Environmental Geochemistry and Health (2006) 28:421–430
Springer 2006

DOI 10.1007/s10653-006-9041-5

Human teeth as historical biomonitors of environmental and dietary lead: some

lessons from isotopic studies of 19th and 20th century archival material

J.G. Farmer1,4, A.B. MacKenzie2 & G.H. Moody3

1
School of GeoSciences, University of Edinburgh, Edinburgh, EH9 3JJ, Scotland, UK
2
Scottish Universities Environmental Research Centre, East Kilbride, G75 0QF, Scotland, UK
3
Edinburgh Dental Institute, University of Edinburgh, Edinburgh, EH3 9HA, Scotland, UK;
4
Author for correspondence (tel.: +44-131-650-4757; fax: +44-131-650-4757;
e-mail: J.G.Farmer@ed.ac.uk)

Received 25 October 2005; Accepted 6 February 2006

Key words: biomonitor, human, lead, lead isotopes, teeth

Abstract
The lead isotopic composition of various sections (crown, crown base, root) of teeth was determined in
specimens collected from 19th century skulls preserved in museum collections and, upon extraction or
exfoliation, from humans of known ages residing in Scotland in the 1990s. For most 20th century samples,
calculation of accurate crown-complete or root-complete dates of tooth formation ranging from the 1920s
to the 1990s enabled comparison of 206Pb/207Pb ratios for teeth sections (crown base root) with corre-
sponding decadally averaged data for archival herbarium Sphagnum moss samples. This showed that the
teeth sections had been significantly influenced by incorporation of non-contemporaneous (more recent)
lead subsequent to the time of tooth formation, most probably via continuous uptake by dentine. This
finding confirmed that separation of enamel from dentine is necessary for the potential of teeth sections as
historical biomonitors of environmental (and dietary) lead exposure at the time of tooth formation to be
realised. Nevertheless, the mean 19th century value of 1.172±0.007 for the 206Pb/207Pb ratio in teeth was
very similar to the corresponding mean value of 1.173±0.004 for 19th century archival moss, although
relative contributions from environmental sources – whether direct, by inhalation/ingestion of dust con-
taminated by local lead smelting (206Pb/207Pb1.17) and coal combustion (206Pb/207Pb1.18) emissions, or
indirect, through ingestion of similarly contaminated food – and drinking/cooking water contaminated by
lead pipes of local origin, cannot readily be determined. In the 20th century, however, the much lower
values of the 206Pb/207Pb ratio (range 1.100–1.166, mean 1.126±0.013, median 1.124) for the teeth collected
from various age groups in the 1990s reflect the significant influence of imported Australian lead of lower
206
Pb/207Pb ratio (1.04) and released to the environment most notably through car-exhaust emissions
arising from the use of alkyl lead additives (206Pb/207Pb1.06–1.09) in petrol in the U.K. from ca. 1930 until
the end of the 20th century.

Introduction Needleman et al. 1979; Paterson et al. 1988;

Human teeth have often been used to monitor
human exposure to the harmful heavy element
lead. Studies have ranged from the deliberate use
of shed deciduous teeth to monitor the extent
of childhood exposure in recent decades (e.g.
Fergusson et al. 1989; Tsuji et al. 2001) to the
opportunistic use of permanent teeth found at
ancient burial sites to investigate the degree of
human exposure up to several thousand years ago
(e.g. Patterson et al. 1991; Budd et al. 1998, 2000,
2004). In addition to concentration, the isotopic
422 j.g. farmer et al.
composition of the lead present in teeth has
sometimes been used to estimate the contributions
of different sources and pathways of exposure to
lead (e.g. Alexander et al. 1993; Delves & Camp-
bell 1993; Farmer et al. 1994; Gulson & Wilson
1994; Gulson 1996; Åberg et al. 1998, 2001; Budd
et al. 1998, 2004; Yoshinaga et al. 1998). In Brit-
ain, variations in stable lead isotope ratios, in
particular the 206Pb/207Pb atom ratio, have been
exploited because of the considerable interest,
especially during the 1980s and 1990s, in evaluat-
ing the relative magnitude of contributions
from petrol lead (206Pb/207Pb = 1.06–1.09) and
drinking water in contact with lead pipes
(206Pb/207Pb = 1.16–1.18) (e.g. Alexander et al.
1993). This difference in the 206Pb/207Pb ratio
arose from the distinctiveness of the isotopic
signature of the original sources of the lead used,
e.g. the influence of the comparatively low 206Pb-
content Australian ores (206Pb/207Pb = 1.04)
used, along with British Columbian ores
(206Pb/207Pb = 1.16), in the manufacture of petrol
lead additives from the 1920s until their complete
withdrawal in 2000 and, at the other extreme, the
typical 1.16–1.18 range for the 206Pb/207Pb ratio of
British and other ores used in the manufacture of
many products, including lead pipes for domestic
plumbing systems, in Victorian times (Campbell &
Delves 1989; Sugden et al. 1993; Farmer et al.
2000). Such differences in the lead isotopic com-
position of various sources of lead release to the
atmosphere (e.g. car exhausts, smelting, combus-
tion of coal (206Pb/207Pb1.18)) have also con-
tributed to temporal variations in the 206Pb/207Pb
ratio of atmospherically deposited lead, e.g. as
preserved in the accumulating layers of lake sedi-
ments (e.g. Farmer et al. 1996; Eades et al. 2002),
peat bogs (e.g. Farmer et al. 1997) and peaty soils
(Farmer et al. 2005) and in collected specimens of
vegetation, such as moss, now stored in herbaria
(e.g. Farmer et al. 2002).
Notwithstanding the successful use of lead iso-
topic composition in both source apportionment
studies of lead in modern teeth (e.g. Alexander
et al. 1993; Delves & Campbell 1993; Farmer et al.
1994; Gulson 1996) and investigations of the spe-
cific source of lead in ancient teeth (e.g. Budd et al.
1998, 2004), however, it has long been recognised
that there can be difficulties in interpretation of
tooth lead data. These may be related to the po-
sition in the mouth of the particular tooth selected,
the part (e.g. whole, crown, root) and chemical
component (e.g. enamel, dentine) of the tooth
analysed and, for teeth not specifically collected
for analysis at the time of exfoliation, extraction or
death, the effects of possible contamination or
alteration during storage or burial (Delves et al.
1982; Purchase & Fergusson 1986; Fergusson &
Purchase 1987; Paterson et al. 1988; Bercovitz &
Laufer 1991; Budd et al. 1998, 2000, 2004; Lee
et al. 1999; Arora et al. 2004). Although lead in
tooth enamel is thought to reflect the presence of
lead at the time of tooth formation (Hillson 1996)
concerns have been raised at various times about
the possibility of subsequent uptake of lead ions
(e.g. from food and water) into surface enamel
(e.g. Purchase & Fergusson 1986; Fergusson &
Purchase 1987; Budd et al. 1998, 2004; Brown
et al. 2004). Tooth dentine, which is in direct
contact with blood, is thought to reflect exposure
to lead during the lifetime of the tooth (Needleman
et al. 1979; Purchase & Fergusson 1986; Fergusson
& Purchase 1987; Paterson et al. 1988; Fergusson
et al. 1989; Tsuji et al. 2001). The purposes of the
particular study and the origins and nature of the
material available must be taken into account.
In seeking to investigate the use of human tooth
lead as a historical biomonitor of environmental
and dietary lead exposure, via isotopic analysis of
19th and 20th century archival material, we here
report previously unpublished tooth lead data and
compare them with our extensive database of the
lead isotopic composition of the Scottish envi-
ronment during the 19th and 20th centuries.
Materials and methods
19th century teeth
Specific parts of teeth were provided from 25
adults in total. Five samples, consisting of crown
slices/chips, came from molar teeth in five Scottish
skulls held in the Turner Collection, a human
cranial collection in the Department of Anatomy,
Medical School, University of Edinburgh, which
also provided crown slices/chips of five Inuit molar
teeth collected in 1896. The Royal College of
Surgeons of Edinburgh permitted samples (discs
taken 2 mm up from the apex of the tooth root) to
be cut from teeth (67% of which were bicuspid) in
15 Scottish skulls held in its Greig Collection. It
 human teeth as historical biomonitors of environmental and dietary lead 423
was a condition of sampling that there be minimal
disturbance to the appearance of the museum
specimens and collection from the root also re-
duced the risk of surface contamination arising
from display under museum conditions. If the age
of the person at death was known, then the posi-
tion of the tooth in the mouth enabled dating of
the (root complete) formation of that part of the
tooth to within ±10 years for 19th century teeth.
20th century teeth
Samples from adult teeth, extracted at dental
surgery from people currently living in Scotland,
were provided in three batches. The first, collected
in 1993, consisted of slices through the crowns of
17 teeth from 17 adults aged 19–64 years. The
second, collected in 1994, consisted of discs taken
2 mm up from the apex of the tooth root in 40
teeth (70% of which were molar) from 39 adults
aged 18–69 years. This enabled dating of the (root
complete) formation of that part of the tooth to
±1 year. The third, collected in 1995, consisted of
2-mm-thick discs taken from the area adjacent to
the tooth neck, near the base of the crown, in 52
teeth (75% of which were molar) from 50 adults
aged 20–75 years. This was thought to minimise
the possibility of surface contamination of the
crown enamel and enabled dating of the (crown
complete) formation of that particular part of the
tooth to ±1 year.
Children’s extracted deciduous teeth were pro-
vided in two batches. The first of these, collected in
1993, consisted of slices through the crowns of
seven teeth from seven children aged 5–12 years.
The second, collected in 1994, consisted of discs
taken 2 mm up from the apex of the tooth root in
10 teeth from 8 children aged 3–7 years. This en-
abled calculation of the date of (root complete)
formation of that part of the tooth to ±0.5 year.
Preparation of teeth for analysis
The sections of modern 20th century teeth were
soaked in 10% v/v hydrogen peroxide in 5 mL
beakers overnight to remove blood and fat. The
19th century samples were not so treated as any
blood and fat had long since turned to dust and
they were too brittle and fragile for this treat-
ment. All sections, 19th and 20th century, were
then rinsed well with deionised water, heated in
2% v/v Decon 90 solution in a water bath at
80 C for 30 min, and then rinsed six times with
MilliQ deionised water, before drying in a drying
cabinet at 30 C for 3–4 days. Each section was
weighed into a 5 mL beaker to which 1 mL 8 M
nitric acid was added before heating at 80 C on
a hotplate to enable dissolution and evaporation
to dryness. The residues were ashed at 460 C for
16 h in a muffle furnace before redissolution in
0.5 mL 8 M nitric acid and repetition of the
process. The resultant residues were dissolved in
0.2 mL 16 M nitric acid and transferred by
Pasteur pipette to 10 mL volumetric flasks. After
addition of 2 mL of lead-free acidified 10% m/v
ammonium citrate solution and adjustment of
pH to 4.5 using concentrated ammonia and nitric
acid as required, 1 mL 10% ammonium pyrrol-
idine dithiocarbamate was added to each to
complex the lead as Pb(II) (Delves et al. 1982).
This was extracted by adding 1 mL methyl
isobutyl ketone to each and shaking for 1 min.
Once the phases had separated, the organic layer
was transferred by Pasteur pipette into a clean
5 mL beaker and evaporated slowly at 40–50 C
on a hotplate, after which the residue was re-
dissolved in 2% nitric acid and the volume made
up to 5 mL in a volumetric flask.
Analysis
Lead concentrations were determined by atomic
absorption spectrometry with a Pye Unicam SP9-
800 spectrometer at 217.0 nm, using a lead hollow
cathode lamp (5 mA), an air-acetylene flame plus
slotted tube atom trap for enhancement of sensi-
tivity, a deuterium arc background correction sys-
tem, and a spectral bandpass setting of 0.5 nm. Lead
calibration standards were prepared by dilution of a
1000 mg L)1 standard solution and sample solution
concentrations calculated from the linear portion of
the absorbance–concentration calibration curve,
prior to conversion to tooth lead concentrations
expressed in mg kg)1 dry tooth. A mean lead con-
centration of 2.8±0.9 mg kg)1 was obtained for the
IAEA Certified Reference Material Animal Bone
H-5, for which the certified lead concentration is
3.1 mg kg)1 (confidence interval 2.6–3.7 mg kg)1).
After dilution of sample solutions to
<50 lg L)1, determination of lead isotope ratios
(206Pb/207Pb, 208Pb/207Pb, 208Pb/206Pb) was carried
out using a PlasmaQuad 2 Turbo Plus ICP-MS
424 j.g. farmer et al.
(Fisons Instruments, Winsford, UK). The forward
plasma power was 1350 W, with reflected power
<5 W. Argon flow rates for cool gas, auxiliary gas
and nebuliser flow were 14 L min)1, 1 L min)1 and
0.9 mL min)1, respectively. The electron multiplier
detector voltage was 2800 V and scanning over the
mass range 201.6–211.4 a.m.u. consisted of 2000
sweeps with a dwell time per channel of 80 ls over
the 20 channels allocated per a.m.u. Correction for
mass bias was carried out using a solution of the
National Institute of Standards and Technology
(NIST) common Pb isotopic reference standard
SRM 981 (206Pb/207Pb = 1.093, 208Pb/206Pb =
2.168, 208Pb/207Pb = 2.370). Each sample solution
was analysed five times, with the mean isotope ratios
and associated standard deviations being calculated
automatically by the IBM PC-AT data handling
system. The average precision on measurements of
the above lead isotopic ratios by ICP-MS was
±0.5%. For the IAEA Animal Bone H-5, which is
not certified for lead isotopic composition, mean
values of 1.127±0.004, 2.411±0.009 and
2.138±0.006 were obtained for the 206Pb/207Pb,
208
Pb/207Pb and 208Pb/206Pb ratios, respectively.
Results
The lead concentration and 206Pb/207Pb data for
19th century teeth (crown slices in the case of the
Inuit teeth and the Scottish teeth from the Turner
collection; root disc samples in the case of the
206
Table 1. Summary of lead concentrations and Pb/207Pb ratios for 19th century teeth.
Collection n Pb concn. range mg kg)1 Pb concn. mean mg kg)1 (±1SD)
Inuita 5 1.6–12.8 6.7±4.6
Turnera 5 1.3–8.6 4.6±3.0
Greigb 15 0.9–27 8.8±7.7
Turnera+Greigb 20 0.9–27 7.7±7.0
a
Crown slices from teeth in Turner Colletion.
b
Root discs from teeth in Greig Collection.
206
Table 2. Summary of lead concentrations and Pb/207Pb ratios for 20th century teeth collected in 1993a.
Age group (years) n Pb concn. range mg kg)1 Pb concn. mean mg kg)1 (±1SD)
64 2 14.1–17.3 15.7±2.3
30–39 6 0.8–6.2 2.4±2.1
19–24 9 0.3–3.7 1.4±1.3
5–12 7 2.2–4.9 3.7±1.2
a
Crown slices from extracted or exfoliated teeth.
Scottish teeth from the Greig Collection) are
summarised in Table 1. The mean value of the
206
Pb/207Pb ratio for the Turner and Greig col-
lection teeth was 1.172±0.007 (±1SD).
The lead concentration and 206Pb/207Pb data for
20th century teeth (crown slices) collected from
people in 1993 are summarised for four different
age groups (64, 30–39, 19–24 and 5–12 years) in
Table 2. Lead concentrations were markedly
greater for the two oldest individuals born ca.
1930, but the mean values of the 206Pb/207Pb ratio
for the different age groups differed only slightly,
from 1.133±0.021 for the oldest group to
1.119±0.009 for the youngest category (deciduous
teeth of children).
The lead concentration and 206Pb/207Pb data for
sections of 20th century teeth (root disc in the case
of those collected from people in 1994 and crown
base disc in the case of those collected from people
in 1995) are summarised for eight different decades
(1920s to 1990s), with respect to the time of for-
mation of the different teeth, in Table 2. Lead
concentrations were markedly greater for the
1920s and 1930s, averaging 20 mg kg)1, and
decreased steadily to £1.1 mg kg)1 in the 1990s.
The mean values of the 206Pb/207Pb ratio for the
different decades decreased from 1.139±0.012 for
the 1920s and 1.136±0.002 for the 1930s to a
minimum value of 1.119±0.010 in the 1970s, be-
fore increasing to 1.139±0.009 in the 1990s.
The 206Pb/207Pb data for all Scottish teeth
samples are plotted against the corresponding
206
Pb/207Pb range 206
Pb/207Pb mean (±1SD)
1.157–1.169 1.164±0.005
1.159–1.175 1.167±0.006
1.156–1.181 1.173±0.007
1.156–1.181 1.172±0.007
206
Pb/207Pb range 206
Pb/207Pb mean (±1SD)
1.118–1.148 1.133±0.021
1.111–1.153 1.132±0.015
1.104–1.130 1.120±0.007
1.106–1.131 1.119±0.009
 human teeth as historical biomonitors of environmental and dietary lead 425
208
Pb/206Pb data (n = 141) in the three-isotope
plot of Figure 1. The best straight line through the
points is defined by 208Pb/206Pb = 2.983–
0.735(206Pb/207Pb), with r = )0.560. Also plot-
ted are mean values of the corresponding ratios
for: Australian lead ore (Chow et al. 1975); Aus-
tralian-lead-influenced leaded additives in petrol
on sale in Scotland from 1989 to 1998 resulting in
car-exhaust emissions of lead to the atmosphere
(Farmer et al. 2000); indigenous lead ore from
Leadhills/Wanlockhead in Scotland (Moorbath,
1962; Sugden et al. 1993); drinking water with
domestic lead water pipes (Sugden et al. 1993;
Farmer et al. 1994); and Scottish coal (Farmer
et al. 1999) the combustion of which resulted in
emissions of lead to the atmosphere.
Discussion
The value of 1.172±0.007 for the mean
206
Pb/207Pb ratio for 19th century Scottish adult
teeth (Turner+Greig collection, n = 20, Table 1)
is in agreement with the corresponding mean value
of 1.173±0.004 for 19th century archival Scottish
Sphagnum moss samples (n = 34) (Farmer et al.
2002). The latter, which is itself in agreement with
other proxy values for atmospheric lead in 19th
century Scotland, for example from dated om-
brotrophic peat bogs (Farmer et al. 1997), repre-
sents the average 206Pb/207Pb ratio of lead emitted
to the atmosphere primarily via the smelting of
indigenous lead ores and the combustion of coal
(Farmer et al. 1999). In the case of the lead in
human teeth, however, it cannot be stated with
certainty that the source (and route of uptake) of
the lead was environmental, either directly via
inhalation of atmospheric lead emitted by smelt-
ing/coal combustion or indirectly via ingestion of
atmospheric lead deposited upon, or taken up by
the roots of, food plants. There could also have
been a contribution from the direct ingestion of
drinking (or cooking) water contaminated with
lead as a result of the contact of soft, acidic water

2.30
Australian
lead ore

2.25

2.20
Pb/206Pb

2.15 Scottish
leaded petrol
208

2.10
Drinking water in
contact with lead pipes
Scottish lead ore (L/W) Scottish
coal
2.05

2.00
1.04 1.06 1.08 1.10 1.12 1.14 1.16 1.18 1.20
206 207
Pb/ Pb

Fig. 1. A three-isotope plot of 208Pb/206Pb versus 206Pb/207Pb for individual 19th century crown slices (adult, n = 5, d), 20th cen-
tury crown slices (adult and children, n = 24, s), 19th century root discs (adult, n = 15, .), 20th century root discs (adult and
children, n = 48, ,) and 20th century crown base discs (adult, n = 52, h), along with mean values for Australian lead ore, leaded
additives in petrol on sale in Scotland from 1989 to 1998 (Farmer et al. 2000), indigenous lead ores from Leadhills/Wanlockhead
(L/W) in Scotland (Moorbath 1962; Sugden et al. 1993), water in contact with domestic lead water pipes (Sugden et al. 1993;
Farmer et al. 1994), and Scottish coal (Farmer et al. 1999).
426 j.g. farmer et al.
206
Table 3. Summary of lead concentrations and Pb/207Pb ratios for sections of 20th century teeth collected in 1994a,b and 1995c.

206
Date of formation n Pb concn. range Pb concn. mean Pb/207Pb range 206
Pb/207Pb mean
of tooth mg kg)1 mg kg)1 (±1SD) (±1SD)

1920s
Roota – – – – –
Crownc 2 21.5–22.1 21.8±0.4 1.130–1.147 1.139±0.012
All 2 21.5–22.1 21.8±0.4 1.130–1.147 1.139±0.012
1930s
Roota 1 6.1 6.1 1.138 1.138±0.003
Crownc 3 14.7–35.9 24.4±10.7 1.133–1.138 1.135±0.003
All 4 6.1–35.9 19.9±12.7 1.133–1.138 1.136±0.002
1940s
Roota 1 19 19 1.127 1.127±0.004
Crownc 1 3.2 3.2 1.118 1.118±0.002
All 2 3.2–19 11.1±11.2 1.118–1.127 1.123±0.006
1950s
Roota 4 4.2–6.3 5.3±0.9 1.123–1.140 1.132±0.007
Crownc 15 0.1–13.8 4.7±3.9 1.108–1.143 1.124±0.010
All 19 0.1–13.8 4.8±3.4 1.108–1.143 1.125±0.010
1960s
Roota 7 1.3–7.4 3.8±2.4 1.110–1.134 1.124±0.010
Crownc 13 0.8–7.9 3.1±2.0 1.100–1.166 1.127±0.018
All 20 0.8–7.9 3.3±2.1 1.100–1.166 1.126±0.015
1970s
Roota 11 0.3–5.8 2.2±1.7 1.111–1.143 1.123±0.011
Crownc 12 0.2–6.7 2.2±1.9 1.102–1.125 1.116±0.007
All 23 0.2–6.7 2.2±1.8 1.102–1.143 1.119±0.010
1980s
Roota 14 0.5–6.6 1.7±1.8 1.106–1.157 1.125±0.015
Crownc 6 0.4–4.7 2.4±1.8 1.111–1.137 1.125±0.009
All 20 0.4–6.6 1.9±1.8 1.106–1.157 1.125±0.013
1990s
Roota 2 0.5–0.8 0.7±0.2 1.134–1.146 1.140±0.008
Rootb 8 0.3–1.1 0.4±0.3 1.124–1.150 1.139±0.009
Crownc – – – – –
All 10 0.3–1.1 0.5±0.3 1.124–1.150 1.139±0.009
a
Root discs.
b
Root discs (deciduous); all from extracted or exfoliated teeth.
c
Crown base discs from extracted teeth.

with the lead pipes of Victorian plumbing systems
manufactured from indigenous lead ores
(206Pb/207Pb = 1.16–1.18).
On the basis of the decline in 206Pb/207Pb ratios
from the mean value for the 19th century Scottish
adult teeth (1.172±0.007 for the Turner+Greig
collections, Table 1) to the mean values for the
20th century Scottish adult teeth (1.126±0.013 for
crown slices from all adults, n = 17, in Table 2)
collected in 1993, the influence of lead of much
lower 206Pb/207Pb ratio (e.g. the 1.04 of Australian
lead, as used in combination with other sources in
the leaded petrol additives introduced from the
1930s) is apparent. Previously, a mean value of
1.124±0.014 was found for teeth from four young
adults in Edinburgh (Farmer et al. 1994). For the
children’s teeth (n = 7) collected in 1993, mean
values of 3.7±1.2 mg kg)1 and 1.119±0.009 for
lead concentration and 206Pb/207Pb ratio (Table 2),
respectively, were in excellent agreement with the
corresponding mean values of 3.2±1.2 mg kg)1
and 1.131±0.015 reported by Farmer et al. (1994)
for whole teeth (n = 7) collected from ‘low-lead
exposure’ children in Edinburgh earlier in the
1990s. At that time, the mean 206Pb/207Pb ratio
was considered to be intermediate between the
then observed values for leaded petrol
(1.075±0.013) and tap water in contact with
 human teeth as historical biomonitors of environmental and dietary lead 427
1.20
1.18
1.16
Pb/207Pb
206
1.14
1.12
1.10
1800 1850
Fig. 2. A comparison of temporal trends in the decadal average 206Pb/207Pb ratios (±1SD) for teeth (root and crown base discs,
Table 3, d; time of root complete or crown complete formation) and Scottish Sphagnum moss samples (s) (Farmer et al. 2000)
during the 19th and 20th centuries.
leaded pipes (1.160±0.012) and comparable with
that reported for lead in food (1.12–1.13).
In using teeth, the question arises of the
appropriateness of different parts for use as bio-
monitors of value over varying lengths of time in
studies aimed at source identification or appor-
tionment. Both from our previous work (Farmer
et al. 1994) and the comparison of crown slice data
in this study, it seemed clear to us that separation
into enamel and dentine components would be
beneficial. We were concerned, however, that
crown enamel, although potentially easier to relate
to a specific period of tooth formation and com-
pletion, might have been subject to post-eruption
oral exchange and uptake of lead from lead-con-
taminated food and water (e.g. Fergusson & Pur-
chase 1987). This is why we decided, at least for
the teeth collected in 1994, to try to sample from a
section of tooth (root) equally easy to relate to a
specific period of tooth formation and completion
but much less subject to surficial exchange and
uptake. In the subsequent attempt, in 1995, we
hoped that the collection of discs from near the
base of the crown, near the neck of the tooth,
would also minimise lead contamination of surface
enamel. As, in both cases, our attempts to separate
enamel and dentine components by a simple and
1900 1950 2000
date
previously successful physical separation method
(Farmer et al. 1994) failed, it is the suitability and
applicabilty of whole root discs and crown base
discs that comes under scrutiny below.
Table 3 demonstrates general agreement
between the 206Pb/207Pb data for root discs and
crown base discs over the decades of the 20th
century for which data were available. Figure 2,
therefore, compares the decadally averaged tem-
poral trend in the combined 206Pb/207Pb data for
the root and crown base disc samples (Table 3) for
the 20th (and 19th – root disc only) centuries with
that subsequently established for archival herbar-
ium Sphagnum moss samples from Scotland
(Farmer et al. 2002). There is good agreement for
the 19th century data, where the decadal values
ranged from 1.167 to 1.176 for teeth and 1.169 to
1.176 for moss (Figure 2). However, as is clear
from the corresponding comparison of 20th cen-
tury 206Pb/207Pb data, most notably from the
1920s to 1970s (Figure 2), there was a strong
subsequent contributory influence from lead that
was non-contemporaneous (more recent) to the
time of tooth formation and completion, possibly
also reflected in the increased lead concentration
with age of tooth (Table 3). Bearing in mind the
practical difficulty experienced in separating
428 j.g. farmer et al.
enamel from dentine, this most probably occurred
via dentine incorporation of lead circulating in the
blood. While the bulk of the root section analysed
would have been primary dentine, the pulpal cav-
ity at the centre of the tooth where lead can enter
from the bloodstream is lined with circumpulpal or
secondary dentine, which has a higher organic
content and tends to have a much higher lead
concentration (w/w) than in primary dentine or
enamel (e.g. Fergusson & Purchase 1987). The lead
isotopic data for the root disc samples (Table 3),
therefore, probably reflect some more recent
accumulation of lead in the secondary dentine
surrounding the pulp cavity with age through the
years. Similarly, the crown base discs were no
more successful in terms of generating specific
date-related lead isotopic data that were uninflu-
enced by subsequent exposure to and incorpora-
tion of lead, suggesting that they too have been
affected to a similar extent by accumulation of lead
deposited in coronal dentine in the years after
crown completion. It is noticeable, however, that
the difference between teeth and moss 206Pb/207Pb
data for samples from younger people with more
recent dates of tooth formation (1980s to 1990s)
was much lower (Figure 2), there being less time
for subsequent accumulation of lead of potentially
different 206Pb/207Pb ratio.
The lessons to be learned from this tooth study
have emerged as a consequence of the establish-
ment of the trends in the 206Pb/207Pb ratio of
atmospheric lead in the Scottish environment over
the past two centuries. To some extent, these
reinforce the recommendations or findings of other
subsequent studies of lead in teeth. In particular,
if, as in this historical biomonitoring study, the
isotopic composition of lead that prevailed in the
environment (and to which humans were exposed)
over a short interval of time (i.e. corresponding to
tooth formation) is sought, then, for adult teeth,
the crown enamel (after removal of the surface
enamel) should be preferred for analysis (Budd
et al. 1998, 2000, 2004), as dentine integrates
exposure to lead over the lifetime of the tooth.
Removal of the surface enamel removes contri-
butions resulting from the subsequent ion-ex-
change uptake of lead from contaminated drinking
water in contact with lead pipes and, in longer-
term archaeological studies, from post-burial ion-
exchange with lead in soil (Budd et al. 1998, 2000,
2004). Of course, where there has been exposure to
lead from the consumption of drinking water
contaminated by lead pipes during the formation
of the tooth, the lead isotopic composition of the
treated crown enamel will not necessarily reflect
that of the environment (e.g. the atmosphere) at
that time. Furthermore, the dietary intake of lead
at that time also need not necessarily reflect the
prevailing isotopic composition of atmospheric
lead as dietary sources such as food plants may
have accumulated previously deposited lead from
soil.
Nevertheless, in these teeth samples, it is clear
that direct (inhalation/ingestion) or indirect (die-
tary) exposure to environmental lead, in addition
to the (unknown) possibility of exposure to lead
present in drinking water from contact with lead
pipes, has been significant. This is demonstrated
by the mean value of 1.126±0.013 (median 1.124)
for the 206Pb/207Pb ratio of teeth (n = 141) col-
lected from various age groups during the 1990s
(Tables 2, 3), reflecting the influence of releases of
imported Australian lead of lower 206Pb/207Pb ra-
tio (1.04) to the environment primarily through
car-exhaust emissions arising from the use of alkyl
lead additives (206Pb/207Pb1.06–1.09) in petrol.
The mean value of 1.126 is intermediate between
that of 1.076 for leaded petrol (Farmer et al. 2000)
and the 1.170 for local indigenous lead ore and
1.181 for Scottish coal (Farmer et al. 1999). Al-
though source apportionment calculation was not
the purpose of this study, the demonstration of
significant contributions to human exposure from
the petrol-derived source and pathways of lead has
been borne out by the remarkable observed decline
in blood lead not only in the U.K. but elsewhere in
the world where unleaded petrol has been intro-
duced and leaded petrol banned (Thomas et al.
1999). Measures to reduce exposure to lead from
other sources, for example from water and
adventitious sources such as leaded paint), have
also been important. On the basis of our previous
tooth study (Farmer et al. 1994), the presence of a
few outliers in the 208Pb/206Pb versus 206Pb/207Pb
plot of Figure 1 could possibly reflect the influence
of exposure to leaded paint.
Conclusions
The influence of changes in environmental sources
of lead of varying isotopic composition during the
 human teeth as historical biomonitors of environmental and dietary lead 429
20th century could readily be discerned in the lead
isotopic composition of sections (crown, crown
base, root disc) of teeth of known date of forma-
tion. The demonstrable non-contemporaneous
uptake of lead, however, probably via incorpora-
tion in dentine, serves to preclude the use of such
teeth sections as accurate biomonitors (for periods
corresponding to the time of tooth formation) of
(i) the prevailing isotopic composition of atmo-
spheric lead and (ii) environmental and dietary
exposure to lead. Accordingly, the findings indi-
rectly support the use of the enamel component of
teeth in historical biomonitoring studies, and the
use of the dentine component of short-lived
deciduous teeth in surveys of recent childhood
exposure to lead.
Acknowledgements
We thank the Department of Anatomy and the
Edinburgh Dental Institute, University of Edin-
burgh, and the Royal College of Surgeons of
Edinburgh for provision of teeth samples,
K. Cotton, J. Hitchen and E. Kane for chemical
preparation of samples in the laboratory and
K. Sampson for ICP-MS analysis.
References
Åberg G, Fosse G, Stray H. 1998 Man, nutrition and mobility:
a comparison of teeth and bone from the Medieval era and
the present from Pb and Sr isotopes. Sci Total Environ 224,
109–119.
Åberg G, Charalampides G, Fosse G, Hjelmseth H. 2001 The
use of Pb isotopes to differentiate between contemporary and
ancient sources of pollution in Greece. Atmos Environ 35,
4609–4615.
Alexander LM, Heaven A, Delves HT, Moreton J, Trenouth
MJ. 1993 Relative exposure of children to lead from dust and
soil. Arch Environ Health 48, 392–400.
Arora M, Chan SWY, Kennedy BJ, Sharma A, Crisante D,
Walker DM. 2004 Spatial distribution of lead in the roots of
human primary teeth. J Trace Elem Med Biol 18, 135–139.
Bercovitz K, Laufer D. 1991 Age and gender influence on lead
accumulation in root dentine of human permanent teeth.
Arch Oral Biol 36, 671–673.
Brown CJ, Chenery SRN, Smith B, Mason C, Tomkins A,
Roberts GJ, Sserunjogi L, Tiberindwa JV. 2004 Environ-
mental influences on the trace element content of teeth –
implications for disease and nutritional status. Arch Oral Biol
49, 705–717.
Budd P, Montgomery J, Cox A, Krause P, Barreiro B, Thomas
RG. 1998 The distribution of lead within ancient and modern
human teeth: implications for long-term and historical
exposure monitoring. Sci Total Environ 220, 121–136.
Budd P, Montgomery J, Evans J, Barreiro B. 2000 Human
tooth enamel as a record of the comparative lead exposure of
prehistoric and modern people. Sci Total Environ 263, 1–10.
Budd P, Montgomery J, Evans J, Trickett M. 2004 Human lead
exposure in England from approximately 5500 BP to the 16th
century AD. Sci Total Environ 318, 45–58.
Campbell MJ, Delves HT. 1989 Accurate and precise determi-
nation of lead isotopic ratios in clinical and environmental
samples using inductively coupled plasma mass spectrometry.
J Anal Atomic Spectrometry 4, 235–256.
Chow TJ, Snyder CB, Earl JL. 1975. Isotope ratios of lead as
pollutant source indicators. In Isotope Ratios as Pollutant
Source Indicators. Vienna: IAEA, pp. 95–107.
Delves HT, Campbell MJ. 1993 Identification and apportion-
ment of sources of lead in human tissue. Environ Geochem
Health 15, 75–83.
Delves HT, Clayton BE, Carmichael A, Bubear M, Smith M.
1982 An appraisal of the analytical significance of tooth-lead
measurements as possible indices of environmental exposure
of children to lead. Ann Clin Biochem 16, 329–337.
Eades LJ, Farmer JG, MacKenzie AB, Kirika A, Bailey-Watts
AE. 2002 Stable lead isotopic characterisation of the histor-
ical record of environmental lead contamination in dated
freshwater lake sediment cores from northern and central
Scotland. Sci Total Environ 292, 55–67.
Farmer JG, Sugden CL, MacKenzie AB, Moody GH, Fulton
M. 1994 Isotopic ratios of lead in human teeth and sources of
lead exposure in Edinburgh. Environ Technol 15, 593–599.
Farmer JG, Eades LJ, MacKenzie AB, Kirika A, Bailey-Watts
AE. 1996 Stable lead isotope record of lead pollution in Loch
Lomond sediments since 1630 A.D. Environ Sci Technol 30,
3080–3083.
Farmer JG, MacKenzie AB, Sugden CL, Edgar PJ, Eades LJ.
1997 A comparison of the historical lead pollution records in
freshwater lake sediments and peat cores from central
Scotland. Water Air Soil Poll 100, 253–270.
Farmer JG, Eades LJ, Graham MC. 1999 The lead content and
isotopic composition of British coals and their implications
for past and present releases of lead to the U.K. environment.
Environ Geochem Health 21, 257–272.
Farmer JG, Eades LJ, Graham MC, Bacon JR. 2000 The
changing nature of the 206Pb/207Pb isotopic ratio of lead in
rainwater, atmospheric particulates, pine needles and leaded
petrol in Scotland 1982–1998. J Environ Monitor 2, 49–57.
Farmer JG, Eades LJ, Atkins H, Chamberlain DF. 2002
Historical trends in the lead isotopic composition of archival
Sphagnum mosses from Scotland (1838–2000). Environ Sci
Technol 36, 152–157.
Farmer JG, Graham MC, Vinogradoff SI, Bacon JR, Dunn
SM, MacKenzie AB. 2005 Isotopic characterisation of the
historical lead deposition record at Glensaugh, an organic-
rich, upland catchment in rural N.E. Scotland. Sci Total
Environ 346, 121–137.
Fergusson JE, Purchase NG. 1987 The analysis and levels of
lead in human teeth: a review. Environ Poll 46, 11–44.
Fergusson JE, Kinzett N, Ferguson DM, Horwood LT. 1989 A
longitudinal study of dentine lead levels, intelligence, school
performance and behaviour: the measurement of dentine
lead. Sci Total Environ 80, 229–241.
430 j.g. farmer et al.
Gulson BL. 1996 Tooth analyses of sources and intensity of
lead exposure in children. Environ Health Persp 104, 306–312.
Gulson BL, Wilson D. 1994 History of lead exposure in
children revealed from isotopic analyses of teeth. Arch
Environ Health 49, 279–283.
Hillson S. 1996 Dental Anthropology, Cambridge: Cambridge
University Press.
Lee KM, Appleton J, Cooke M, Keenan F, Sawicka-Kapusta
K. 1999 Use of laser ablation inductively coupled plasma
mass spectrometry to provide element versus time profiles in
teeth. Anal Chim Acta 395, 179–185.
Moorbath S. 1962 Lead isotope abundance studies on mineral
occurrences in the British Isles and their geological signifi-
cance. Philos Trans Royal Soc London A 254, 295–360.
Needleman HL, Gunnoe C, Leviton A, Reed R, Peresie H,
Maher C, Barrett P. 1979 Deficits in psychologic and
classroom performance of children with elevated dentine
lead levels. N Eng J Med 300, 689–695.
Paterson LJ, Raab GM, Hunter R, Laxen DPH, Fulton M, Fell
GS, Halls DJ, Sutcliffe P. 1988 Factors influencing lead
concentrations in shed deciduous teeth. Sci Total Environ 74,
219–233.
Patterson C, Ericson J, Manea-Krichten M, Shirahata H. 1991
Natural skeletal levels of lead in Homo sapiens sapiens
uncontaminated by technological lead. Sci Total Environ 107,
205–236.
Purchase NG, Fergusson JE. 1986 Lead in teeth: the influence
of the tooth type and the sample within a tooth on lead levels.
Sci Total Environ 52, 239–250.
Sugden CL, Farmer JG, MacKenzie AB. 1993 Isotopic ratios of
lead in contemporary environmental material from Scotland.
Environ Geochem Health 15, 59–65.
Thomas VM, Socolow RH, Fanelli JJ, Spiro TG. 1999 Effects
of reducing lead in gasoline: an analysis of the international
experience. Environ Sci Technol 33, 3942–3948.
Tsuji LJS, Karagatzides JD, Katapatuk B, Young J, Kozlovic
DR, Hanning RM, Nieboer E. 2001 Elevated dentine-lead
levels in deciduous teeth collected from remote First Nation
communities located in the western James Bay region of
northern Ontario, Canada. J Environ Monitor 3, 702–705.
Yoshinaga J, Yoneda M, Morita M, Suzuki T. 1998 Lead in
prehistoric, historic and contemporary Japanese: stable
isotopic study by ICP mass spectrometry. Appl Geochem
13, 403–413.