The Science of the Total Environment 299 (2002) 177–190
Lead within ecosystems on metalliferous mine tailings in Wales
and Ireland
Adrian Miltona, Michael S. Johnsona, John A. Cookeb,*
a
School of Biological Sciences, Life Sciences Building, University of Liverpool, Liverpool, L69 7ZB, UK
School of Life and Environmental Sciences, George Campbell Building, University of Natal, Durban 4041, South Africa
b
Received 9 December 2001; accepted 5 June 2002
Abstract
A comparative study of the concentrations of lead in ecosystems developed on metalliferous mine tailings was
undertaken. Mine soils, vegetation, ground-dwelling invertebrates and Apodemus sylvaticus from nine abandoned
mines in Wales and a modern Irish mine site were sampled in order to evaluate and compare exposure risks to
wildlife. The mine sites had a wide range of relatively high concentrations of total lead in their tailings (from 1058
to 46 630 mg kgy1) but the extractable lead fractions were extremely variable and not clearly related or proportional
to the total values. The high soil concentrations were reflected in vegetation collected from most of the sites with the
exception of the modern mine, but there was no statistical relationship, on a site basis, between available soil lead
and that in plant leaf samples. The highest plant concentrations were found in litter, which in all but one of the
Welsh sites exceeded the threshold guideline value of 150 mg kgy1. Food-chain transfer was shown by high
concentrations of lead in invertebrates and A. sylvaticus from the abandoned Welsh mines. A highly significant
relationship existed between lead in grass and the grasshopper, Chorthippus brunneus. Adverse effects on soil
invertebrates, essential to the decomposition processes and cycling of essential nutrients, were identified as probably
the major obstacle to natural ecosystem development on the abandoned Welsh sites. Toxicological risk of lead to the
small mammals from the Welsh sites, but not the modern Irish tailings, is indicated given the high lead concentrations
in dietary items and the resultant residues in kidney with some evidence of renal oedema in animals from two sites.
The absence of a significant relationship between the estimated dietary lead concentration, calculated on a site basis,
and the total body concentration in A. sylvaticus, was attributed, in part, to the large size of the home range and the
partial feeding of individual animals off the contaminated mine site.
䊚 2002 Elsevier Science B.V. All rights reserved.
Keywords: Lead; Food chains; Ecotoxicological risk; Mine tailings; Invertebrates; Apodemus sylvaticus
*Corresponding author. Tel.: q27-312-603-192; fax: q27-312-602-029.
E-mail address: [email protected] (J.A. Cooke).
0048-9697/02/$ - see front matter 䊚 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 0 4 8 - 9 6 9 7 Ž 0 2 . 0 0 2 5 3 - X
A. Milton et al. / The Science of the Total Environment 299 (2002) 177–190
178
natural sources rarely cause such elevated levels
of lead in the substrate (Andrews et al., 1989).
This study compares the environmental significance of lead at a selection of historic and abandoned Welsh mine sites last worked between 1866
and 1924 (WDA, 1978), and a typical modern
tailings facility at a leadyzinc mine in the Republic
of Ireland (Brady, 1993). The primary aim was to
assess the potential ecotoxicological risk associated
with lead at a range of mine sites in the context
of these sites being colonised by wildlife. A second
aim was the comparison of the modern site in
Ireland with the much older abandoned sites. This
sought to inform the restoration practice currently
in operation at the modern site, in terms of the
production of a sustainable low-maintenance grassland and the potential impacts residual lead may
have in realising this goal. Cadmium was also a
significant contaminant of these wastes and will
be considered in a separate paper.
1. Introduction
The main contemporary uses of lead are battery
manufacture, cable sheathing, solders, bearings,
radiation shielding, plastics, ceramics and even
cosmetics (Stewart, 1994). As with other metals
the major increase in lead consumption and associated pollution events came with the Industrial
Revolution, and the mining and smelting of metalliferous ores has resulted in approximately 4000
km2 of agricultural land being contaminated in the
UK alone (Thornton, 1980). Concerns about the
environmental impacts of technogenic lead have
caused it to be one of the most investigated of
pollutants. Special attention has been given to the
environmental and human health issues surrounding alkyl-lead additives to motor fuels, in particular
the neurological and developmental defects exhibited by children living in areas of dense traffic
(Lansdown, 1986). Another area of environmental
concern has been the exposure of birds to lead
used as fishing weights or gun shot, sources to
which bird mortalities have been linked (Pain,
1996; Scheuhammer and Norris, 1996).
Ecosystem distribution of lead has been investigated in mining and related refining industries
(Roberts and Johnson, 1978; Andrews et al., 1989;
Purcell et al., 1992), road verges (Purcell et al.,
1992) and aquatic systems (Gerhardt, 1993). However, mine tailings, the waste from the processing
of mineral ores, represent an exceptional source of
metal contamination since other anthropogenic or
2. Materials and methods
Site selection was based on the results of an
earlier study of metal mines in Wales (WDA,
1978), using the criteria of size ()0.5 ha) and
soil levels of residual metals to ensure comparisons
across a broad range of contamination levels
(Table 1). A reference site (OS Grid Ref: SN
665734), with a similar grass flora to the Welsh
mine sites, was also selected. The modern tailings
dam, located in the Republic of Ireland, covers
Table 1
Characteristics of Welsh mine sites (WDA, 1978)
Site
name
Ordinance
survey
Grid Ref.
Date last
worked
Area
(ha)
Pba
(mg kgy1)
dry wt.
Cda
(mg kgy1)
dry wt.
Zna
(mg kgy1)
dry wt.
Cwmerfin
Grogwnion
Cwmsymlog
Frongoch
Cwmrheidol
Ystum Tuen
Rhosesmor
Trelogan
East Halkyn
SN 703824
SN 714725
SN 700837
SN 721744
SN 730782
SN 733788
SJ 214684
SJ 127802
SJ 211698
1889
1889
1901
1901
1924
1919
1866
1909
1910
0.5
1
2
4
2
0.75
0.8
2.3
0.5
7800
7100
14 000
7600
4219
7800
138 500
13 800
5500
14
10
8
6
35
18
165
131
135
2500
625
7850
7200
6600
9870
40 320
46 200
39 000
a
Total concentrations in fine tailings (-2 mm).
A. Milton et al. / The Science of the Total Environment 299 (2002) 177–190
some 78 ha to an average tailings depth of 14 m.
It is covered by a sown and managed, predominantly grass, sward dominated by red fescue (Festuca rubra), creeping bent (Agrostis stolonifera)
and white clover (Trifolium repens) (Brady, 1993).
Field sampling was conducted in the late summer and autumn of 1995. Bulk samples (500 g)
of surface soil (0–10 cm) were collected from at
least four densely vegetated areas at each mine
site, but with selectivity towards the core area so
as to avoid the margins and boundaries. Dried
samples (0.5 g) of tailings were sieved (1 mm)
and sequentially extracted with 10 ml of double
distilled water (DDW), 0.5 M ethanoic acid and
concentrated nitric acid (Davies, 1971). Soil suspensions were agitated for 1 h and centrifuged at
10 000 rev. miny1 for 30 min with each extractant
before decanting and diluting to 25 ml with DDW.
A range of vegetation samples was collected
from the Welsh mine sites, including the leaves of
dominant grasses, red fescue (Festuca rubra) and
common bent (Agrostis capillaris). In order to
determine dietary lead intake by the wood mouse,
Apodemus sylvaticus, the principal food items were
also collected including grass seeds, blackberries
(Rubus fruticosus), seeds of rowan (Sorbus aucuparia), gorse (Ulex europaeus) and sycamore
(Acer pseudoplatanus) trees, young heather (Calluna vulgaris) shoots and flowers, and a composite
sample of the leaves and flowers of young forbs
(Smal and Fairley, 1980; Hunter et al., 1987a;
Rogers and Gorman, 1995). Invertebrates were
also sampled since this provides a seasonal food
source of A. sylvaticus. The estimated dietary
composition took account of the food items available at each site and was adapted accordingly.
Invertebrates were collected using pitfall traps
containing 2% formalin positioned in each of the
vegetation zones. Traps were exposed for 12 weeks
in late summer and autumn, animals were sorted,
identified to species level and then oven-dried to
constant weight. The terrestrial gastropods, Arion
ater and Cepaea nemoralis, and adult (5th instar)
common field grasshoppers (Chorithippus brunneus), were also collected manually from each
site.
Vegetation and plant litter were collected as
small (2–15 g wet wt.) bulk samples, each com-
179
prising pooled, randomly selected material from
an area of 100 m2 to 0.5 ha according to the total
area of the mine site. Samples were oven-dried
(48 h, 80 8C), then ground to pass a 0.5-mm
screen. Gastropods were allowed to passively evacuate their alimentary tract, bulked to give three
replicates of six animals of each species per site,
and then homogenised before acid digestion.
Grasshoppers were not depurated, digested in acid
individually after oven drying. Only invertebrate
species present at several sites, and in sufficient
numbers for statistical analysis, had their metal
concentrations determined.
Adult specimens of A. sylvaticus ()15-22 g
wet wt.) were caught in late summer using baited
live traps located selectively in the densest areas
of vegetation. The sparse population density of
mammals was addressed by an intensive campaign
using 150 traps set for 2–5 days, with twice daily
inspection. Animals were killed and immediately
frozen to prevent tissue lysis and metal redistribution. The liver, kidney, muscle and bone (femur)
were analysed as well as the residual carcass from
which the gut contents were removed and discarded. Muscle tissue was manually removed from the
femurs, which were then cleaned enzymically with
papain solution and oven-dried (Pankakoski and
Hanski, 1989).
All biological samples were subjected to acid
digestion, involving overnight cold dissolution in
concentrated HNO3, followed by boiling at 120 8C
for 2 h. MERCK ‘Analar’ grade acid was used for
soil analysis, whereas ‘Aristar’ grade was used for
vegetation, invertebrate and mammal samples. Soil
extracts and acid digests were analysed for metals
using flame atomic absorption spectrometry
(AAS), matrix-matched calibration standards and
deuterium background correction for non-atomic
absorption. Seed, invertebrate and small mammal
tissue digests were analysed by graphite furnace
atomic absorption spectrometry (GFAAS), using
the L’vov platform furnace technique (Voth-Beach,
1985). Zeeman background correction and a mixed
chemical modifier of reduced palladium, magnesium nitrate and ammonium dihydrogen phosphate
solution was used to minimise interference for all
GFAAS analyses. Operational conditions are given
in more detail in Milton et al. (1998).
180
A. Milton et al. / The Science of the Total Environment 299 (2002) 177–190
Table 2
Concentration (mg kgy1 dry wt."S.E.M.) of lead in surface soils‡ from the reference and mine sites
Site
n
Double distilled
H2O
0.5 M
CH3COOH
Concentrated
HNO3
Reference
Cwmerfin
Grogwnion
Cwmsymlog
Frongoch
Cwmrheidol
Ystum Tuen
Rhosesmor
Trelogan
East Halkyn
Irish tailings
7
4
4
5
4
4
4
4
4
4
10
0.07"0.03
4.8"1.3**††
18"11**††
11"4.6**††
5.8"1.1**†
34"21***†††
24"12***†††
8.3"1.8**††
3.4"1.0*†
4.7"1.4*†
-0.05
7.6"2.0†††
5800"1788***†
7440"2866***†
4500"1292***†
1294"460***
174"83***†
558"130***
4420"649***†
5970"333***†
2510"2405***
912"167***
31"7.3†††
12 550"3620***††
8930"2634***††
24 470"12003***††
4160"1202***†
9510"3882***††
5740"811***††
46 630"11070***†††
13 780"4115***††
17 410"12849***††
1058"173***
*, ** and *** denote significant difference from reference site at P-0.05, -0.01 and -0.001, respectively. † , †† and ††† denote
significant difference from Irish tailings at P-0.05, -0.01 and -0.001, respectively. ‡ ANOVA computed on log10 transformed
data.
All glassware was acid-washed and quality
assurance was provided by running double blanks
and certified reference materials (CRMs) with all
sample batches. BCR 142 (light sandy soil), BCR
277 (estuarine sediment) and IAEA soil 7 were
included in each batch of tailings samples and
recoveries of lead were between 88 and 96% of
the certified value. The CRMs included with vegetation samples were BCR 60 (aquatic plant,
Lagarosiphon major), BCR 279 (sea lettuce, Ulva
lactuca) and NIST 1573a (tomato leaves). Recoveries of lead were between 88 and 114%. For
invertebrates, the CRM used was TORT-1 (lobster
hepatopancreas) for which lead recoveries were
between 90 and 97%. The CRMs adopted for
mammal tissues were BCR 185 (bovine liver) and
BCR 186 (pig kidney). Recoveries were between
83 and 118%.
Mammal tissue results were converted from wet
to dry weight using pre-determined conversion
factors (carcass, 2.72; kidney, 3.69; liver, 3.67;
muscle, 3.83). Total body lead concentrations
(TBC) were calculated by adding individual tissue
burdens to residual carcass values. Gastropod concentrations of lead were similarly converted to dry
weight using a factor of 6.8.
Due to variations in site area and the abundance
of animals, replication (n) for each site varied,
and is given in Tables 2–7. All data sets are
presented using arithmetic means and standard
errors and were log10 transformed and subjected
to a one-way analysis of variance (ANOVA), using
site as the effect factor. Probability data therefore
relate to analyses undertaken on normalised distributions. A Tukey’s–Kramer test was used for
comparison between sites for each sample type
(Day and Quinn, 1989). Values of half the detection limit (FAASs0.05 mg gy1) were substituted
in statistical analysis for samples with lead levels
below machine detection limits. All data from the
present study are presented on a dry weight basis
unless stated otherwise.
3. Results
3.1. Soil
Concentrations of lead in the mine wastes are
summarised in Table 2. With the exception of the
Irish tailings, all sites had significantly more water
extractable lead than the reference site (P-0.05).
For the Irish tailings, water soluble lead was below
a detection limit of 0.05 mg gy1 for all replicates
(ns10). The ethanoic- and nitric acid extractable
lead for all mine sites were significantly elevated
with respect to the reference site (P-0.001).
Similarly, most of the old Welsh mine wastes
contained more lead than the Irish tailings (P0.05), the exceptions being the ethanoic acid
extractable fraction at Frongoch, Ystum Tuen and
A. Milton et al. / The Science of the Total Environment 299 (2002) 177–190
181
Table 3
Concentration of lead in plants and litter from the reference and mine sites (mg kgy1 dry wt."S.E.M.)
Site
F. rubra‡
Agrostis spp‡
Composite
ground
cover‡
Litter‡
Composite
seeds
Reference
3.6"1.6
(3)
29"4.6***†††
(3)
48"10***†††
(3)
49"12***†††
(4)
88"30***†††
(3)
14"3.6***†††
(4)
48"5.6***†††
(3)
87"13***†††
(3)
178"46***†††
(3)
68"4.2***†††
(3)
1.3"0.2
(6)
4.3"0.2
(3)
156"41***†††
(3)
89"3.7***†††
(3)
281"69***†††
(4)
161"38***†††
(3)
21"11*††
(4)
81"7.6***†††
(3)
58"27***†††
(3)
131"36***†††
(3)
53"2.6***†††
(3)
-0.50**
(3)
2.3"0.4
(9)
87"34***†††
(9)
121"19***†††
(6)
16"2.3***†††
(8)
156"18***†††
(9)
8.3"4.5
(12)
61"10***†††
(4)
136"15***†††
(4)
177"39***†††
(4)
158"22***†††
(6)
1.4"0.2
(8)
10"1.3†††
(6)
1925"595***†††
(5)
790"285***†††
(6)
382"74***†††
(8)
896"308***†††
(6)
72"19***
(8)
346"55***†††
(6)
523"29***†††
(4)
1890"1220***†††
(4)
163"25***†
(6)
60"4.3***
(4)
2.0"0.5
(14)
6.4"1.8
(12)
2.2"0.6
(9)
2.4"0.5
(12)
46"15***†††
(9)
1.5"0.1
(3)
0.8"0.8
(4)
14"4.3**†
(9)
16"3.8**†
(12)
21"5.0***††
(12)
3.1"0.3
(9)
Cwmerfin
Grogwnion
Cwmsymlog
Frongoch
Cwmrheidol
Ystum Tuen
Rhosesmor
Trelogan
East Halkyn
Irish tailings
‡
ANOVA performed on log10 transformed data. Replication (n) in parentheses. *, ** and *** denote significant difference from
reference site at P-0.05, -0.01 and -0.001, respectively. † , †† and ††† denote significant difference from Irish tailings at P-0.05,
-0.01 and -0.001, respectively.
East Halkyn where there were no differences (P)
0.05), and the same fraction at Cwmrheidol which
was lower than for the Irish material (P-0.05).
The extractable and total concentrations of lead
in soils from the reference site were within the
normal ranges for British soils of: 0.13–16 mg
kg –1 and 10.9–145 mg kgy1, respectively (Archer
and Hodgson, 1987), and the total value was very
similar to the baseline of 30 mg kgy1 calculated
for soils worldwide (Kabata-Pendias and Pendias,
1992). Combining the sequentially extracted fractions gave a total lead concentration in the Irish
tailings similar to the 2320 mg kgy1 reported
previously for the same site (Brady, 1993). Total
lead levels from the Welsh mine sites were also
similar to those previously reported (Table 1)
(WDA, 1978).
3.2. Vegetation and litter
Lead concentrations in vegetation and litter samples are summarised in Table 3. The concentration
in litter samples was always considerably higher
than any other vegetation sample type, and at all
sites, except the Irish tailings, composite seeds
showed the lowest levels of lead contamination.
The grasses F. rubra and Agrostis spp had significantly elevated levels of lead at all the Welsh
mine sites compared to both the reference and
Irish sites (P-0.05). Similarly, concentrations of
lead in composite ground cover vegetation were
significantly higher (P-0.001) at all Welsh mine
sites except Cwmrheidol, whilst there was no
difference between the Irish and reference sites
(P)0.05). Litter showed a similar pattern, with
all Welsh sites and the Irish tailings dam significantly elevated with respect to the reference site
(P-0.001) and all Welsh sites except Cwmrheidol
higher than the Irish tailings (P-0.05). The concentrations of lead in composite seeds showed less
differences between sites with only samples from
Frongoch, Rhosesmor, Trelogan and East Halkyn,
182
A. Milton et al. / The Science of the Total Environment 299 (2002) 177–190
Table 4
Concentration of lead in herbivorous and detritivorous invertebrates from the reference and mine sites (mg kgy1 dry wt."S.E.M.)
Site
A. aulica
C. brunneus
A. ater
C. nemoralis
Reference
Grogwnion
0.7"0.1†††
(6)
5.1"0.9***
(5)
NT
12"1.0
(3)
109"19**†††
(4)
NT
Cwmsymlog
NT
Frongoch
Ystum Tuen
222"56***†††
(5)
136"32***†††
(7)
NT
9.5"0.3††
(3)
108"19***††
(3)
63"12***
(3)
14"5.0††
(4)
126"21***†††
(6)
8.0"0.4††
(3)
NT
Rhosesmor
NT
Trelogan
41 133"8644***†††
(5)
3847"583***†††
(6)
8.6"2.9***
(8)
2.6"0.5††
(5)
33"4.7***
(8)
48"9.3***††
(4)
30"4.0***†
(11)
61"18***††
(7)
13"4.7**
(8)
34"14***
(4)
77"15***†††
(5)
63"15***††
(8)
72"23***††
(4)
12"2.0**
(10)
Cwmerfin
Cwmrheidol
East Halkyn
Irish Tailings
65"29†
(5)
109"23**†††
(5)
5.6"4.1*†††
(4)
22"1.4
(3)
38"5.3†
(3)
NT
79"10**††
(3)
9.2"0.7
(5)
NT
25"4.5*
(3)
44"3.1***
(4)
32"5.2**
(9)
NT, not trapped or biomass too low for comparison. Replication (n) in parentheses. *, ** and *** denote significant difference
from reference site at P-0.05, -0.01 and -0.001, respectively. † , †† and ††† denote significant difference from Irish tailings at P0.05, -0.01 and -0.001, respectively.
being significantly higher than the reference and
Irish sites (P-0.05).
The concentrations of lead in plant material
followed the pattern described previously for vegetation on a fluorspar tailings lagoon surface
(Andrews et al., 1989), with the highest concentrations being found in the plant litter. This is
likely to reflect both a genuine increase in internal
metal burden, and also some degree of surface
contamination. Levels of lead in reference site
vegetation were all similar to those reported previously for clean sites (Andrews et al., 1989).
Seasonal variation in vegetation concentrations of
lead have been described for contaminated sites,
and the summer is when lowest values occur, in
response to growth dilution (Hunter et al., 1987b).
The samples reported here were collected during
the summer and so may represent lower dietary
exposure conditions for consumers than those
experienced at other times of the year. Seasonal
fluctuations may have implications for the inter-
pretation of animal tissue concentrations of lead
since the composition of the diet itself, and also
the metal loadings, will cause changes in metal
intake, tissue- and total body burdens.
3.3. Invertebrates
Table 4 summarises the lead concentrations in
herbivorous invertebrates. Concentrations of lead
in the beetle Amara aulica and the grasshopper
Chorthippus brunneus, varied greatly between sites
with all mine waste animals showing significantly
higher values than the reference site (P-0.01).
Animals taken from all the historic Welsh mine
sites, except Cwmerfin, Cwmrheidol and Ystum
Tuen, were also significantly higher than those
from the Irish tailings (P-0.05).
Specimens of Arion ater from Cwmerfin, Frongoch, and East Halkyn had elevated levels of lead
compared to the reference site, whilst animals from
Cwmrheidol were significantly lower (P-0.05).
A. Milton et al. / The Science of the Total Environment 299 (2002) 177–190
183
Table 5
Concentration of lead in carnivorous invertebrates from the reference and mine sites (mg kgy1 dry wt."S.E.M.)
Site
F. nigrita
P. mirabilus
O. parietinus
Reference
0.6"0.1
(22)
14"0.9***
(6)
160"39***†††
(5)
NT
NT
17"1.7
(5)
1998"860***†††
(4)
NT
Cwmerfin
Grogwnion
Cwmsymlog
NT
61"25
(8)
2755"696†††
(7)
326"97
(11)
241"52†††
(5)
48"24
(5)
NT
1594"21†††
(3)
3253"1775†††
(6)
75"14
(6)
175"44***†††
(3)
60"18
(7)
159"36***†††
(4)
NT
NT
Frongoch
Cwmrheidol
Ystum Tuen
Rhosesmor
Trelogan
1532"65***†††
(4)
16"2.1***
(22)
East Halkyn
Irish Tailings
207"107*
(3)
755"216***††
(3)
NT
NT
NT
NT
NT
48"8.1*
(3)
NT, not trapped or biomass too low for analysis. Replication (n) in parentheses. * and *** denote significant difference from
reference site at P-0.05, and -0.001, respectively. †† and ††† denote significant difference from Irish tailings at P-0.01 and 0.001, respectively.
Animals from all the Welsh sites were higher than
the Irish tailings, except Ystum Tuen (P)0.05),
and Cwmrheidol where animals were significantly
lower (P-0.05). A similar pattern was found for
C. nemoralis with all sites except Cwmsymlog and
Cwmrheidol higher than the reference site (P-
0.05). Cwmerfin and Frongoch were the only sites
with a higher lead concentration than the Irish site
(P-0.01), while Cwmsymlog and Cwmrheidol
were significantly lower (P-0.01).
Lead concentrations in carnivorous invertebrates
are summarised in Table 5. The beetle, Ferona
Table 6
Lead concentrations in tissues of Apodemus sylvaticus from the reference and mine sites (mg kgy1 dry wt."S.E.M.)a
Site
n
Liver
Kidney
Bone
Muscle
Reference
Cwmerfin
Grogwnion
Cwmsymlog
Frongoch
Cwmrheidol
Ystum Tuen
Trelogan
East Halkyn
Irish tailings
8
6
5
4
4
5
5
8
8
11
0.4"0.1
4.4"2.0**
3.2"0.9**
3.4"0.7**
4.4"1.8**
4.3"0.9**†
2.7"0.7**
2.7"0.5**
4.7"0.6***†
1.3"0.5
1.2"0.2
15"1.4***†††
14"1.7***†††
43"7.5***†††
40"12***†††
9.6"3.6***†
10"2.5***†
14"3.8***††
6.8"0.6***†
3.2"0.7
2.2"0.3†††
531"162***†††
165"21***††
573"84***†††
603"305***†††
123"62***†
158"59***†
111"31***†
84"26***
27"8.0***
0.04"0.02†
1.5"0.3***
1.4"0.4***
0.6"0.4
0.8"0.1*
0.6"0.2
0.4"0.1
0.6"0.1
0.3"0.1
0.8"0.2*
a
Excluding Rhosesmor mine where no animals were caught.
*, ** and *** denote significant difference from reference site at P-0.05, -0.01 and -0.001, respectively. † , †† and
significant difference from Irish tailings at P-0.05, -0.01 and -0.001, respectively.
†††
denote
A. Milton et al. / The Science of the Total Environment 299 (2002) 177–190
184
Table 7
Total body concentration (TBC) and estimated daily lead
intake by A. sylvaticus (mean"S.E.M.)
Site
N
TBC
(mg kgy1)
Pb intake
(mg dayy1)
Reference
Cwmerfin
Grogwnion
Cwmsymlog
Frongoch
Cwmrheidol
Ystum Tuen
Trelogan
East Halkyn
Irish tailings
8
6
5
4
4
5
5
8
8
11
0.73"0.3
29.4"11.3
3.9"0.8
23.3"5.0
37.5"19.8
2.1"0.8
1.6"0.4
7.6"2.5
5.6"2.0
3.4"0.8
12.5"0.8
1380"167
1934"193
1491"66
774"39
117"17
351"25
16 571"1151
1179"104
178"23
nigrita, had significantly elevated lead levels at all
mine sites when compared with reference animals
(P-0.001). For the wolf spider, Pisaura mirabilus, only animals from Cwmsymlog, Trelogan and
East Halkyn had concentrations significantly higher than the Irish site (P-0.001). Similarly, for the
harvestman, Opilione parietinus, all Welsh mine
sites recorded concentrations higher than the reference site (P-0.05).
3.4. Apodemus sylvaticus
Mean body wet weights (g"S.E.M.) ranged
from 21"2.4 for the Irish tailings to 26.7"1.3 for
Frongoch, with a predominant weight range for
individuals of 20–24 g. There was no significant
difference between sites in terms of mean body
wet weights (P)0.05).
Concentrations of lead in the tissues of A.
sylvaticus are summarised in Table 6. Values for
the reference animals are similar to or lower than
those reported previously for clean sites (Talmage
and Walton, 1991; Purcell et al., 1992). The pattern
of tissue accumulation: bone)kidney)liver)
muscle, was the same at all sites and the same as
that reported previously (Roberts et al., 1978;
Andrews et al., 1989; Purcell et al., 1992). For the
whole data set, there was a significant relationship
between concentrations of lead in the total body
and in the kidney (r 2s0.71; ns64; P-0.001)
and bone (r 2s0.62; ns64; P-0.001) reflecting
the importance of these two tissues as the main
depositories for absorbed lead.
Liver, kidney, and bone (femur) lead concentrations in animals from all Welsh mine sites were
higher (P-0.01) than the reference site, although
only bone (P-0.001) and muscle (P-0.05) from
the Irish site were elevated above the levels at the
reference site. Kidney and bone concentrations
from all the Welsh sites were also higher than the
Irish tailings (P-0.05). Only animals trapped at
Cwmrheidol and East Halkyn had lead levels in
liver that were higher than for the Irish site (P0.05). Muscle showed a variable role in its contribution to body lead burden with Cwmerfin,
Grogwnion, Frongoch and the modern Irish tailings
significantly higher than the reference site (P0.05). However, all liver values were lower than
concentrations reported previously for small mammals from both contaminated and clean sites (Roberts et al., 1978; Storm et al., 1994). Muscle tissue
is not known to have a prominent role in the
storage of lead in the mammalian body other than
through its higher proportional contribution to total
body weight than that made by most other tissues.
3.5. Diet estimation
In view of the similarities in vegetation at the
Welsh mine sites and reference site, at least as
regards species present, a common dietary component model was adopted. However, the different
vegetation of the Irish tailings dam meant that a
separate model was needed to accommodate the
reduced floristic diversity. Hunter et al. (1987a)
described the seed part of the diet of A. sylvaticus
inhabiting a contaminated grassland as comprising
the seeds of R. fruticosus, rose bay willow herb
(Epilobium angustifolium) and U. europaeus, supplemented by wind-blown tree seeds (A. pseudoplatanus) from surrounding areas. This collection
of species is very similar to that encountered at
the Welsh mine sites, and the diet model advocated
by Hunter et al. (1987a) (60% seed, 30% grass
and 10% invertebrate material) was adopted
accordingly. The diet of A. sylvaticus has also been
described for animals living in ‘set-aside’ grassland
and on revegetated lignite mining waste where
grass and seeds account for over 95% of the food
consumed (Halle, 1993; Rogers and Gorman,
1995). A similar, predominantly grass-based, diet
A. Milton et al. / The Science of the Total Environment 299 (2002) 177–190
was assumed for animals on the revegetated Irish
tailings surface (45% seed, 50% grass and 5%
invertebrate material).
In addition to the three food components of the
diet (seeds, grass, and invertebrates) incidental
soil ingestion can comprise a significant proportion
(approx. 2%) of the diet (Beyer et al., 1994; Erry
et al., 2000). An estimate of Pb intake from soil
was also included into the dietary model. The
intake of all four dietary components was based
on body weight.
There were no significant correlations (P)0.05)
between estimated dietary intake of lead (Table 7)
and the calculated total body concentration or any
individual tissue value. However, high diet levels
were generally associated with increased tissue and
whole body burdens.
4. Discussion
There have been few investigations into the
impact and biological transfer of lead to wildlife
where the lead has been derived from soils and
vegetation associated with mine tailings (Roberts
and Johnson, 1978; Andrews et al., 1989; Purcell
et al., 1992). Most of the research into soil lead
pollution originating from the metals industry has
focused on smelter deposition from the atmosphere
to normal soils (e.g. Strojan, 1978; Ma et al.,
1991; Storm et al., 1994; Rabitsch, 1995a,b).
In this study, bioavailable lead, represented by
the water and ethanoic acid extractable metal,
showed a large variation in the percentage of the
total lead it represents; for example available lead
from Cwmrheidol was approximately 2% of the
total while for Grogwnion and the Irish tailings
85% of the total was available. The bioavailability
of heavy metals is influenced by many factors in
unpolluted soils including substrate surface area as
a function of particle size, pH, clay content,
calcium carbonate, organic matter and levels of
iron and manganese (Hughes et al., 1980; Sillan¨ ¨ and Jansson, 1992; McLaughlin et al., 2000).
paa
In polluted soils the chemical and physical nature
of the soil lead itself will also be important
(Rieuwerts et al., 2000) and in mine tailings this
will also included the physical and chemical nature
of discrete high lead particles (Davis et al., 1992).
185
Relationships between substrate levels of metals
and those in vegetation growing on mine waste
are also strongly influenced by plant factors such
as transpiration rates, root growth rates and architecture, root exudate production and root surface
lead precipitation, and the presence of associated
mycorrhizal fungi. Thus, the complex interactions
between the geochemistry of the wastes and plant
species or ecotype influence plant uptake of lead.
In this study, there was no correlation between
either the total or available lead in mine waste and
the concentrations in vegetation growing on them
(P)0.05). The concentration in vegetation itself
appears to be the only reliable indicator of the
potential for lead transfer from the substrate
through the trophic levels of the ecosystem developed on the mining wastes.
Animals play an important role in ecosystem
functioning. In the context of ecosystem development on mine wastes these roles, inter alia, include
(Majer, 1989): decomposition and nutrient cycling;
improvement in soil structure; and plant pollination
and seed dispersal. The detrimental effects of
heavy metals on the invertebrate communities
responsible for decomposition have been well documented and the lead levels present in soils and
litter from all the mine sites in this study generally
exceed those reported to impact upon decomposition processes (see below and Babich and Stotzky,
1985; Bengtsson et al., 1988). Inhibited incorporation of organic matter into the substrate can also
have long term implications since the adhesion of
mineral particles to organic matter is essential for
the development of the soil crumb structure that
determines aeration and drainage.
Conservative threshold concentrations of soil
lead above which adverse effects may be expected
in soil-dwelling invertebrates are surprisingly low,
and range from 25 to 300 mg kgy1 (De Vries and
Bakker, 1996; McLaughlin et al., 2000). Using the
lowest observed effect concentration (LOEC) from
field and laboratory studies 150 mg kgy1 is a
consensus critical threshold value for leaf litter
and soil humus horizons (Bengtsson and Tranvik,
1989; De Vries and Bakker, 1996). Given the
higher values of substrate and litter lead values in
relation to these critical values, in this study, this
may partly explain why pitfall catches of inverte-
186
A. Milton et al. / The Science of the Total Environment 299 (2002) 177–190
brates were so reduced in diversity and abundance
at all mine sites. Heavy metals have also been
associated with the total absence of earthworms
(Genus-Scherotheca) from metal contaminated
soils (Rida and Bouche, 1995) and such was the
case for the majority of pitfall traps located at
mine sites in the present study, though earthworms
were encountered frequently in pitfalls at the reference site. However, pitfall traps are not a particularly efficient way of assessing earthworm
abundance.
Terrestrial gastropods, represented by Arion ater
and Cepaea nemoralis, are responsible for the
initial breakdown of a significant proportion of the
litter pool and so form an important part of the
detritivorous as well as herbivorous fauna (Russell-Hunter, 1983; Laskowski and Hopkin, 1996).
However, though slugs and snails have been
reported as being relatively insensitive to elevated
levels of metals in laboratory exposures (Laskowski and Hopkin, 1996), they are often absent
from highly polluted sites. This absence has been
attributed to delayed reproduction causing eventual
extinction, although starvation-induced mortalities
due to the avoidance of contaminated foods is also
an important factor (Laskowski and Hopkin,
1996). Laskowski and Hopkin (1996) also highlight the importance of synergistic impacts from
interactions between metals in polymetallic waste
when assessing the ecotoxicological risk to terrestrial molluscs. All mine tailings in this study
display this polymetallic character with relatively
high levels of cadmium and zinc.
Given the importance of ground and soil-living
invertebrates to decomposition and nutrient
cycling, the high levels of metals in the substrata
could, through their impact upon invertebrate
diversity and abundance, emphasise deficiencies in
the labile pool of nutrients in the grassland ecosystems. This would add to the inherited problems
caused by the initially low concentrations of plant
nutrients in the mine wastes at the time of deposition, especially in respect of nitrogen and phosphorus. The net effect of retarded decomposition
rates will be to further reduce the prospect of
diversification through plant recruitment into the
system at unmanaged sites. Along with the requirement for metal tolerance, this must partly explain
the largely barren nature of historic abandoned
mine sites, even 100 years after the termination of
waste disposal.
Members of the Carabidae family are thought
not to accumulate metals, and they have even been
termed ‘deconcentrators’ (Dallinger, 1993). Levels
of lead greater than 20 mg kgy1 have been
recorded only rarely in field-caught animals (Hopkin, 1989; Rabitsch, 1995a), although high concentrations ()600 mg kgy1) have been reported
in adult Carabidae, and also in Staphylinidae ()
1000 mg kgy1) (Rabitsch, 1995a). The potential
to bioaccumulate lead to very high concentrations
is evident in the present study. Specimens of A.
aulica with the highest concentrations of lead
originated from the most contaminated sites but
there was no overall significant relationship wr 2s
0.45 (soil), r 2s0.15 (litter), ns7; P)0.05x
between the lead burden of the beetles and soily
litter concentrations. The metal detoxification
mechanism that may enable beetles to survive in
such extreme habitats is the local accumulation of
lead in cells or vesicles which are then excreted
via the faeces (Dallinger, 1993). In addition, Carabids have also been found with 63–82% of total
body lead deposited in the mainly chitinous exoskeleton (Roberts and Johnson, 1978). Immobilisation of lead in this manner may constitute a
secondary detoxification system that is brought
into being when the primary mechanisms are
overloaded by extreme dietary lead intake.
It is difficult to estimate the natural diet components and therefore the intake–excretion dynamics of invertebrates. One exception is the
grasshopper Chorithippus brunneus because of its
simple, grass dominated diet ()90%). Moreover,
C. brunneus forms an important part of the diets
of birds, small mammals and some larger invertebrates so the metal accumulation dynamics of this
animal are significant to food chain transfer at
contaminated sites (Hunter et al., 1987c). A significant relationship (r 2s0.60; ns11; P-0.05)
was found between the concentration of lead in
the combined and averaged (1:1 wyw; Festuca:
Agrostis; Table 3) grass diet and the whole bodies
of C. brunneus (Table 4).
Carnivorous invertebrates are exposed to levels
of lead that reflect environmental levels as miti-
A. Milton et al. / The Science of the Total Environment 299 (2002) 177–190
gated by the ability of their prey to physiologically
regulate absorption and retention of the metal.
Spiders assimilate only the soft tissues of their
prey, where most metals are stored preferentially,
and are thus exposed to higher dietary concentrations of lead than other species (Hopkin, 1989).
This is reflected in P. mirabilus which showed
higher lead concentrations than the herbivorous
invertebrates at all but one site. P. mirabilus feeds
on small, soft-chitinised insects including the Collembola and Isopoda which, although not analysed
in this study, accumulate metals in line with
environmental exposure (Hopkin, 1989). There is
evidence that P mirabilus and other lycosid spiders
are sensitive to high soil metal contamination
(Read et al., 1998). The harvestman O. parietinus
also feeds on similar prey (Adams, 1984) and this
was reflected in the body lead burdens being higher
than for any of the herbivorous invertebrates
(Tables 4 and 5). This trophic pattern is similar to
that described elsewhere in respect of cadmium
for which carnivorous taxa generally had higher
concentrations than species from lower trophic
levels (Hunter et al., 1987d).
Tissue concentrations of lead in wild small
mammals are very much dietary dependant which
in turn means that exposure is via the transfer of
lead in food chains (Ma, 1996). Generally, as in
Apodemus sylvaticus in this study, animals from
the more contaminated sites showed the highest
levels of lead in soft tissues and, in some cases,
concentrations were sufficient to induce clinical
symptoms of lead toxicity based on precedent.
Lead administered in drinking water has been
shown to induce the formation of renal nuclear
inclusion bodies, increased d-aminolevulinic acid
excretion, reticolocytosis, renal oedema, reduced
sperm counts, irregular dioestrous, growth retardation and impaired learning in laboratory mice
and rats, and similar observations have been reported for small mammals taken from contaminated
areas (Shore and Douben, 1994).
Clinical signs of lead toxicosis appear to be
associated with relatively low concentrations of 5
and 15 mg kgy1 dry wt. in liver and kidneys,
respectively (Ma, 1996). Histological changes in
kidney include altered proximal kidney tubular
cells, oedema and nuclear inclusion bodies. In a
187
study by Ma (1989) the critical value of lead in
the kidneys of small mammals as regards nuclear
inclusion bodies was 25 mg kgy1 dry wt. This
figure was only exceeded at two of the historic
Welsh mine sites suggesting a low probable incidence of lead poisoning on this criterion alone.
However, another sensitive indicator of toxicological risk is the organ-to-body weight ratio (Ma,
1989) and a comparison of the kidney weight as
a proportion of total body weight (renal oedema)
indicated that animals from Cwmsymlog
(1.8%"0.1) and Frongoch (1.9%"0.1) were significantly different from the reference site
(1.3%"0.0) (P-0.01). Cwmsymlog and Frongoch were the sites with the highest kidney lead
concentrations both in excess of the 25-mg kgy1
‘threshold’ level. Roberts et al. (1978) also found
evidence of renal oedema in animals with a kidney
lead concentration of 35–55 mg kgy1. Neither
Cwmsymlog nor Frongoch animals had elevated
kidney cadmium or zinc concentrations compared
to the reference site.
In laboratory feeding trials, rats fed 200 mg
kgy1 of lead in their diets exhibited increased
relative kidney weight, along with the development
of intranuclear inclusion bodies, although the kidney lead concentration was only 10.8 mg kgy1
(Mahaffey et al., 1981). These apparent conflicts
in critical values for renal inclusion bodies and
oedema highlight the large number of factors that
influence trace element metabolism including the
dietary concentrations of other metals, total metal
burden, age, period of exposure, and metabolic
condition.
This study also confirms the importance of bone
and kidney as the major repositories for absorbed
lead. For the reference site and Irish tailings, only
4 and 6%, respectively, of the total body lead
burden was located in these two tissues combined.
However, at the most contaminated Welsh sites
the bone and kidney account for between 16 and
52%, respectively, of the body burden. Lead deposited in the skeleton of mammals poses a relatively
low toxicological risk as it is bound within the
hydroxyapatiteycollagen bone matrix and is physiologically inert. Immobilisation of lead inhibits
translocation within the body to metabolically
188
A. Milton et al. / The Science of the Total Environment 299 (2002) 177–190
active centres where a toxicological response could
result.
Liver is the ‘first’ organ to encounter absorbed
dietary lead distributed via the bloodstream. In
contrast to kidney there is greater regulation of
hepatic lead in small mammals and, in a review
of a number of studies, the lead concentrations in
liver did not exceed 15 mg kgy1 even when kidney
values in the same individual animal were as high
as 60 mg kgy1 (Shore and Douben, 1994). However, concentrations of lead in liver found in this
study did not exceed 5 mg kgy1 even when kidney
values were of the order of 40 mg kgy1 and thus
liver values here were lower than in most previous
investigations at contaminated sites, and indeed
similar to some reported reference site concentrations (Talmage and Walton, 1991). The reason for
these low liver values is not known and it is
improbable that the lead concentrations reported
here would cause any disruption to normal hepatic
functioning.
Animals trapped, even at the centre of a large
contaminated area, may only be taking part of
their food from highly polluted terrain. Attuquayefio et al. (1986) found that small mammals
inhabiting unproductive but unpolluted ecosystems
had significantly larger (P-0.001) home ranges
(12 290–36 499 m2) than those inhabiting productive and species-rich woodland (1719–6276 m2).
The low primary productivity of mine sites probably means that animals may be forced to forage
in adjacent areas of lower contamination, where
the abundance of food is greater. The consequential
dilution of lead intake, as compared with exposure
estimated by the analysis of dietary items taken
only from the contaminated site itself, may explain
the lack of correlation between dietary lead intake
and bodyytissue concentrations. A similar reduction in metal loading when home ranges are
increased has been conceptualised for cadmium
(Marinussen and Vanderzee, 1996).
The data presented here indicate that lead is an
influential component of the mine tailings ecosystem. The interaction between lead and nutrient
cycling is likely to be a major obstacle to what
natural colonisation can be achieved on historic
abandoned sites and even to the long-term sustainability of any reclamation scheme on modern sites.
For the modern Irish tailings, where an agricultural
end use for the reclaimed land has been tested and
verified as an option (Brady, 1993), fertiliser input
would be a normal management practise and would
circumvent problems of inhibited decomposition
processes. For lower maintenance end uses including wildlife habitat, it may be that the regular
input of fertilisers to build-up a nutrient capital is
all that is needed to achieve a sustainable ecosystem, albeit one with suppressed nutrient dynamics
and litter decomposition. Many species of invertebrate, fungi and bacteria have been reported as
tolerant to elevated levels of heavy metals (Tyler
et al., 1989), and it might be possible to introduce
selected species and populations from elsewhere
to promote decomposition. This approach would
parallel the use of tolerant grasses for providing
an initial vegetation cover on metal-contaminated
mine wastes (Johnson et al., 1994).
The concentrations of lead in tissues of A.
sylvaticus provide evidence that there may be a
toxicological risk associated with inhabiting some
of the historic mine sites. Tissue concentrations
are certainly consistent with conditions that induce
the clinical symptoms of plumbism in small mammals. However, given the uncertainties in quantifying the many factors that determine whether the
potential for lead poisoning is realised in practice,
and the fact that animals can emigrate from contaminated areas to feed, the overall hazard to the
viability of indigenous populations across the
many hundreds of disused and abandoned sites is
considered to be low. Moreover, the improved
mineral processing technology which is expressed
in the lower levels of lead in tailings and mammals
from the modern Irish mine, suggests that the
benefits of technological change may eventually
be reflected in the diversity, structure and functioning of ecosystems established on mine wastes in
the future.
References
Adams J. The habitat and feeding ecology of woodland
harvestmen (Opiliones) in England. Oikos 1984;42:361 –
370.
Andrews SM, Johnson MS, Cooke JA. Distribution of trace
element pollutants in a contaminated grassland ecosystem
A. Milton et al. / The Science of the Total Environment 299 (2002) 177–190
established on metalliferous fluorspar tailings. Lead Environ
Pollut 1989;58:73 –85.
Archer FC, Hodgson IH. Total and extractable trace element
contents of soils in England and Wales. J Soil Sci
1987;38:421 –431.
Attuquayefio DK, Gorman ML, Wolton RJ. Home range sizes
in the wood mouse Apodemus sylvaticus: habitat, sex and
seasonal differences. J Zool Lond 1986;210:45 –53.
Babich H, Stotzky G. Heavy metal toxicity to microbemediated ecologic processes: a review and potential application to regulatory policies. Environ Res 1985;36:111 –
137.
Bengtsson G, Tranvik L. Critical metal concentrations for
forest soil invertebrates; a review of limitations. Water Air
Soil Pollut 1989;47:381 –417.
Bengtsson G, Berden M, Rundgren S. Influence of soil animals
and metals on decomposition processes: a microcosm experiment. J Environ Qual 1988;17:113 –119.
Beyer WN, Connor EE, Gerould S. Estimates of soil ingestion
by wildlife. J Wildlife Mange 1994;58:375 –382.
Brady E. Revegetation and environmental management of
lead–zinc mine tailings. M.Phil. thesis. England: University
of Liverpool, 1993:201 pp.
Dallinger R. Strategies of metal detoxification in terrestrial
invertebrates. In: Dallinger R, Rainbow SP, editors. Ecotoxicology of metals in invertebrates. Boca Raton, USA: Lewis,
1993. p. 245 –290.
Davies BE. Trace metal content of soils affected by base metal
mining in the west of England. Oikos 1971;22:366 –372.
Davis A, Ruby MV, Bergstrom PD. Bioavailability of arsenic
and lead in soils from the Butte, Montana, mining district.
Environ Sci Technol 1992;26:461 –468.
Day RW, Quinn GP. Comparisons of treatments after analysis
of variance in ecology. Ecol Monogr 1989;59:433 –463.
De Vries W, Bakker DJ. Manual for calculating critical loads
of heavy metals for soils and surface waters. Preliminary
guidelines for environmental quality criteria, calculation
methods and input data. Wageningen. DLO Winand Staring
Centre Rep 1996;114:173.
Erry BV, Macnair MR, Meharg AA, Shore RF. Arsenic
contamination in wood mice (Apodemus sylvaticus) and
bank voles (Clethrionomys glareous) on abandoned mine
sites in southwest Britain. Environ Pollut 2000;110:179 –
187.
Gerhardt A. Review of impact of heavy metals on stream
invertebrates with special emphasis on acid conditions.
Water Air Soil Pollut 1993;66:289 –314.
Halle S. Wood mice (Apodemus sylvaticus) as pioneers of
recolonization in a reclaimed area. Oecologia 1993;94:120 –
127.
Hopkin SP. Ecophysiology of metals in terrestrial invertebrates.
London: Elsevier, 1989. (366 pp).
Hughes MK, Lepp NW, Phillips DA. Aerial heavy metal
pollution in terrestrial ecosystems. Adv Ecol Res
1980;11:218 –327.
189
Hunter BA, Johnson MS, Thompson DJ. Ecotoxicology of
copper and cadmium in a contaminated grassland ecosystem.
3. Small mammals. J Appl Ecol 1987;24:601 –614.
Hunter BA, Johnson MS, Thompson DJ. Ecotoxicology of
copper and cadmium in a contaminated grassland ecosystem.
1. Soil and vegetation contamination. J Appl Ecol
1987;24:573 –586.
Hunter BA, Hunter LM, Johnson MS, Thompson DJ. Dynamics of metal accumulation in the grasshopper Chorthippus
brunneus in contaminated grasslands. Arch Environ Contam
Toxicol 1987;16:711 –716.
Hunter BA, Johnson MS, Thompson DJ. Ecotoxicology of
copper and cadmium in a contaminated grassland ecosystem.
2. Invertebrates. J Appl Ecol 1987;24:587 –599.
Johnson MS, Cooke JA, Stevenson J. Revegetation of metalliferous wastes and land associated with metal mining. In:
Hester RE, Harrison RM, editors. Issues in environmental
science and technology, vol. 1. Mining and its environmental
impact. Cambridge, UK: Royal Society of Chemistry, 1994.
p. 31 –48.
Kabata-Pendias A, Pendias H. Trace elements in soils and
plants. Boca Raton: Lewis, 1992. (365 pp).
Lansdown R. Lead, intelligence, attainment and behaviour. In:
Lansdown R, Yule W, editors. The lead debate: the environment, toxicology and child health. London: Croom Helm,
1986. p. 235 –271.
Laskowski R, Hopkin SP. Effect of Zn, Cu, Pb, and Cd on
fitness in snails (Helix aspersa). Ecotox Environ Saf
1996;34:59 –69.
Ma WC. Effect of soil pollution with metallic lead pellets on
lead bioaccumulation and organybody weight alterations in
small mammals. Arch Environ Contam Toxicol
1989;18:617 –622.
Ma WC. Lead in mammals. In: Beyer N, Heinz G, RedmondNorwood A, editors. Environmental contaminants in wildlife: interpreting tissue concentrations. Boca Raton: Society
of Environmental Toxicology and Chemistry Special Publication. Lewis, 1996. p. 281 –296.
Ma WC, Denneman W, Faber J. Hazardous exposure of
ground-living small mammals to cadmium and lead in
contaminated terrestrial ecosystems. Arch Environ Contam
Toxicol 1991;20:266 –270.
Mahaffey KR, Caper SG, Gladen BC, Fowler BA. Concurrent
exposure to lead, cadmium, and arsenic: effects on toxicity
and tissue metal concentrations in the rat. J Lab Clin Med
1981;98:463 –481.
Majer JD. Animals in primary succession: the role of fauna in
reclaimed lands. Cambridge: Cambridge University Press,
1989. (450 pp).
Marinussen MPJC, Vanderzee SEATM. Conceptual approach
to estimating the effect of home-range on the exposure of
organism to spatially-variable soil contamination. Ecol
Modell 1996;87:83 –89.
McLaughlin MJ, Hamon RE, McLaren RG, Speir TW, Rogers
SL. Review: A bioavailability-based rationale for controlling
metal and metalloid contamination of agricultural land in
190
A. Milton et al. / The Science of the Total Environment 299 (2002) 177–190
Australia and New Zealand. Aust J Soil Res 2000;38:1037 –
1086.
Milton AM, Collings SE, Johnson MS. The determination of
lead and cadmium in biota by electrothermal atomic absorption spectrometry. Anales Quim 1998;94:349 –353.
Pain DJ. Lead in waterfowl. In: Beyer N, Heinz G, RedmondNorwood A, editors. Environmental contaminants in wildlife: interpreting tissue concentrations. Boca Raton: Society
of Environmental Toxicology and Chemistry Special Publication. Lewis, 1996. p. 251 –264.
Pankakoski E, Hanski I. Metrical and non-metrical skull traits
of the common shrew Sorex araneus and their use in
population studies. Ann Zool Fenn 1989;26:433 –444.
Purcell PW, Hynes MJ, Fairley JS. Lead levels in Irish small
rodents (Apodemus sylvaticus and Clethrionomys glareolus)
from around a tailings pond and along motorway verges.
Proc R Ir Acad 1992;92B:79 –90.
Rabitsch WB. Metal accumulation in arthropods near a leady
zinc smelter in Arnoldstein, Austria. 1. Environ Pollut
1995;90:221 –237.
Rabitsch WB. Metal accumulation in arthropods near a leady
zinc smelter in Arnoldstein, Austria. 3. Arachnida. Environ
Pollut 1995;90:249 –257.
Read HJ, Martin MH, Rayner MV. Invertebrates in woodlands
polluted by heavy metals—an evaluation using canonical
correspondabnce analysis. Water Air Soil Pollut
1998;106:17 –42.
Rida AMMA, Bouche MB. The eradication of an earthworm
genus by heavy-metals in southern France. Appl Soil Ecol
1995;2:45 –52.
Rieuwerts JS, Farago ME, Cikrt M, Benko V. Differences in
lead bioavailability between a smelting and a mining area.
Water Air Soil Pollut 2000;122:203 –229.
Roberts RD, Johnson MS. Dispersal of heavy metals from
abandoned mine workings and their transference through
terrestrial food chains. Environ Pollut 1978;16:293 –309.
Roberts RD, Johnson MS, Hutton M. Lead contamination of
small mammals from abandoned metalliferous mines. Environ Pollut 1978;15:61 –69.
Rogers LM, Gorman ML. The diet of the wood mouse
Apodemus sylvaticus on set-aside land. J Zool Lond
1995;235:77 –83.
Russell-Hunter WD. Overview: planetary distribution of and
ecological constraints upon the Mollusca. In: Russell-Hunter
WD, editor. The Mollusca, vol. 6: Ecology. London: Academic Press, 1983. p. 1 –27.
Scheuhammer AM, Norris SL. The ecotoxicology of lead shot
and lead fishing weights. Ecotoxicology 1996;5:279 –295.
Shore RF, Douben PET. Predicting ecotoxicological impacts
of environmental contaminants on terrestrial small mammals. Rev Environ Contam Toxicol 1994;134:49 –89.
¨ ¨ M, Jansson H. Status of cadmium, lead, cobalt and
Sillanpaa
selenium in soils and plants of thirty countries. FAO Soils
Bulletin 65. Rome: Food and Agriculture Organization,
1992. (195 pp).
Smal CM, Fairley JS. Food of wood mice (Apodemus sylvaticus) and bank voles (Clethrionomys glareolus) in Oak and
Yew woods at Killarney, Ireland. J Zool Lond
1980;191:413 –418.
Stewart DM. Does lead have a future—A 20-year vision. J
Power Sources 1994;48:17 –27.
Storm GL, Fosmire GJ, Bellis ED. Persistence of metals in
soils and selected vertebrates in the vicinity of the Palmerton
zinc smelters. J Environ Qual 1994;3:508 –514.
Strojan CL. The impact of zinc smelter emissions on forest
litter arthropods. Oikos 1978;31:41 –46.
Talmage SS, Walton BT. Small mammals as monitors of
environmental contaminants. Rev Environ Contam Toxicol
1991;119:47 –145.
Thornton I. Geochemical aspects of heavy metal pollution and
agriculture in England and Wales. In: inorganic pollution
and agriculture, MAFF Reference Book No. 326. London:
HMSO, 1980:105–194.
¨
¨ ¨ E,
Tyler G, Balsberg Pohlsson
AM, Bengtsson G, Baath
Tranvik L. Heavy-metal ecology of terrestrial plants, microorganisms and invertebrates. Water Air Soil Pollut
1989;47:189 –215.
Voth-Beach LM. The reduction of matrix interference utilising
the platform in graphite furnace AAS. Varian Instruments
at Work, No. AA-54. London: Varian-Techtron Pty Ltd,
1985. (15 pp).
WDA. Survey of abandoned metalliferous mines in Wales.
Cardiff, UK: Welsh Development Agency, 1978. (90 pp).