonymous forebears
for local agency and local identity, within the structure
of wider changes elsewhere, a relationship which we
et al. this
issue). Our sense of this narrative is becoming more
and it is hard not to believe that it was similar for the
Neolithic people involved.
Acknowledgements
The radiocarbon dating programme reported here was
samples. The re-analysis of the human bone assemblage was
Notes
missing and there is no evidence of sampling on any
have poor convergence, with some of the distributions
from the primary mortuary deposits falling slightly
for these dates at West Kennet are strongly bi-modally
sampler is unable to determine whether the actual
age of these samples falls in the mid-thirty-seventh
.
, is also slightly
more probable in a model which does not allow for
the possibility of secondary deposition but in which
convergence and so is unstable, although an earlier
date in the mid-thirty-seventh century cal. is always
century cal. .
References
Archaeological Research Agenda for the World
Heritage Site
Nuclear Instru-
ments and Methods in Physics Research B
collec-
tion. Radiocarbon
The
Enclosure, Wiltshire
long barrows near Avebury, Wiltshire. Proceedings of
the Prehistoric Society
-
and other places, in
-
fordshire
decade integrating radiocarbon dates into chronologi-
cal models, in
for Crossing Disciplinary Boundaries
Dating
Science and Stonehenge
for the Neolithic.
-
sis of stratigraphy. Radiocarbon
Radiocarbon
Development of the radiocarbon
calibration program. Radiocarbon
-
Archaeometry
Archaeometry
Radiocarbon
Radio-
carbon
Proceedings
et al.
of the Prehistoric Society
Wiltshire Archaeological and Natural History Magazine
Distribution, morphology and artefacts, in Megalithic
Enquiries in the West of Britain,
The Megalithic Chambered Tombs of the
Cotswold-Severn Region
Land Snails in Archaeology-
nar Press.
Journal of Archaeological Science
Archaeometry Archaeometry
Avebury -
worth.
The date of
the West Kennet long barrow. Antiquity
ArchaeometryArchaeometry
involvement with human remains, in Forensic Tapho-
, eds.
search strategy considerations for locating missing
teeth, in
Human Remains,
-
developments in routine dating. Archaeometry
-
treatment system and the pretreatment of older and
contaminated samples. Radiocarbon
-
humations, in Death, Decay and Reconstruction, eds. A.
-
Report on Geophysical Survey, January 2001. (Centre for
Heritage.
.
C measurement of
C variations from
. Radiocarbon
1955–56
on Overton Hill, Avebury. Wiltshire Archaeological and
Natural History Magazine
Landscape
-
. Radiocarbon
to understanding mortuary behaviour, in Advances in
Perspectives
Long Cairn of the Cotswold-Severn Group-
lish Heritage.
Windmill Hill and Avebury-
don Press.
C data. Ra-
diocarbon
A computer program for ra-
diocarbon age calculation. Radiocarbon
C data base
C age calibration program.
Radiocarbon
-
. Radiocarbon
Kennet revisited. Oxford Journal of Archaeology 5,
long barrow at West Kennet, Wiltshire. Archaeologia
facility. Nuclear Instruments and Methods in Physics
Research B
preparation for Accelerator-based radiocarbon dating.
Journal of Archaeological Science
Procedures for comparing
Archaeometry
The West Ken-
56
Date of the West Kennet Long Barrow
-
Oxford Journal of Archaeology
Wiltshire Archaeological and Natural History Magazine
Sacred Mound, Holy Rings. Silbury Hill and
Complex in North Wiltshire
-
events, processes and changing worldviews from the
thirty-eighth to the thirty-fourth centuries cal. in
southern Britain.
