Soil and Biota of Serpentine: A World View
2009 Northeastern Naturalist 16(Special Issue 5):93–110
Elemental Concentrations of Eleven New Caledonian
Plant Species from Serpentine Soils:
Elemental Correlations and Leaf-age Effects
Robert S. Boyd1,* and Tanguy Jaffré2
Abstract - We investigated accumulation of elements (Ca, Co, Cr, Cu, Fe, K, Mg,
Mn, P, Pb, and Zn) in leaves of different ages for 11 evergreen woody plant species
from serpentine soils of New Caledonia. Species were classified into four categories
of Ni accumulation ability: one species was a non-accumulator (<100 mg Ni/
kg), three were accumulators (100–1000 mg Ni/kg), two were hyperaccumulators
(1000–10,000 mg Ni/kg), and five were hypernickelophores (>10,000 mg Ni/kg).
We harvested leaves from each species, separating them into three (four in one case)
relative age categories based upon their position along branches (younger toward
the apex, older far from it). Leaf samples were dried, ground, and dry-ashed, and
their elemental concentrations were determined by inductively coupled plasma
spectrometry (all elements except Ni) or atomic absorption spectrophotometry (Ni).
Great variation was found for most elements both within and among species, but
Ni varied most (1050-fold between species for oldest leaves). Correlations between
Ni and other transition metals showed no significant relationships within samples
of any species, but, we found significant positive correlations between Ni and Pb
(correlation coefficient = 0.97) and Ni and Fe (correlation coefficient = 0.87) among
species. Leaf Ni concentrations varied significantly with leaf age for two species, the
hypernickelophores Geissois pruinosa and Homalium kanaliense. We conclude that
Ni concentration varies markedly between species, but generally does not vary with
leaf age within species. We also suggest that four Ni accumulation category terms—
non-accumulator, hemi-accumulator, hyperaccumulator, and hypernickelophore—be
used to subdivide the wide variation found in Ni concentrations in plant leaves.
Introduction
Plants vary greatly in elemental makeup. Factors influencing this variation
are many, including species differences, environmental conditions, plant
physiological state, variation among plant organs, and others. The extremely
large variation in the concentrations of metals in plant tissues has stimulated
considerable scientific interest (Brooks 1987). Many high-metal plants grow
on serpentine soils derived from ultramafic rocks (high in Mg and Fe, but
low in Si and Ca), but other plants with much lower metal concentrations
can be found growing side-by-side with high-metal plants (Alexander et al.
2007, Brooks 1987, Kruckeberg 2002).
1Department of Biological Sciences, 101 Rouse Life Sciences Building, Auburn
University, AL 36849-5407 USA. 2Laboratoire de Botanique et d’ Ecologie Végétale
Appliquées, Institut de Recherche pour le Développement (IRD), Centre de Nouméa,
BP A5, Nouméa 98848, New Caledonia. *Author for correspondence - boydrob@
auburn.edu.
94 Northeastern Naturalist Vol. 16, Special Issue 5
The wide variation in metal concentration of plant species has stimulated
attempts to divide that variation into categories. For Ni, the term “hyperaccumulator”
was coined by Brooks et al. (1977) to describe plants containing
>1000 mg Ni/kg dry mass. The concept has since been extended to define
hyperaccumulation for a number of metals and other elements (Reeves and
Baker 2000), but Ni is the most commonly hyperaccumulated element,
comprising about 75% of the >400 taxa of hyperaccumulator plants listed
by Reeves and Baker (2000). Jaffré and Schmid (1974) used the term “hypernickelophore”
for plants containing extremely large (>10,000 mg/kg) Ni
concentrations. The term “accumulator” has been applied to plants that take up
100–1000 mg/kg (Berazaín 2004), whereas Reeves and Baker (2000) defined
“normal” Ni concentrations as 1–10 mg/kg. Thus, the concentration of Ni in
plant tissues shows extraordinary variation (>4 orders of magnitude), from as
little as 1 mg Ni/kg in normal plants to as much as 60,000 mg Ni/kg in leaves
of some hypernickelophore Phyllanthus species (Reeves and Baker 2000).
The enormous variation in plant Ni concentrations undoubtedly reflects
different physiological uptake and sequestration mechanisms among species.
Considerable effort is being made to understand these mechanisms
(Salt 2004), including the relatively new effort termed “metallomics”
(Szpunar 2004): the study of the location, identity, quantity, and complexation
of metals (including Ni) in cells. Variation in plant Ni concentrations
also must have ecological consequences (Boyd and Martens 1998), some of
which stem from the toxicity of Ni to organisms that consume living high-
Ni plant tissues (plant natural enemies). One consequence may be plant
defense; Martens and Boyd (1994) suggested that high levels of metal in
plant tissue provide an “elemental defense” against plant natural enemies,
and research to date has shown defensive effects for several hyperaccumulated
metals (Boyd 2007). Documenting how plant defenses vary in time and
space is vital to our understanding of plant defense strategies (e.g., Brenes-
Arguedas and Coley 2005), yet few studies have been conducted regarding
such variation in elemental defenses (for exceptions, see Boyd et al. 1999,
2004; Galeas et al. 2007).
The effectiveness of elemental defenses may be influenced by levels of
organic defenses or by high concentrations of other elements in plant tissues.
Boyd (2007) suggested that these “joint effects” may be ecologically important
and may have played a role in the evolution of hyperaccumulation by
plants. Simultaneously elevated levels of more than one metal may enhance
their protective effects against a single natural enemy or may protect against
multiple natural enemies (Boyd 2007). Examining hyperaccumulators for
concentrations of multiple metals provides information relevant to this relatively
new concept.
Finally, plants with enhanced Ni uptake abilities have applied uses
in phytoremediation (Pilon-Smits 2004, Raskin and Ensley 2000) or
phytomining (Nicks and Chambers 1998). Typically, these technologies
use hyperaccumulator plants to remove metal (such as Ni) from soils.
2009 R.S. Boyd and T. Jaffré 95
Knowledge of how metal accumulation varies within and among metal accumulating
species is important basic information for the applied uses of
these species (Whiting et al. 2004).
The archipelago of New Caledonia contains remarkably high levels of
biological endemism in many groups of organisms (Mittermeier et al. 2005),
including plants (Jaffré 2005). Nickel hyperaccumulators are particularly
numerous in New Caledonia, which ranks second (behind Cuba) in hosting
the greatest number of Ni hyperaccumulator taxa (Reeves 2003). New
Caledonian species also are important to the historical development of the
terms used in reference to hyperaccumulators, as both “hyperaccumulator”
and “hypernickelophore” were developed from investigations that focused
primarily on New Caledonian high-Ni plants. We selected 11 New Caledonian
plant species varying greatly in leaf Ni concentration and asked the
following sets of questions:
(1) What are the element concentrations in leaves of these 11 species, and do
they vary significantly between species? We were particularly interested
in documenting differences in Ni levels between species to subdivide
them into categories of Ni accumulation (normal or non-accumulator,
accumulator, hyperaccumulator, hypernickelophore).
(2) Given the wide range in mature leaf Ni concentrations documented in
answer to question 1, are there significant relationships between the concentrations
of Ni and other metals in mature plant leaves? Correlations
between metals may reflect the specificity of underlying physiological
mechanisms of metal uptake, transport, and sequestration. They also may
have ecological ramifications: positive correlations of concentrations of
two metals may lead to enhanced plant defense due to joint effects (Boyd
2004, 2007). For example, Jhee et al. (2006) demonstrated additive toxicity
of Ni and Zn to caterpillars of a leaf-chewing insect.
(3) Do leaf metal levels vary with leaf age within species, and are there
consistent patterns among plant species from different categories of Ni
accumulation? This question is relevant to determine if metal-based defenses
vary during leaf lifespans (Boyd 1998), as illustrated by research
into similar variation in the organic defense compounds of some tropical
forest species (e.g., Brenes-Arguedas et al. 2006).
Methods
We selected 11 species of plants (Table 1) growing on serpentine soils
at the southern end of Grand Terre, the main island of the New Caledonian
archipelago. Species were selected to include a diversity of Ni accumulation
categories. Samples of five species: Psychotria baillonii, Casearia silvana,
Homalium guillainii, Hybanthus austrocaledonicus, and Sebertia acuminata
(Sève Bleue) (see Table 1 for nomenclatural authorities) were collected
from a site within the Parc Territorial de la Rivière Bleue, which contains
areas of humid tropical forest. Jaffré and Veillon (1991) provide a description
of the vegetation of this forest type. These samples were collected
from a site located at Kaori Géant (a very large Agathis lanceolata Linley
ex Warb. tree). This site includes one of the study plots described by Jaffré
96 Northeastern Naturalist Vol. 16, Special Issue 5
and Veillon (1991) and is remarkable because six high-Ni species co-occur
in two layers of the vegetation (Boyd et al. 1999). Several other studies of
hyperaccumulators and their ecological relationships (Boyd and Jaffré 2001;
Boyd et al. 1999, 2006; Davis et al. 2001) have taken advantage of the large
number of co-occurring Ni hyperaccumulators found at this site. The shrub
layer contains the hypernickelophores Psychotria douarrei (Beauvis.) Däniker
(Rubiaceae), Hybanthus austrocaledonicus, and Casearia silvana. The
overstory contains Sebertia acuminata, as well as Homalium guillainii and
Geissois hirsuta Brongn. and Gris.
The remaining six species (Table 1) were sampled from serpentine sites
outside of the park. Leaves of two species, Grevillea gillivrayi and Geissois
pruinosa, were collected from plants growing at several serpentine sites
along the main road from Noumea to Yaté, before reaching the junction of
that road with the road leading northwest to the park entrance. Two other
species, Agatea longipedicillata and Garcinia amplexicaulis, were collected
from a site at the junction of the road from Noumea to Yaté and the road
leading to the park entrance. Samples of the remaining two species, Xanthostemon
aurantiacus and Homalium kanaliense, were collected from sites
along the road leading to the park entrance.
Species were classified into the following Ni accumulation categories:
normal or non-accumulator (<100 mg Ni/kg), accumulator (100–999 mg
Ni/kg), hyperaccumulator (1000–9999 mg Ni/kg), and hypernickelophore
(≥10,000 mg Ni/kg). Following the criteria of Reeves (1992) for defining
Table 1. Species selected for sampling in this study. The “Reference” column provides the citation
used to classify each species as either an accumulator of Ni (at least one sample contained
100–999 mg Ni/kg), hyperaccumulator of Ni (at least one sample contained 1000–9999 mg Ni/
kg), or hypernickelophore (at least one sample contained >10,000 mg Ni/kg).
Species Family Reference
Non-accumulator (< 100 mg Ni/kg)
Grevillea gillivrayi Hook. and Arn. Proteaceae Jaffré 1980
Ni accumulators (100–999 mg Ni/kg)
Garcinia amplexicaulis Vieill. Clusiaceae This study
Pyschotria baillonii Schltr. Rubiaceae This study
Xanthostemon aurantiacus (Brongn. and Myrtaceae This study
Gris) Schltr.
Ni hyperaccumulators (1000–9999 mg Ni/kg)
Agatea longipedicillata Baker f. (referred Violaceae Jaffré 1980
to as A. deplanchei by Jaffré)
Casearia silvana Schltr. Flacourtiaceae Jaffré 1980
Hypernickelophores (>10,000 mg Ni/kg)
Geissois pruinosa Brongn. Cunoniaceae Jaffré et al. 1979
Homalium guillainii Briq. Flacourtiaceae Jaffré 1980
Homalium kanaliense (Vieill.) Briq. Flacourtiaceae This study
Hybanthus austrocaledonicus (Vieill.) Violaceae Brooks et al. 1977
Schinz and Guillamin ex Melchior
Sebertia acuminata Pierre ex Baillon Sapotaceae Jaffré et al. 1976
2009 R.S. Boyd and T. Jaffré 97
hyperaccumulation, which he defined as “a plant in which a nickel concentration
of at least 1000 mg/kg has been recorded in the dry matter of any
above-ground tissue in at least one specimen growing in its natural habitat,”
we classified each species into Ni accumulation categories (Table 1) based
on either values reported in the literature or the greatest Ni value generated
among all samples in our dataset.
Leaf sampling and analysis
We searched sample sites for plants with leafy branches low enough to
be reached from the ground. In the case of tree species, we often used relatively
young individuals growing in roadside disturbed areas. Leaves were
divided into relative age categories, basing leaf-age on the position of leaves
along a branch. Samples were collected by clipping a leafy branch (at a point
closer to the trunk from the attachment point of the oldest leaf) and removing
the leaves in order from the tip to the cut end of the branch, dividing the
leaves into three groups of approximately equal biomass. Those closest to
the cut end were labeled “old,” those at the apex were “young,” and those
in between were “intermediate” in age. For one species, Geissois pruinosa,
we were able to define four leaf age categories due to the presence of rapidly
expanding light-colored young leaves at the distal ends of branches, these
last being considered “very young” leaves.
Leaf samples were dried for at least 72 h at 60 °C, ground, dry-ashed at
485 °C, and further oxidized using 1 M HNO3, and the residue then was redissolved
in 1 M HCl. Concentrations of Ca, Co, Cr, Cu, Fe, K, Mg, Mn, P, Pb,
and Zn were determined by an inductively coupled argon plasma spectrometer
(Jarrell-Ash, ICAP 9000). Nickel concentrations were determined using an
atomic absorption spectrophotometer (Instrumentation Laboratory, IL 251).
Data analysis
Data from old leaves were used to compare element concentrations among
species and to investigate associations among levels of elements. We focused
these analyses on old leaves because our definition of intermediate and young
leaves was contingent upon our initial identification of leaves comprising the
“old” category; thus, old leaves provide a relatively consistent leaf-age category
for comparisons between species. Element concentrations of old leaves
were compared among species using a separate one-way analysis of variance
(ANOVA) for each element. Fisher’s protected least significant difference
(PLSD) test was used for post-hoc mean separations.
We searched for associations between concentrations of Ni and concentrations
of other transition metals (Co, Cr, Cu, Fe, Mg, Mn, Pb, Zn) using
correlations of values for old leaves of each species. Again, we used data
from old leaves because we believed that the “old” age category was defined
most consistently among species. These correlations were conducted for
data from each species separately, and then for all species using the mean
metal concentrations of old leaves for each species. Finally, we examined the
influence of leaf age on the concentrations of all transition metals analyzed
98 Northeastern Naturalist Vol. 16, Special Issue 5
(Co, Cr, Cu, Fe, Mg, Mn, Ni, Pb, Zn) using separate one-way ANOVAs for
each species followed by Fisher’s PLSD test.
Statistical analyses used StatView 5.0 (Abacus Concepts 1998). Because
of the large number of statistical tests comprising all the preceding analyses,
we used a more conservative alpha (0.001) as our criterion for statistical significance for all analyses and mean separations to decrease our probability of
committing Type I error (Zar 1996).
Results
Elemental concentrations
Considerable variation in concentration was observed for all elements
(Table 2). The highest mean value of an element in any species was 20,000
mg Ni/kg in Hybanthus austrocaledonicus, although some high values for
Ca were found for some species (16,000 mg/kg for Grevillea gillivrayi and
Geissois pruinosa; Table 2). Comparing mean values by dividing the greatest
species’ mean by the smallest mean in Table 2 showed Ni to vary more
between species than any other element (1050-fold). Means of two metals
(Co and Pb) could not be compared in that way because the lowest mean
values were below detection limits for at least one species.
All elements excepting P varied significantly among species (Table 2).
Calcium and K varied comparatively little, 4–fold for Ca and 7–fold for
K, and no general trends in levels of these elements among the Ni accumulator
categories were discernable (Table 2). Among transition metals,
variation was least for Cu and greatest for Ni, with Mn, Co, and Cr also varying
greatly (60–fold or greater). As expected, mean Ni values increased from
non-accumulator to accumulator to hyperaccumulator to hypernickelophore
categories (Table 2), but no trends in other metal concentrations emerged
from inspection of Table 2. There was considerable variation in Ni values
within and among species, so that there were no clear divisions of species’ Ni
concentrations that corresponded to Ni accumulation categories (non-accumulator,
accumulator, etc.). Besides the extremely high Ni concentrations of
some hypernickelophores (Hybanthus austrocaledonicus with 20,000 mg/kg
and Sebertia acuminata with 14,000 mg/kg), we documented extremely high
concentrations of Mn in Garcinia amplexicaulis (6300 mg/kg) and Grevillea
gillivrayi (3500 mg/kg), of Co in Homalium kanaliense (450 mg/kg) and of
Cr in Agatea longipedicillata (170 mg/kg).
Relationships between concentrations of Ni and other metals
Comparing correlations of leaf Ni concentrations against concentrations
of other metals in old leaves of each species revealed no significant relationships
for any species (results were P > 0.001 in all cases: data not shown).
On the other hand, testing correlations of the mean Ni concentrations of
old leaves of all species against concentrations of other metals showed two
significant positive relationships (P < 0.001): with Fe and Pb. Correlation
coefficients were 0.87 for Ni and Fe, and 0.97 for Ni and Pb. Correlations
2009 R.S. Boyd and T. Jaffré 99
of Ni against all other transition metals (Co, Cr, Cu, Mg, Mn, Zn) were not
significant (P = 0.034 for Co, P > 0.26 in all other cases; data not shown).
Effect of leaf age on metal concentrations
Leaf age affected concentrations of at least one transition metal for eight
of the 11 species examined (Table 3). Species for which no age effect was
found for any metal included one accumulator (Garcinia amplexicaulis), one
hyperaccumulator (Casearia silvana), and one hypernickelophore (Sebertia
acuminata), revealing no clear trend of leaf-age patterns among species in
different Ni accumulation categories. Two hypernickelophores had signifi-
cant leaf-age effects for a large number of metals (six metals for Geissois
pruinosa and five for Homalium kanaliense), whereas all other species
for which we detected significant leaf-age effects had significant patterns
for only one or two metals (Table 3).
Among metals, Fe was most frequently affected by leaf age, as five species
(Grevillea gillivrayi, Xanthostemon aurantiacus, Geissois pruinosa
and both species of Homalium) had significantly greater Fe concentrations
in older leaves (Table 3). Four species, the Ni hyperaccumulator Agatea
longipedicillata and three hypernickelophores (Geissois pruinosa, Homalium
kanaliense, and Hybanthus austrocaledonicus), showed significant
increases in Pb concentration with increased leaf age. Effects of leaf age
on Co, Cr, and Cu concentrations were found for three species each; these
relationships showed increases in Co and Cr concentrations with increased
leaf age, but decreases in Cu concentrations as leaf age increased (Table 3).
Two species (Geissois pruinosa and Homalium kanaliense, both hypernickelophores)
showed increases in leaf Ni with age, although these were not the
species with greatest leaf Ni concentrations (those were Hybanthus austrocaledonicus
and Sebertia acuminata). One species (Geissois pruinosa) had
increased leaf Zn concentrations with leaf age. Two metals, Mg and Mn, did
not vary significantly with leaf age for any species examined.
Discussion
We found wide differences in the levels of variation of different elements.
Levels of Ni varied more among species than for any other element
documented (Table 2), showing that we captured considerable variability
among the species chosen for sampling. Variation also was high for most elements,
including Ni, among samples within a species, indicating that other
factors besides leaf age and plant identity affect leaf element concentrations.
These other factors may include plant size, rooting depths, variation in soil
properties between sites, or genetic variability within species.
Our results for leaf age (Table 3) were generally consistent with what is
known about the mobility of the elements examined (Kabata-Pendias 2000).
For example, Fe was the metal for which a significant leaf-age effect was
found for five of the 11 species, and Fe is generally considered an immobile
element (Kabata-Pendias 2000). Departures from these expectations may be
100 Northeastern Naturalist Vol. 16, Special Issue 5
Table 2. Mean element concentrations of old leaves for all sampled species (in mg/kg dry mass, SE in parentheses below each mean). Mean values for elements
that differ significantly between species (Fisher’s PLSD test, α = 0.001) are denoted by differing superscripts. Sample sizes (number of plants sampled) for
each species’ means are reported under each species’ name. Concentrations of Co and Pb that were below detection limits (<1 mg/kg) are denoted by “N.D.”
(not detected).
Element
Species Ca K P Co Cr Cu Fe Mg Mn Ni Pb Zn
Non-accumulator
Grevillea gillivrayi 16,000C 2000A 210 N.D. 3.9A 2.2AB 130AB 3300AB 3500B 19A 1.1A 15A
(n = 6) (2300) (340) (17) (1.2) (0.29) (17) (560) (820) (6.7) (0.70) (3.3)
Ni accumulators
Garcinia amplexicaulis 7000AB 3800AB 160 3.9A 14AB 3.8ABC 180AB 1300A 6300C 110AB 6.4AB 130AB
(n = 7) (1100) (480) (21) (2.8) (9.5) (0.32) (84) (170) (990) (25) (2.9) (17)
Pyschotria baillonii 15,000C 11,000DE 570 N.D. 22AB 6.5C 330AB 8400DE 34A 550AB 2.9A 93AB
(n = 7) (1000) (1800) (17) (4.3) (0.26) (55) (1300) (13) (110) (0.77) (49)
Xanthostemon aurantiacus 12,000BC 4400ABC 140 N.D. 3.4A 1.9A 140AB 1200A 110A 160AB N.D. 24A
(n = 7) (2100) (480) (7.8) (1.2) (0.22) (17) (170) (64) (37) (3.9)
Ni hyperaccumulators
Agatea longipedicillata 8400ABC 9100CDE 510 7.7A 170D 13D 180AB 11,000DE 1100A 1000ABC 6.8ABC 230AB
(n = 4) (450) (1000) (12) (1.1) (22) (1.3) (66) (2000) (59) (270) (1.1) (20)
Casearia silvana 10,000BC 5900ABCD 1800 3.1A 2.8A 5.5ABC 320AB 12,000E 130A 1100ABC 5.1AB 260B
(n =7) (820) (1900) (1300) (0.87) (0.49) (0.53) (97) (1000) (20) (220) (0.81) (82)
2009 R.S. Boyd and T. Jaffré 101
Table 2, continued.
Element
Species Ca K P Co Cr Cu Fe Mg Mn Ni Pb Zn
Hypernickelophores
Homalium guillainii 12,000BC 5100ABC 490 20A 12A 6.0C 570BC 6900BCD 80A 6300CD 21DE 150AB
(n = 7) (740) (850) (16) (2.8) (1.9) (0.37) (92) (440) (27) (720) (1.5) (12)
Homalium kanaliense 11,000BC 14,000E 230 450C 4.3A 5.5ABC 390AB 1700A 560A 9900DE 17CD 510C
(n = 6) (2700) (1800) (21) (54) (1.5) (0.33) (36) (300) (83) (1000) (1.8) (110)
Geissois pruinosa 16,000C 7900BCD 280 37A 41ABC 5.5ABC 300AB 4200ABC 280A 8800DE 16BCD 67AB
(n = 6) (3100) (330) (18) (6.0) (12) (0.47) (53) (380) (73) (1800) (2.6) (8.0)
Hybanthus austrocaledonicus 8200ABC 5000ABC 630 150B 87C 6.3C 1700C 7900CDE 280A 20,000F 41F 120AB
(n = 6) (910) (930) (38) (54) (18) (0.33) (580) (470) (37) (1600) (2.9) (7.6)
Sebertia acuminata 3700A 3800ABC 420 32A 58BC 6.9C 1600C 3000A 250A 13,000E 30E 93AB
(n = 8) (660) (750) (70) (6.8) (16) (1.7) (460) (780) (64) (2000) (4.2) (11)
Highest species mean 16,000 14,000 1800 450 170 13 1700 12,000 6300 20,000 41 510
Lowest species mean 3700 2000 140 N.D. 2.8 1.9 130 1200 34 19 N.D. 15
Comparison (high/low)† 4× 7× 13× ≥450׆† 61× 7× 13× 10× 185× 1050× ≥41׆† 34×
†This row compares the highest species mean for an element to the lowest species mean. Since the lowest species means for Co and Pb were below detection
limits (<1 mg/kg), definite comparisons of highest and lowest species means could not be made for these elements.
††These values are estimates because the lowest values for these metals were below detection limits (<1 mg/kg), thus comparisons of highest and lowest species
means for these metals are probably underestimates.
102 Northeastern Naturalist Vol. 16, Special Issue 5
Table 3. Mean metal concentrations (in mg/kg dry mass, SE in parentheses after each mean). Means for which a significant effect of leaf age was detected are
denoted with a superscripted letter. Mean values that differ significantly between leaves of different age for a species (Fisher’s PLSD test, α = 0.001) are denoted
by differing superscripts. Sample sizes (number of plants sampled) are reported for each species.
Metal
Species Co Cr Cu Fe Mg Mn Ni Pb Zn
Grevillea gillivrayi (n = 6; Non-accumulator)
Young 0.0 (0.0) 0.0 (0.0) 4.2 (0.28) 50A (3.5) 2200 (150) 870 (250) 10 (3.2) 0.0 (0.0) 17 (2.4)
Intermediate 0.0 (0.0) 1.0 (0.64) 2.9 (0.52) 98AB (14) 3100 (350) 2300 (620) 15 (3.3) 0.33 (0.19) 12 (1.0)
Old 0.0 (0.0) 3.9 (1.2) 2.2 (0.29) 130B (17) 3300 (560) 3500 (820) 19 (7.8) 1.1 (0.70) 15 (3.3)
Psychotria baillonii (n = 7; Ni accumulator)
Young 0.0 (0.0) 1.6A (0.73) 8.3 (0.52) 130 (20) 5200 (860) 0.0 (0.0) 210 (63) 0.0 (0.0) 140 (34)
Intermediate 0.0 (0.0) 13AB (2.1) 7.0 (0.75) 250 (40) 9100 (1500) 4.0 (2.0) 290 (93) 1.8 (0.87) 270 (140)
Old 0.0 (0.0) 22B (4.3) 6.5 (0.26) 330 (55) 8400 (1300) 34 (13) 550 (110) 2.9 (0.77) 93 (49)
Xanthostemon aurantiacus (n = 7; Ni accumulator)
Young 0.0 (0.0) 0.0 (0.0) 5.5B (0.69) 41A (1.9) 1800 (270) 7.2 (3.6) 31 (9.8) 0.0 (0.0) 23 (1.6)
Intermediate 0.0 (0.0) 0.60 (0.46) 3.7AB (0.62) 93AB (15) 1400 (190) 21 (9.9) 78 (12) 0.0 (0.0) 26 (2.7)
Old 0.0 (0.0) 3.4 (1.2) 1.9A (0.22) 140B (17) 1200 (170) 110 (64) 160 (37) 0.0 (0.0) 24 (3.9)
Garcinia amplexicaulis (n = 7; Ni accumulator)
Young 0.45 (0.45) 1.2 (0.51) 6.8 (0.89) 45 (5.7) 1300 (110) 4300 (600) 37 (8.0) 1.5 (0.74) 69 (19)
Intermediate 1.5 (1.5) 2.7 (0.58) 5.0 (0.84) 63 (7.1) 1100 (180) 5800 (410) 140 (87) 2.9 (0.94) 110 (15)
Old 3.9 (2.8) 4.8 (1.4) 3.8 (0.32) 180 (84) 1300 (170) 6300 (990) 110 (25) 6.4 (2.9) 130 (17)
Agatea longipedicillata (n = 4; Ni hyperaccumulator)
Young 0.0A (0.0) 20 (5.0) 37 (8.6) 75 (8.3) 5700 (340) 330 (57) 330 (61) 0.0A (0.0) 91 (4.2)
Intermediate 4.7AB (1.2) 140 (35) 13 (1.2) 250 (110) 9900 (1400) 720 (220) 600 (150) 5.5AB (0.55) 190 (25)
Old 7.7B (1.1) 170 (22) 13 (1.3) 180 (66) 11,000 (2000) 1100 (59) 1000 (270) 6.8B (1.1) 230 (20)
2009 R.S. Boyd and T. Jaffré 103
Table 3, continued.
Metal
Species Co Cr Cu Fe Mg Mn Ni Pb Zn
Casearia silvana (n = 7; Ni hyperaccumulator)
Young 1.1 (0.57) 1.9 (1.7) 6.4 (0.93) 140 (21) 8400 (920) 95 (21) 460 (88) 2.2 (0.93) 72 (8.5)
Intermediate 2.2 (0.44) 2.1 (1.2) 5.4 (0.24) 270 (64) 11,000 (650) 150 (32) 850 (150) 5.1 (0.45) 100 (22)
Old 3.1 (0.87) 2.8 (0.49) 5.5 (0.53) 320 (97) 12,000 (1000) 130 (20) 1100 (220) 5.1 (0.81) 260 (82)
Homalium guillainii (n = 7; Ni hyperaccumulator)
Young 8.2 (2.7) 0.0A (0.0) 7.9 (0.94) 73a (12) 4200 (990) 43 (22) 3100 (530) 14 (5.5) 110 (16)
Intermediate 20 (3.5) 2.2A (0.86) 6.9 (0.28) 210a,b (33) 5900 (580) 85 (29) 5800 (440) 22 (0.91) 160 (14)
Old 20 (2.8) 12B (1.9) 6.0 (0.37) 570b (92) 6900 (440) 80 (27) 6300 (720) 21 (1.5) 150 (12)
Geissois pruinosa (n = 6; hypernickelophore)
Very young 4.2A (0.82) 1.4 (1.4) 19B (1.1) 100A (28) 4500 (440) 180 (26) 960A (180) 0.16A (0.16) 30A (3.1)
Young 8.3A (2.1) 3.2 (2.7) 15B (1.2) 82A (15) 4600 (230) 200 (54) 1900A (430) 1.8A (1.5) 30A (2.1)
Intermediate 31B (4.2) 34 (8.4) 8.4A (0.63) 180AB (22) 5200 (400) 330 (76) 5400AB (720) 8.6AB (1.5) 47AB (4.1)
Old 37B (6.0) 41 (12) 5.5A (0.47) 300B (53) 4200 (380) 280 (73) 9100B (2000) 16B,C (2.6) 67BC (8.0)
Homalium kanaliense (n = 6; hypernickelophore)
Young 85A (15) 0.0 (0.0) 8.7B (0.49) 65A (6.5) 3100 (310) 210 (33) 2300A (340) 0.58A (0.22) 190 (9.6)
Intermediate 330B (30) 1.4 (0.81) 6.1B (0.32) 240B (28) 2900 (320) 470 (86) 8000B (480) 13B (1.3) 470 (63)
Old 450B (46) 4.3 (1.3) 5.3A (0.32) 380C (31) 1800 (260) 510 (83) 9900B (890) 17B (1.6) 490 (91)
Hybanthus austrocaledonicus (n = 6; hypernickelophore)
Young 39 (14) 5.7A (1.4) 7.2 (0.40) 120 (17) 6800 (970) 120 (21) 17,000 (590) 26A (2.5) 110 (11)
Intermediate 98 (27) 49AB (10) 6.7 (0.31) 660 (170) 9300 (640) 250 (42) 20,000 (1100) 45B (0.86) 130 (7.2)
Old 150 (54) 87B (18) 6.3 (0.33) 1700 (580) 7900 (470) 280 (37) 20,000 (1800) 41B (2.9) 120 (7.6)
Sebertia acuminata (n = 8; hypernickelophore)
Young 14 (1.5) 9.6 (3.5) 8.4 (0.56) 190 (42) 2200 (280) 140 (26) 14,000 (1400) 22 (2.5) 63 (8.9)
Intermediate 18 (0.73) 25 (4.1) 6.6 (1.3) 590 (110) 2200 (660) 150 (28) 12,000 (2300) 25 (3.4) 72 (12)
Old 36 (6.3) 63 (17) 7.4 (1.9) 1800 (500) 3200 (880) 290 (64) 14,000 (2100) 33 (2.6) 98 (10)
104 Northeastern Naturalist Vol. 16, Special Issue 5
due to physiological differences between species, but also may stem from
other factors. These other factors may include differences in growth rate of
the species examined. Because we defined leaf age in a relative manner, effects
of leaf age on element concentrations may have been difficult to detect
for species with considerably long-lived leaves.
Nickel levels for some species reported here differed from those reported
in earlier literature; as a result, we classified some species into Ni accumulation
categories that differ from earlier reports. For example, Jaffré (1980)
reported a value of 21 mg Ni/kg for Xanthostemon aurantiacus, but we found
old leaves had a mean value of 160 mg/kg and thus classified that species
as a Ni accumulator. Jaffré (1980) also reported Homalium kanaliense as a
Ni hyperaccumulator, but we found some samples of old leaves to contain
>10,000 mg Ni/kg (Table 2) and so classified it as a hypernickelophore in
Table 1. In another case, we found lower Ni values than expected: Jaffré
(1980) reported Homalium guillainii as a hypernickelophore, but our mean
values were between 3100 and 6300 mg Ni/kg (Table 3), putting our specimens
into the hyperaccumulator range.
As illustrated above, the Ni-concentration boundaries used to categorize
species are difficult to use consistently. We agree with Reeves (1992)
and Macnair (2003) that the boundaries that separate categories of Ni accumulation
are artificial and are not based upon natural discontinuities.
For example, our data (Tables 2 and 3) did not show a clear separation
of species into Ni hyperaccumulator and hypernickelophore categories.
On the other hand, we think it is helpful to divide the wide range of Ni
(and other metal) concentrations of plants into categories to aid in discussions
of their properties (such as their physiologies or their ecological
effects on other organisms). Using orders of magnitude to define category
boundaries is particularly convenient, and we suggest doing so with Ni
even though this approach probably does not reflect clear biological
boundaries. For Ni, we suggest non-accumulator (<100 mg/kg), hemi-accumulator
(100–999 mg/kg), hyperaccumulator (1000–9999 mg/kg) and
hypernickelophore categories (≥10,000 mg/kg). We particularly suggest
that hypernickelophore be retained, as plants with Ni concentrations of
that magnitude may be particularly important for their ecological effects,
such as mobilizing metals into terrestrial food webs (Boyd 2004). The use
of “accumulator” for plants that accumulate to below hyperaccumulator
status is problematic because the term “accumulator” has been used generically
to denote plants with relatively high metal concentrations (e.g.,
Reeves and Adigüzel 2004). If accumulator is used generally to refer to
plants with greater than normal Ni concentrations, then another term is
needed for those in the sub-hyperaccumulator range. Therefore, we suggest
use of a new term, “hemi-accumulator,” to describe those plants
with 100–999 mg Ni/kg. The Greek-derived prefix means “half” and thus
conveys that these plants accumulate Ni, but not to the extremes of hyperaccumulators
or hypernickelophores.
2009 R.S. Boyd and T. Jaffré 105
To be operationally useful, Ni accumulation categorical terms need stringent
definition. Following the criteria of Reeves (1992), in which he defined
a Ni hyperaccumulator as “a plant in which a nickel concentration of at least
1000 mg/kg has been recorded in the dry matter of any above-ground tissue
in at least one specimen growing in its natural habitat,” we suggest that the
terms “non-accumulator,” “hemi-accumulator,” and “hypernickelophore”
be similarly defined based upon records of Ni concentration of <100 mg/kg,
100–999 mg/kg, and ≥10,000 mg/kg, respectively, in the dry matter of any
above-ground tissue in at least one specimen growing in its natural habitat.
We recognize that, as pointed out by Macnair (2003), only a single specimen
is needed to elevate the classification of a species’ Ni accumulation category,
and thus these categories can become “over-reported.” Still, we conclude
it is necessary to have a discrete definition so that use of these terms can be
standardized.
The inadvertent uptake hypothesis for metal hyperaccumulation (Boyd
and Martens 1992) suggests that metal hyperaccumulation might have
evolved from physiological mechanisms that target uptake and sequestration
of one or more other soil ions. This hypothesis has received some support in
the literature on Zn hyperaccumulation (Macnair 2003, Taylor and Macnair
2006) but to our knowledge, there is no direct evidence of shared uptake and
sequestration mechanisms for Ni and other elements (Callahan et al. 2006).
However, this hypothesis is consistent with strong positive correlations
between tissue levels of Ni and another soil constituent. In our study, correlations
showed significant positive relationships between the concentrations
of Ni and Fe, and Ni and Pb, in leaves of the studied species. These results
may stem from similar uptake and sequestration pathways for these metals.
Because Pb is not known to be an essential element for plants (Pais and
Jones 1997), it is difficult to envision a functional reason for enhancing Pb
uptake (of course, Pb might be taken up inadvertently along with Ni). Iron,
however, is a required plant micronutrient (Pais and Jones 1997), and our
results may reflect linked uptake and sequestration pathways of Ni and Fe in
these species. It is also plausible that Ni uptake and sequestration increase
the Fe requirement of hyperaccumulator and hypernickelophore plants, although
we know of no data that bear directly on this question.
We note that the concept of inadvertent uptake has previously focused
on explaining the high concentrations of hyperaccumulated metals in plants,
suggesting that another element was the target of uptake and sequestration
mechanisms. Our results suggest another possibility: the correlations we
found between Ni and both Pb and Fe may reflect inadvertent uptake not of
Ni, but of Fe and Pb. It is possible that the extremely active Ni uptake and
sequestration mechanisms characteristic of hyperaccumulators and hypernickelophores
(Callahan et al. 2006) result in inadvertent uptake of other
elements (e.g., Pb and Fe). In this scenario, the small amounts of Pb and Fe
(relative to Ni) in these plants (Tables 2, 3) may be viewed as being inadvertently
captured during the acquisition and transport of Ni.
106 Northeastern Naturalist Vol. 16, Special Issue 5
Previous studies (e.g., Brooks and Yang 1984, Yang et al. 1985) investigating
elemental correlations in Ni hyperaccumulators have reported
positive correlations between Ni and some elements (e.g., Ni with Co and Cr).
Yang et al. (1985) reported a relatively weak positive Ni-Fe correlation. We
found no significant correlations of Ni with other element concentrations
when the analyses were done within species, but significant correlations
when we used mean values for our species from differing Ni accumulation
categories. This result suggests that uptake of all three metals (Fe, Ni, and
Pb) differs based upon the Ni accumulation category of these species. We
know of few physiological studies of Ni hyperaccumulators that have investigated
the specificity of Ni transport and sequestration mechanisms. For
example, Gabbrielli et al. (1991) reported competition of Ni with Co and Zn
in roots of the Ni hyperaccumulator Alyssum bertolonii Desv (Brassicaceae).
Assunção et al. (2003) reported variability of Cd, Ni, and Zn uptake among
populations of Thlaspi caerulescens (Alpine Penny-cress) (Brassicaceae),
and concluded this might be due to variable expression of multiple metal
transporters.
Elevated leaf metal levels also can result from the presence of dust on
leaf surfaces (Reeves et al. 1999, 2007). Reeves et al. (2007) pointed out that
samples containing more than 1500 mg Fe/kg (as well as high Cr and Ni)
might indicate dust contamination. Fortunately, in our study, such high Fe
values were found for only two species (Table 2), both hypernickelophores
(Hybanthus austrocaledonicus, Sebertia acuminata). Although we did not
wash our samples, Reeves et al. (1999, 2007) pointed out that dust contamination
is not always readily removed in this way. We do not think the Ni-Fe
correlation we observed across species is due to dust contamination, but it
is possible that dust accumulation may have varied among species in such a
way that those species with greatest Ni concentrations had greater amounts
of dust on their leaves. This possibility seems improbable, however, given
the relatively strong relationship we found between Ni and Fe and the fact
that species with differing Ni accumulation categories were collected at each
field site. Thus, it is unlikely that site-specific contamination differences
would have resulted in the relationship we observed between Ni and Fe
among species in different Ni accumulation categories.
Metal hyperaccumulation may function as an elemental defense against
plant natural enemies (Martens and Boyd 1994). Unfortunately, we do not
know what metal concentration is sufficient to protect against plant natural
enemies (Boyd 2004); it likely varies based on tissue and cellular level metal
distribution in the plant, natural enemy feeding mode, and natural enemy
physiology (Boyd 2007). For Ni, concentrations at hyperaccumulator levels
suffice against some natural enemies but are ineffective against others
(Jhee et al. 2006). In the context of plant elemental defense, our data suggest
several conclusions. First, it is clear that leaf Ni concentrations vary greatly
(1050-fold among old leaves in this study) among species growing on New
Caledonian serpentine soils. If Ni is defensively valuable against some natural
enemies, then some species are much better protected by this defense than
others. Studies of organic chemical defenses in tropical forest trees have illus2009
R.S. Boyd and T. Jaffré 107
trated that variability in defense levels can help explain patterns of herbivore
damage and plant defense strategies (Kursar and Coley 2003). It is also clear
that, in general, leaf age does not significantly affect leaf Ni concentration
(Table 3). Thus, in most species, young and old leaves are defended to an equal
degree by this elemental defense. Exceptions did occur, such as Geissois pruinosa
and Homalium kanaliense (Table 3), in which young leaves contained
significantly less Ni and thus may have been less well defended. Similar cases
of reduced Ni concentrations of young leaves have been reported by Boyd et
al. (1999) for the New Caledonian hypernickelophore Psychotria douarrei
and by Anderson et al. (1997) and Boyd et al. (2004) for the South African hypernickelophore
Berkheya coddii Roessler (Asteraceae).
Our data are relevant to another aspect of elemental plant defense:
combination effects of chemical defenses (Boyd 2007). Boyd (2004) hypothesized
that elemental defenses may not act alone in generating defensive
benefits to plants. The effects of one element may combine with those of
another element, or with those of an organic defense compound, to generate
a greater defensive effect in combination than either chemical alone. To our
knowledge, the study of Jhee et al. (2006) is the only one to experimentally address
this issue (using an artificial insect diet system). They found significant
joint defensive effects between Ni and several other metals (Cd, Pb, Zn). Our
data show that some New Caledonian species can have elevated concentrations
of multiple metals, and we suggest these may enhance the defensive
effect of Ni in some species. For example, we found: Homalium kanaliense
with 9900 mg Ni/kg, 450 mg Co/kg, and 510 mg Zn/kg; Hybanthus autrocaledonicus
with 20,000 mg Ni/kg, 7900 mg Mg/kg, and 150 mg Co/kg;
Casearia silvana with 1100 mg Ni/kg, 12,000 mg Mg/kg, and 170 mg Cr/
kg; and Agatea longipedicillata with 1000 mg Ni/kg and 11,000 mg Mg/kg
(Table 2). Defensive effect enhancement by multiple metals may occur in
two ways. First, a combination of metals may provide a greater defensive effect
against a single natural enemy, as shown by Jhee et al. (2006) against a
leaf-chewing insect. On the other hand, if enemies differ in their sensitivity to
metals, each metal in a combination may provide a defensive benefit against
a different natural enemy. In a hypothetical example, the high level of Co in
Homalium kanaliense leaves may defend against a pathogenic bacterium,
whereas the high level of Ni might defend against a folivorous insect. In this
sense, a combination of metals may extend elemental defenses by being effective
against a broader collection of natural enemies. Additional research
testing these defensive effects, guided by the levels of metals revealed by our
study, may illuminate these functions of metal accumulation by plants.
Acknowledgments
We thank Dr. Michael Wall for assistance with fieldwork, Dr. James E. Watkins,
Jr. and Dr. John Odom for assistance with plant analyses, and Dr. Roger Reeves and
two anonymous reviewers for helpful suggestions regarding an earlier version of the
manuscript. Dr. Roland Dute suggested the prefix for the new term, hemi-accumulator.
We also thank Auburn University for providing travel funds for this research.
108 Northeastern Naturalist Vol. 16, Special Issue 5
Literature Cited
Abacus Concepts. 1998. StatView. SAS Institute, Inc., Cary, NC, USA.
Alexander, E.B., R.G. Coleman, T. Keeler-Wolf, and S.P. Harrison. 2007. Serpentine
Geoecology of Western North America: Geology, Soils, and Vegetation. Oxford
University Press, New York, NY, USA. 512 pp.
Anderson, T.R., A.W. Howes, K. Slatter, and M.F. Dutton. 1997. Studies on the nickel
hyperaccumulator, Berkheya coddii. Pp. 261–266, In T. Jaffré, R.D. Reeves, and
T. Becquer (Eds.). The Ecology of Ultramafic and Metalliferous Areas. Centre
ORSTOM de Noumea, New Caledonia. 306 pp.
Assunção, A.G.L., W.M. ten Bookum, H.J.M. Nelissen, R. Vooijs, H. Schat, and
W.H.O. Ernst. 2003. Differential metal-specific tolerance and accumulation patterns
among Thlaspi caerulescens populations originating from different soil
types. New Phytologist 159:411–419.
Berazaín, R. 2004. Notes on tropical American nickel hyperaccumulating plants.
Pp. 255–258, In R.S. Boyd, A.J.M. Baker, and J. Proctor (Eds.). Ultramafic
Rocks: Their Soils, Vegetation, and Fauna. Science Reviews Ltd, St Albans,
UK. 347 pp.
Boyd, R.S. 1998. Hyperaccumulation as a plant defensive strategy. Pp. 181–201, In
R.R. Brooks (Ed.). Plants that Hyperaccumulate Heavy Metals: Their Role in
Phytoremediation, Microbiology, Archaeology, Mineral Exploration, and Phytomining.
CAB International, Wallingford, UK. 380 pp.
Boyd, R.S. 2004. Ecology of metal hyperaccumulation. New Phytologist 162:563–
567.
Boyd, R.S. 2007. The defense hypothesis of elemental hyperaccumulation: Status,
challenges, and new directions. Plant and Soil 293:153–176.
Boyd, R.S., and T. Jaffré. 2001. Phytoenrichment of soil Ni concentration by Sebertia
acuminata in New Caledonia and the concept of elemental allelopathy. South
African Journal of Science 97:535–538.
Boyd, R.S., and S.N. Martens. 1992. The raison d’être for metal hyperaccumulation
by plants. Pp. 279–289, In A.J.M. Baker, J. Proctor, and R.D. Reeves (Eds.). The
Vegetation of Ultramafic (Serpentine) Soils. Intercept, Andover, UK. 509 pp.
Boyd, R.S., and S.N. Martens. 1998. The significance of metal hyperaccumulation
for biotic interactions. Chemoecology 8:1–7.
Boyd, R.S., T. Jaffré, and J.W. Odom. 1999. Variation of nickel content in the nickelhyperaccumulating
shrub Psychotria douarrei (Rubiaceae) from New Caledonia.
Biotropica 31:403–410.
Boyd, R.S., M.A. Davis, and K. Balkwill. 2004. Nickel accumulation patterns in two
South African Ni hyperaccumulator species. Pp. 275–278, In R.S. Boyd, A.J.M
Baker, and J. Proctor (Eds.). Ultramafic Rocks: Their Soils, Vegetation, and
Fauna. Science Reviews 2000 Ltd., St. Albans, UK. 347 pp.
Boyd, R.S., M.A. Wall, and T. Jaffré. 2006. Nickel levels in arthropods associated
with Ni hyperaccumulator plants from an ultramafic site in New Caledonia. Insect
Science 13:271–277.
Brenes-Arguedas, T., and P.D. Coley. 2005. Phenotypic variation and spatial structure
of secondary chemistry in a natural population of a tropical tree species.
Oikos 108:410–420.
Brenes-Arguedas, T., M.W. Horton, P.D. Coley, J. Lokvam, R.A. Waddell, B.E.
Meizoso-O’Meara, and T.A. Kursar. 2006. Contrasting mechanisms of secondary
metabolite accumulation during leaf development in two tropical tree species
with different leaf-expansion strategies. Oecologia 149:91–100.
2009 R.S. Boyd and T. Jaffré 109
Brooks, R.R. 1987. Serpentine and its Vegetation: A Multi-disciplinary Approach.
Dioscorides Press, Portland, OR, USA. 454 pp.
Brooks, R.R., and X. Yang. 1984. Elemental levels and relationships in the endemic
serpentine flora of the Great Dyke, Zimbabwe and their significance as controlling
factors for the flora. Taxon 33:392–399.
Brooks, R.R., J. Lee, R.D. Reeves, and T. Jaffré. 1977. Detection of nickeliferous
rocks by analysis of herbarium specimens of indicator plants. Journal of Geochemical
Exploration 7:49–57.
Callahan, D.L., A.J.M. Baker, S.D. Kolev, and A.G. Wedd. 2006. Metal ion ligands in
hyperaccumulating plants. Journal of Biological Inorganic Chemistry 11:2–12.
Davis, M.A., S.G. Pritchard, R.S. Boyd, and S.A. Prior. 2001. Developmental and
induced responses of nickel-based and organic defences of the nickel-hyperaccumulating
shrub, Psychotria douarrei. New Phytologist 150:49–58.
Gabbrielli, R., C. Mattioni, and O. Vergnano. 1991. Accumulation mechanisms and
heavy metal tolerance of a nickel hyperaccumulator. Journal of Plant Nutrition
14:1067–1080.
Galeas, M.L., L.H. Zhang, J.L. Freeman, M. Wegner, and E.A.H. Pilon-Smits. 2007.
Seasonal fluctuations of selenium and sulfur accumulation in selenium hyperaccumulators
and related nonaccumulators. New Phytologist 173:517–525.
Jaffré, T. 1980. Étude écologique du peuplement végétal des sols dérivés de roches
ultrabasiques en Nouvelle-Calédonie. Travaux et Documents de L’ORSTOM,
No. 124. O.R.S.T.O.M., Paris, France. 273 pp.
Jaffré, T. 2005. Conservation programmes in New Caledonia, western Pacific: In
place for the dry forest, but urgently needed for the ultramafic vegetation. BGjournal
2:13.
Jaffré, T., and M. Schmid. 1974. Accumulation du nickel par une Rubiacée de
Nouvelle-Calédonie, Psychotria douarrei (G. Beauvisage) Däniker. Comptes
Rendus, Academie des Sciences, Paris, France. Series D 278:1727–1730.
Jaffré, T., and J.-M. Veillon. 1991. Etude floristique et structurale de deux forêts
denses humides sur roches ultrabasiques en Nouvelle-Calèdonie. Bulletin du
Museum National d’Histoire Naturelle de Paris, 4e Serie, Section B, Adansonia
12:243–273.
Jaffré, T., R.R. Brooks, J. Lee, and R.D. Reeves. 1976. Sebertia acuminata: A hyperaccumulator
of nickel from New Caledonia. Science 193:579–580.
Jaffré, T., R.R. Brooks, and J.M. Trow. 1979. Hyperaccumulation of nickel by Geissois
species. Plant and Soil 51:157–162.
Jhee, E.M., R.S. Boyd, and M.D. Eubanks. 2006. Effectiveness of metal-metal and
metal-organic compound combinations against Plutella xylostella (Lepidoptera:
Plutellidae): Implications for plant elemental defense. Journal of Chemical Ecology
32:239–259.
Kabata-Pendias, A. 2000. Trace Elements in Soils and Plants. CRC Press, Boca Raton,
fl, USA. 432 pp.
Kruckeberg, A.R. 2002. Geology and Plant Life: The Effects of Landforms and Rock
Types on Plants. University of Washington Press, Seattle, WA, USA. 362 pp.
Kursar, T.A., and P.D. Coley. 2003. Convergence in defense syndromes of young
leaves in tropical rainforests. Biochemical Systematics and Ecology 31:929–949.
Macnair, M.R. 2003. The hyperaccumulation of metals by plants. Advances in Botanical
Research 40:63–105.
Martens, S.N., and R.S. Boyd. 1994. The ecological significance of nickel hyperaccumulation:
A plant chemical defense. Oecologia 98:379–384.
110 Northeastern Naturalist Vol. 16, Special Issue 5
Mittermeier, R.A., P.R. Gil, M. Hoffman, J. Pilgrim, T. Brooks, C.G. Mittermeier, J.
Lamoreux, and G.A.B. da Fonseca. 2005. Hotspots Revisited: Earth’s Biologically
Richest and Most Threatened Terrestrial Ecoregions. University of Chicago
Press, Chicago, IL, USA. 380 pp.
Nicks, L.J., and M.F. Chambers. 1998. A pioneering study of the potential for phytomining
of nickel. Pp. 313–325, In R.R. Brooks (Ed.). Plants that Hyperaccumulate
Heavy Metals: Their Role in Phytoremediation, Microbiology, Archaeology,
Mineral Exploration, and Phytomining. CAB International, Wallingford, UK.
380 pp.
Pais, I., and J.B. Jones, Jr. 1997. The Handbook of Trace Elements. St. Lucie Press,
Boca Raton, fl, USA. 223 pp.
Pilon-Smits, E. 2004. Phytoremediation. Annual Reviews in Plant Biology 56:15–39.
Raskin, I., and B.D. Ensley. 2000. Phytoremediation of Toxic Metals. John Wiley and
Sons, New York, NY, USA. 304 pp.
Reeves, R.D. 1992. The hyperaccumulation of nickel by serpentine plants. Pp.
253–277, In A.J.M. Baker, J. Proctor and R.D. Reeves (Eds.). The Vegetation of
Ultramafic (Serpentine) Soils. Intercept, Andover, UK. 509 pp.
Reeves, R.D. 2003. Tropical hyperaccumulators of metals and their potential for
phytoextraction. Plant and Soil 249:57–65.
Reeves, R.D., and N. Adigüzel. 2004. Rare plants and nickel accumulators from
Turkish serpentine soils, with special reference to Centaurea species. Turkish
Journal of Botany 28:147–153.
Reeves, R.D., and A.J.M. Baker. 2000. Metal-accumulating plants. Pp. 193–229,
In I. Raskin and B.D. Ensley (Eds.). Phytoremediation of Toxic Metals: Using
Plants to Clean up the Environment. John Wiley and Sons, New York, NY, USA.
304 pp.
Reeves, R.D., A.J.M. Baker, A. Borhidi, and R. Berazaín. 1999. Nickel hyperaccumulation
in the serpentine flora of Cuba. Annals of Botany 83:29–38.
Reeves, R.D., A.J.M. Baker, and R. Romero. 2007. The ultramafic flora of the Santa
Elena Peninsula, Costa Rica: A biogeochemical reconnaissance. Journal of Geochemical
Exploration 93:153–159.
Salt, D.E. 2004. Update on plant ionomics. Plant Physiology 136:2451–2456.
Szpunar, J. 2004. Metallomics: A new frontier in analytical chemistry. Analytical and
Bioanalytical Chemistry 378:54–56.
Taylor, S.I., and M.R. Macnair. 2006. Within and between population variation for
zinc and nickel accumulation in two species of Thlaspi (Brassicaceae). New
Phytologist 169:505–514.
Whiting, S.N., R.D. Reeves, D. Richards, M.S. Johnson, J.A. Cooke, F. Malaisse, A.
Paton, J.A.C. Smith, J.S. Angle, R.L. Chaney, R. Ginocchio, T. Jaffré, R. Johns,
T. McIntyre, G.W. Purves, D.E. Salt, H. Schat, F.J. Zhao, and A.J.M. Baker. 2004.
Research priorities for conservation of metallophyte biodiversity and their potential
for restoration and site remediation. Restoration Ecology 12:106–116.
Yang, X.H., R.R. Brooks, T. Jaffré, and J. Lee. 1985. Elemental levels and relationships
in the Flacourtiaceae of New Caledonia. Plant and Soil 87:281–291.
Zar, J.H. 1996. Biostatistical Analysis. Prentice-Hall, Englewood Cliffs, NJ,
USA. 662 pp.