Effect of cd and ni on soybean seed development

Effects of the metal pollutants cadmium and nickel on
soybean seed development
H. L. Malan and J. M. Farrant*
Dept. Botany, University of Cape Town, Private Bag, Rondebosch, 7701, South Africa
Abstract
The chloride salts of Cd or Ni were added to the nutrient
solution in which soybean (
Glycine max
) plants were
grown and the response of the plants to these pollutants
examined. Both metals markedly reduced plant biomass
and seed production. Accumulation was mostly in the
roots. Nickel was more mobile than Cd, reaching higher
levels in all plant parts, especially seeds. Within the
tissues of mature seeds, the highest concentrations of Ni
were found in the axis and testa. The highest
concentrations of Cd were in the testa and cotyledon,
and the lowest in the axis. When expressed on a per
seed basis, metal contents of these organs increased
with developmental age. Nickel amounts were lower in
the pods than the seeds for all growth stages, however
there was no significant difference for Cd. Cadmium
reduced mature seed mass. This effect was mostly due
to decreased yields of lipids, protein and carbohydrates.
Although the number of seeds per pod declined as a
response to Ni, seed mass was unaffected and there
was no apparent effect on storage reserves.
Keywords: metal pollutants, cadmium, nickel, heavy
metal,
Glycine max,
soybean, seed

Introduction
There are several metal pollutants that are considered
to be of potential threat to environmental systems.
These include Cd, Cr, Cu, Hg, Ni, Zn and Pb
(Marschner, 1982; Friedland, 1990). Due to their
distinct chemistry and characteristics, each represents
a rather different hazard to the environment. In this
study, the effect of Cd and Ni on the development of
soybean seeds was examined.
Cd is a non-essential element in plants (Verkleij
and Schat, 1990). It is recognized as one of the most
potentially hazardous of all metal pollutants since it is
extremely toxic to humans and other animals (Rascio
et al., 1993; Cieslinski et al., 1996) and is known to
accumulate in mammalian kidneys (Quaife, 1981).
Exposure is due mainly to high amounts in the diet,
although tobacco smoking and occupational exposure
to CdO fumes are also important sources (Alloway,
1990). The fact that this metal is fairly readily taken up
by plants and translocated to aerial organs facilitates
its entry into the food chain (Rauser and Meuwly 1995;
Salim et al., 1995).
Nickel, on the other hand, although a serious
environmental pollutant (Sajwan et al., 1996) and
phytotoxic at high concentrations (L’Huillier et al.,
1996) is considerably less toxic to living organisms
than Cd. It has been found by some researchers to be
an essential micronutrient in certain plant species,
especially when grown on urea-based media (Breckle,
1991; Gerendas and Sattelmacher, 1997). In

comparison to Cd, Ni is even more mobile within
plants (Marschner, 1982).
Natural amounts of Cd in the environment are
generally low, however anthropogenic activities can
drastically increase these levels (Woolhouse, 1982).
Such activities include: zinc mining and smelting, use
of sewage sludge for agricultural fertilization,
motoring (car exhaust fumes), combustion of fossil
fuels, application of phosphate fertilizers, industrial
and manufacturing processes (Lund, 1981; Xian, 1989;
Rascio et al., 1993; Marchiol et al., 1996).
Nickel is generally more naturally abundant than
Cd. Some native soils, specifically mafic and
ultramafic (serpentine) soils, have high indigenous
amounts of this element (Mishra and Kar, 1974; Steyn
et al., 1996). Specially adapted species and populations
of plants have evolved to survive these conditions
(Peterson, 1983). Localized high contents do occur as a
result of mining, burning of fossil fuels, fertilizer
application, automobiles (McIlveen and Negusanti,
1994) and industrial activities such as the manufacture
of Ni-steel alloys (stainless steel), electronic
components and batteries (McGrath and Smith, 1990).
Since the late 1960s extensive research has been
carried out on the threat posed by metal pollutants to
the environment (Marschner, 1982; Tjell and
Christensen, 1992). However, very little research has
Seed Science Research (1998) 8, 445–453 445
*Correspondence
E-mail: ;

been carried out on the effect of metal pollutants on
seed development (Siegel and Siegel, 1985). A database
survey on papers concerned with uptake, accumulation
and translocation of heavy metals by vascular plants,
revealed that fewer than 11% of the almost 25 000 listed
papers studied the effect of metals on reproductive
parts (Nellessen and Fletcher, 1993). Research that has
been carried out has been limited almost exclusively to
analyses of total metal content within seeds and
whether this poses a potential risk to consumers. There
are few reports concerning the effect metal pollutants
may exert on metabolic and developmental processes
occurring within the seed. Soybeans, because of their
high nutritive value, are an important agricultural crop,
grown increasingly in developing countries as a food
source (Gupta, 1983; Odendaal et al., 1984). Thus it is
important to assess the impact of metal pollution on
seed production, and in this study we examined the
effect of Ni and Cd on soybean seed development.
Materials and methods
Seeds were harvested from soybean plants (Glycine
max (L.) Merr. cv. Crawford), grown in a modified
Hoagland’s nutrient solution amended with either Ni
or Cd. Details for the production of these seeds are
given below.
Plant cultivation and seed production
Plants were grown in a controlled environment
chamber at a 25
ЊC day and 20ЊC night temperature,
12-h photoperiod and PAR of 800 µmol m

Ϫ2
s
Ϫ1
. Seven
days after germination, seedlings were transferred to
one litre plastic jars filled with nutrient solution. The
concentration of macronutrients was as follows: 1 m
M
KH
2
PO
4
, 2 mM MgSO
4
.7H
2
O, 4 mM CaNO
3
, 4 mM
KNO
3
and micronutrients: 89.9 µM FeNaEDTA, 46 µM
H
3
BO
3
, 9.1 µM MnCl
2
.4H
2

O, 0.8 µM ZnSO
4
.7H
2
O,
0.3 µ
M CuSO
4
.5H
2
O, 0.1 µM H
2
MoO
4
.H
2
O. After two
weeks in the starting jars, seedlings were transferred
to a circulating nutrient system, which consisted of a
25-litre growth tank, in which the roots were
immersed. This was attached by tubing to a reserve
tank. Four plants were allocated to each growth tank
and the total volume of nutrient solution was 40 litres.
The pH of each circulating system was adjusted to 6.0
every 1–2 days and deionized water added to bring
the total volume back to 40 litres. Fresh nutrient
solution was made up every 10 days and the growth
tanks were constantly aerated. The same composition
of nutrient solution was used as in the starting
containers except that the chloride salts of either Cd or

Ni were added to give resulting metal pollutant
concentrations of 0 mg/litre (control) 0.05 mg/litre Cd
or 1 mg/litre Ni. The Cd-stress experiments were first
carried out utilizing one growth tank as the control
treatment and another three tanks for the metal
treatment. Subsequently the Ni-stress experiments
were conducted using the same tanks as control and
treatment tanks. Care was taken to ensure that the
environmental and growth parameters were constant.
In addition, the growth tanks were washed after the
Cd-stress experiments and the rinse water analysed
for this metal using the standard procedure (see
below). Cadmium contamination of the growth tanks
was minimal and is therefore not discussed further.
Seeds were harvested at four distinct stages in
development which were determined by the size and
morphology of the pod and seed. Pods were measured
with regard to length, depth and thickness, as well as
the extent to which the depth of the locule was filled
by the developing ovule. This was based on the
method of Miles et al. (1988). Approximate DAF (days
after flowering) for each stage are also given.
Immature pods (IP) – pods dark green in colour, at
least 50 mm in length and 10 mm in depth. Ovules
4–6 mm in depth, i.e. filling half the depth of the
locule. Seeds in rapid growth stage. DAF approxi-
mately ϭ 16–17.
Expanded pods (EP) – pods light green in colour,
fully expanded (
> 7 mm in thickness) and turgid.

Ovules filling the entire locule depth, green and
showing no yellowing. Seeds at mid-seed fill
period. DAF approximately ϭ 30.
Yellow pods (YP) – pods light yellow in colour
and pliable. Seeds detached from the funiculus,
soft and bean shaped. Seeds physiologically
mature. Maturation drying in progress. DAF
approximately ϭ 51.
Brown pods (BP) – pods light brown in colour
and brittle. Seeds mature, dry, hard and round in
shape. DAF approximately ϭ 54.
Seeds were harvested at each growth stage, freeze-
dried for 48 h and stored at Ϫ80ЊC until further
processing.
Uptake of metal pollutants
Plants were grown in nutrient solution amended with
either 0.05 mg/litre Cd or 1 mg/litre Ni and the visual
toxicity symptoms noted. At senescence (when all
remaining pods were at the BP stage), plants were
separated into roots, leaves and pods. These were
washed under running deionized water for 20 s and
oven-dried at 70ЊC for 48 h. Material was finely
ground, and 0.5-g samples ashed in a muffle furnace
for 5 h at 500ЊC. Freeze-dried seed samples were
processed in a similar manner, except that the oven
drying step was omitted and 2-g samples were used.
All samples were then digested for 24 h in
concentrated HNO
3
at a temperature of 150ЊC and

made up to a final volume of 25 ml in 0.1
M HNO
3
.
446 H. L. Malan and J. M. Farrant
Samples were analysed for Cd or Ni using a Jobin
Yvon JY138 ultratrace ICP-AES (inductively coupled
plasma - atomic emission spectrophotometer).
The effect of metal pollutants on seed development
Seeds, harvested from metal-treated plants at various
stages of development, were compared with those
from control plants. The following parameters were
examined: total seed yield, mass, moisture content,
germination and storage reserve accumulation. Lipid
determination was carried out on freshly harvested
seeds according to a modified method of Christie
(1973). Total extractable carbohydrates were
determined on freeze-dried tissue according to the
method of Adams et al. (1980). Total N (nitrogen) was
assayed using the standard micro-Kjeldahl method
(Stock and Lewis, 1986) and the crude protein content
estimated by multiplying the nitrogen content by a
factor of 5.49 which is appropriate for soybean seeds
(Mossé and Pernollet, 1983).
Statistical treatment of the data
Significant differences between means were examined
using Student’s t test at the 95% confidence limit. In
cases where sample size was small, Wilcoxon’s rank
sum test was employed.
Results

Visual toxicity symptoms
Visual toxicity symptoms exhibited by the leaves on
exposure to the two metals were similar to those
previously described for white beans by Rauser (1978).
The pods and seeds produced by metal-treated plants
were for the most part indistinguishable in appearance
from those of the controls. However, plants treated
with Ni occasionally produced deformed terminal
racemes, composed of pods greatly reduced in size
(approximately 10 mm compared to 50 mm). These
abnormal pods remained green and contained either
no seeds or those that were rudimentary and non-
viable.
Effect on plant biomass
Table 1 shows the effects of Cd and Ni on plant growth
parameters as represented by pod yields as well as by
the root dry mass per plant. Both metals markedly
decreased pod production and root biomass relative to
the controls. Cadmium appears to be more toxic than
Ni since a lower concentration of the former was
required to elicit the same degree of pod and root
biomass depression.
Distribution of metal within the plant
Table 2 shows the Cd and Ni content of roots, leaves,
pods and mature (BP) seeds taken from metal-treated
and control plants. Metal content in all parts was
higher than in the equivalent organ of controls.
Considerable Cd enrichment occurred in the roots,
accumulating to a concentration of 130 µg/g dm from
the 0.05 mg/litre present in the nutrient solution.

Cadmium values for the aerial portions of the plant
were low in comparison to the roots, the concentration
in the leaves being 30-fold lower than the roots.
Cadmium contents were lowest in the reproductive
tissues. Nickel was also concentrated in the roots with
lower levels in the shoots and seeds. Nickel values for
the leaves were 20-fold less than for the roots. In
general the amounts in all parts were much higher
than for Cd. Leaves and seeds accumulated similar Ni
concentrations but pod contents were considerably
lower.
The distribution of Cd and Ni within mature
soybean seeds is shown in Table 3. Cadmium
enrichment occurred in the testa and cotyledons with
very little accumulating in the axis. Nickel concen-
trations on the other hand were highest in the axis,
intermediate in the testa and lowest in the cotyledons.
The effect of seed growth stage on metal
accumulation
The concentration of the two metals in seeds and pods
at each developmental growth stage was examined
(Table 4). Concentrations were always significantly
higher in treated, relative to control, seeds and pods.
For all developmental stages Ni contents were higher
in the seed than in the pod. However, there appeared
to be little difference between pods and seeds of
Cd-treated plants whatever the stage of development.
Metal concentrations (calculated per gram dry mass)
were higher in young (IP) seeds but declined
significantly by the EP stage. However if the results

are expressed on a per seed basis, metal content of
treated seeds generally positively correlated with seed
age.
The effect of metal pollutants on seed growth
parameters
Table 5 summarizes the effect of Cd and Ni on seed
mass and the average number of seeds per pod for
mature seeds. The seed number did not change during
development and so results for earlier developmental
stages are not given. Cadmium had a significant effect
on the average size of BP seeds, resulting in decreased
seed mass relative to the controls (P ≥ 0.001). However
this metal did not affect the average number of seeds
per pod. On the other hand, although Ni did not exert
Metal pollutants and soybean seed development 447
448 H. L. Malan and J. M. Farrant
Table 1. Effect of Cd and Ni on seed yield and root dry mass for plants grown in 0.05
mg/litre Cd or 1 mg/litre Ni. SD given in parenthesis. n ϭ 3 for metal treatments, n ϭ 4 for
control
Growth parameter Cd-treated plants Ni-treated plants Control
No. pods/plant 67.5 (Ϯ 13.3) 68.7 (Ϯ 9.1) 93.5 (Ϯ 7.20)
Root dry mass (g) 6.8 (Ϯ 3.3) 6.9 (Ϯ 0.12) 10.6 (Ϯ 2.1)
Table 2. Distribution of Cd and Ni in various parts of plants grown in 0.05 mg/litre Cd or 1
mg/litre Ni. SD given in parenthesis. Minimum sample size ϭ 3
Cadmium (␮g/g dm) Nickel (␮g/g dm)
Plant part
Treatment Control Treatment Control
Roots 130.09 (Ϯ 34.2) 1.31 (Ϯ 0.19) 1100 (Ϯ 40.0) 3.1 (Ϯ 0.1)
Leaves 3.80 (Ϯ 0.08) 0.43 (Ϯ 0.17) 48.1 (Ϯ 1.5) ND
Pods 0.78 (Ϯ 0.24) 0.48 (Ϯ 0.01) 12.5 (Ϯ 0.69) ND

Mature seeds 0.96 (Ϯ 0.15) 0.12 (Ϯ 0.04) 49.1 (Ϯ 5.75) 0.2 (Ϯ 0.01)
ND ϭ Not detectable
Table 3. Distribution of Cd and Ni within the tissues of mature (BP) soybean seeds
harvested from plants grown in 0.05 mg/litre Cd or 1 mg/litre Ni. Because of the low
amounts of metal present in the seed, the dry mass of tissue required per sample was high
and thus sample size was small (n ϭ 2 for axes, n ϭ 3 for other tissues). The approximate
number of seeds required per sample is given for each treatment. The same mass for
treatment and the equivalent control was used. SD given in parenthesis
Cadmium (␮g/g dm) Nickel (␮g/g dm)
Seed tissue
Treatment Control seeds/ Treatment Control seeds/
sample sample
Testa 1.52 (Ϯ 0.51) 0.04 (Ϯ 0.01) 40 77 (Ϯ 3.0) ND 20
Cotyledon 1.53 (Ϯ 0.19) 0.05 (Ϯ 0.01) 15 55.7 (Ϯ 1.9) ND 15
Axis 0.04 (Ϯ 0.06) 0.01 (Ϯ 0.00) 80 99.2 (Ϯ 3.4) 0.98 (Ϯ 1.2) 50
ND ϭ Not detectable
Table 4. Effect of seed development stage on metal concentration in seeds and pods
harvested from plants grown in 0.05 mg/litre Cd or 1 mg/litre Ni. Seed concentrations
given both on a ␮g/g dm and per seed basis. IP ϭ immature pod, EP ϭ expanded pod,
YP ϭ yellow pod, BP ϭ brown pod. Complete descriptions of developmental stages given
under Materials and methods. Ni control values omitted for clarity as all were below
detection limit. SD given in parenthesis. Minimum sample size = 3
Seed
Cadmium Nickel
development Control Treatment Treatment
stage (␮g/g dm) (␮g/g dm) (␮g/g dm)
Pod Seed Pod Seed
␮g Cd
Pod Seed
␮g Ni

/seed /seed
IP 0.46 0.37 0.95 1.25 0.015 31.03 71.2 0.926
(Ϯ 0.01) (Ϯ 0.03) (Ϯ 0.02) (Ϯ 0.12) (Ϯ 0.00) (Ϯ 5.8) (Ϯ 0.5) (Ϯ 0.01)
EP 0.47 0.34 1.41 0.84 0.122 17.3 49.8 7.07
(Ϯ 0.01) (Ϯ 0.06) (Ϯ 0.29) (Ϯ 0.09) (Ϯ 0.01) (Ϯ 1.2) (Ϯ 5.2) (Ϯ 0.74)
YP 0.52 0.33 0.96 0.91 0.208 8.49 40.8 7.83
(Ϯ 0.02) (Ϯ 0.04) (Ϯ 0.15) (Ϯ 0.07) (Ϯ 0.02) (Ϯ 1.8) (Ϯ 2.3) (Ϯ 0.44)
BP 0.48 0.15 0.78 0.87 0.167 12.5 49.1 9.72
(Ϯ 0.01) (Ϯ 0.01) (Ϯ 0.24) (Ϯ 0.15) (Ϯ 0.03) (Ϯ 0.69) (Ϯ 5.8) (Ϯ 1.14)
a significant effect on seed mass, the presence of this
element in the growth medium did decrease the mean
number of seeds per pod (p
≥ 0.001).
The effect of metal pollutants on storage reserves
Figures 1, 2 and 3 show the effect of Cd and Ni on
storage reserves of treated seeds. Protein contents, as
determined from total nitrogen, increased with seed
development in all treatments (Fig. 1). Cadmium
significantly reduced the total protein content of
mature seeds, a reflection of reduced seed mass.
Nickel on the other hand did not result in any
significant changes in protein content.
Lipid content increased with seed age until the YP
stage, thereafter it levelled off (Fig. 2). In the case of
the Cd treated seeds and control seeds, lipid levels
were slightly lower than at the YP growth stage due to
the fact that the BP seeds were slightly smaller in size
(data not shown). Cadmium significantly reduced the
lipid content of the mature seeds relative to the
controls, again a reflection of reduced seed size. A

similar effect was evident in Ni-treated mature seeds.
However the effect of this metal was not statistically
significant.
Soluble carbohydrate (sugars) and insoluble
carbohydrate (starch) levels are given in Figure 3.
Soluble carbohydrates increased with seed
development, reaching a peak at maturity. Starch, on
the other hand, increased during the early growth
stages (IP and EP) and then declined in mature seeds.
Cadmium decreased carbohydrate levels in treated
compared to control seeds, although only the effect on
starch was significant. Nickel decreased the levels of
soluble sugars compared to the controls.
Discussion
The major effect exerted by Cd and Ni in this study
was a general reduction in plant biomass. This was
observed in the form of decreased root mass and
decline in pod yield and was the response to both
Metal pollutants and soybean seed development 449
Table 5. Effect of Cd and Ni on dry mass and seeds per pod for mature (BP) seeds harvested
from plants grown in 0.05 mg/litre Cd or 1 mg/litre Ni. SD and sample size given in
parenthesis
Growth
Cd-treated plants Ni-treated plants
parameter
Treatment Control Treatment Control
No. seeds/pod 1.94 2.12 1.86* 2.27
(Ϯ 0.4, n ϭ 14) (Ϯ 0.19, n ϭ 11) (Ϯ 0.4, n ϭ 24) (Ϯ 0.2, n ϭ 20)
Seed mass (g/seed) 0.192* 0.229 0.198 0.193
(Ϯ 0.04, n ϭ 84) (Ϯ 0.02, n ϭ 53) (Ϯ 0.04, n ϭ 37) (Ϯ 0.05, n ϭ 31)

* Indicates significance at P у 0.001
Figure 1. Effect of Cd and Ni on protein content (mg/seed)
of seeds harvested from plants grown in 0.05 mg/litre Cd or
1 mg/litre Ni. IP and YP values for Ni and Ni control
treatments not determined. (
■ Cd treatment, ❒ Cd control,
vertical stripes ϭ Ni treatment, horizontal stripes ϭ Ni
control). n ϭ 3.
Figure 2. Effect of Cd and Ni on lipid content (mg/seed) of
seeds harvested from plants grown in 0.05 mg/litre Cd or 1
mg/litre Ni. IP and YP values for Ni and Ni control
treatments not determined. (■ Cd treatment, ❒ Cd control,
vertical stripes ϭ Ni treatment, horizontal stripes ϭ Ni
control). n ϭ 3.
metal pollutants. Reduction in plant biomass as a
result of heavy metal stress appears to be an almost
universal finding (MacNicol and Beckett, 1985; Leita et
al., 1993; Dudka et al., 1996; Ouzounidou et al., 1997). A
few cases of yield enhancement due to metal
pollutants have been reported in the literature, but
these were from experiments utilizing extremely low
concentrations of metals (Mishra and Kar, 1974;
Breckle, 1991). Root growth appears to be especially
susceptible to metal toxins compared to the shoots and
has been used extensively as a convenient criterion of
metal tolerance (Ouzounidou et al., 1997). A reduction
in the yield of reproductive tissues has also been
reported for several species (Huang et al., 1974;
Cimino and Toscano, 1993; Singal et al., 1995).
Nickel concentrations in seeds were consistently

higher than those in pods for all growth stages, whilst
there was little difference in the Cd content of the two.
This suggests that the pods pose only a minimal
barrier, and exert little screening effect on metal
pollutants. Other reports in the literature however do
not support these findings. Cimino and Toscano (1993)
examined uptake of Cd, Pb and Cu from sludge- or
metal-amended soils into pea and bean seeds.
Cadmium contents of the pods were significantly
higher than of the seeds for both species. Haghiri
(1973), experimenting with radioactive Cd in soybean
plants, also found that Cd was higher in pods than
seeds. It is possible that the pod to seed ratio may be
dependent on the concentration of Cd supplied to the
plant.
The low Cd content of seeds found in this study is
similar to other values found in the literature, and is
consistent with the general view that plant
reproductive organs tend to be protected from toxic
metals (Marschner, 1982). On the other hand, high
seed concentrations of Ni have also been reported by
other authors (Halstead et al., 1969; Cataldo et al., 1978)
and support the contention that Ni appears to be an
exception to this rule of minimal seed accumulation
(Welch, 1995; Sajwan et al., 1996). Thus Ni appears to
be more mobile within plants than Cd, as shown by
the elevated Ni concentrations of this element in all
plant parts. Whilst the concentration of Ni used in the
nutrient solution was twenty times higher than that of
Cd, calculation of the concentration factor (i.e. the

ratio of the concentration of metal accumulated to that
available for uptake) for seeds in this study, yields
values of 20 for Cd and 50 for Ni. Thus the magnitude
of accumulation in soybean seeds was greater for Ni
than for Cd, and was not simply a result of a higher
supplied concentration.
Many variables such as soil composition,
temperature, pH, chemical form and concentration
have been shown to affect plant uptake of metal
pollutants in the field (Ernst, 1996). In this study,
plants were grown in nutrient solution and the root
environment strictly controlled. Extrapolation of
results from plants grown in such an artificial system
to those in the field can be difficult. In the soil, due to
binding of metal cations by soil components not all the
metal is available for plant uptake (Chaney, 1991). In
nutrient solution systems, on the other hand, the
proportion of bioavailable metal ions is often higher
because of the absence of this binding and thus plant
uptake is often greater from nutrient solution than
from a soil containing the equivalent concentration of
a given metal ion. Reports of Cd values slightly higher
than 1 ␮g/g dm have been reported in the literature
for seeds harvested from plants grown in polluted
areas (Yoshida, 1986; Stefanov et al., 1995). Therefore it
is felt that the levels of metal pollutants used in this
study and the effects exerted by them are comparable
to those that may be found at contaminated sites in the
field. It is of interest that the limit for Cd in legume
crops as recommended by the World Health

Organization in 1992 (Petterson and Harris, 1995) is
0.1 ␮g/g dm.
450 H. L. Malan and J. M. Farrant
Figure 3. Effect of Cd and Ni on the sugar and starch content
(mg glucose/seed) of seeds harvested from plants grown in
0.05 mg/litre Cd or 1 mg/litre Ni. IP and YP values for Ni
and Ni control treatments not determined. (■ Cd treatment,
❒ Cd control, vertical stripes ϭ Ni treatment, horizontal
stripes ϭ Ni control). Minimum n ϭ 2.
Both metal treatments resulted in declined pod
numbers, which in turn affected total seed yield. Thus
the primary effect of these metals was on early events
such as flower production or fruit set. Once committed
to pod formation however, the pollutants had
differing effects on seed development. Nickel
treatment resulted in reduced numbers of seeds per
pod, but seed mass was equivalent to control seeds.
Cadmium treatment resulted in the same number of
seeds per pod as the control, but individual seeds
were smaller. This could be explained in terms of
photosynthate available from the parent for reserve
accumulation. Because Ni treatment reduced the
number of seeds during early development, there was
more photosynthate available per seed for reserve
accumulation. With greater numbers of seeds reaching
the stage of nutrient deposition, Cd treatment resulted
in reduced storage reserve accumulation and this
affected seed mass. Although the total concentration
of Cd in the seeds was low, 83% was located in the
cotyledon, the principal site of storage reserve

deposition. On the other hand, only 43% of the Ni
taken up into the seeds was located in the cotyledons.
This may be the reason that reserves are lowered in
seeds exposed to Cd, but reserve accumulation and
hence seed size were not affected by Ni. Cieslinski et
al. (1996) concluded that yield reduction in strawberry
fruit when grown in Cd-amended soil resulted mainly
from decreases in fruit number rather than average
weight per berry. Moraghan (1993) when investigating
the effect of the same metal pollutant found similar
effects on yield parameters of flaxseed. On the other
hand, Singal et al. (1995), examined the effect of Cd on
seed mass of fenugreek, and found that at all
developmental stages there was a general decrease in
seed size with increasing Cd concentration.
The differing metal distribution patterns found in
the seeds is interesting. When expressed as percentage
of total metal uptake, 42% of the total Ni was localized
in the maternal tissue of the testa and is unlikely to
affect subsequent germination. Fifteen percent was
found in the axis. However, despite this relatively high
value, it was found (data not reported) that seed
germination was not profoundly affected, vigour was
slightly decreased relative to the controls but viability
was the same. In the case of Cd, although the total
concentration of the metal in the seed was low, 83%
was located in the cotyledon and comparatively little
in the testa (17%) or axis (0.1%). Even though the
amount in the axis was extremely low, vigour was also
slightly decreased in these seeds, but there was no

effect on viability. Thus, the results indicate that Cd is
more toxic than Ni and exerts a more pronounced
effect on seed development.
It is possible that the quality of the storage reserves
within the seed are altered as a response to the
presence of either metal pollutant, since only the
quantities of storage reserves were investigated in the
present study. Stefanov et al. (1995) found that lead
altered the lipid content in seeds of green pepper,
shifting the balance between saturated and
unsaturated fatty acids in a complex manner.
Cadmium was found to generally increase lipid
phosphorus (P) and decrease protein P in the seeds of
fenugreek (Singal et al., 1995). Further studies on the
effect of toxic metals on the chemical composition of
soybean seeds may be rewarding, especially if the
nutritional value is affected. Although it is well
documented that soybean seeds contain very little
starch at maturity (Adams et al., 1980) results from
these experiments consistently showed starch contents
up to 20 mg/seed in the oldest growth stage. This may
be due to the cultivar, but is most likely due to
inefficient separation of soluble sugars from starch
during the extraction process.
In conclusion, it can be seen that the presence of
metal pollutants in the nutrient solution greatly
affected the parent plant. At levels of Cd or Ni where
adult plants could survive, seed production was
greatly diminished. Significant amounts of metal did
enter the seeds, especially in the case of Ni due to its

enhanced mobility. This did not markedly reduce the
quality of the seed with respect to the quantity of
nutrients in the case of Ni, however yields of storage
reserves from Cd-treated plants were reduced.
Cadmium appears to have a more profound effect on
seed development than Ni.
Acknowledgements
This work was supported by the FRD (Foundation for
Research Development), South Africa.
References
Adams, C.A., Rinne, R.W. and Fjerstad, M.C. (1980) Starch
deposition and carbohydrase activities in developing
and germinating soya bean seeds. Annals of Botany 45,
577–582.
Alloway, B.J. (1990) Cadmium. pp 100–124 in Alloway, B.J.
(Ed.) Heavy metals in soils. Glasgow, Blackie.
Breckle, S.W. (1991) Growth under stress: Heavy metals. pp
351–373, in Waisel, Y.; Eshel, A.; Kafkafi, U. (Eds) Plant
roots: The hidden half. New York, Marcel Dekker Inc.
Cataldo, D.A., Garland, T.R., Wildung, R.E. and Drucker,
H. (1978) Nickel in plants II: Distribution and chemical
form in soybean plants. Plant Physiology 62, 563–565.
Chaney, R.L. (1991) Metal speciation and interaction among
elements affect trace element transfer in agricultural
and environmental food-chains. pp 219–260 in Kramer,
J.R.; Allen, H.E. (Eds) Metal speciation: Theory, analysis
and application. Chelsea, Michigan, Lewis Publishers
Inc.
Metal pollutants and soybean seed development 451
Christie, W.W. (1973) Lipid analysis: Isolation, separation,

identification and structural analysis of lipids. Oxford,
Pergamon Press.
Cieslinski, G., Van Rees, K.C.J., Huang, P.M., Kozak, L.M.,
Rostad, H.P.W. and Knott, D.R. (1996) Cadmium
uptake and bioaccumulation in selected cultivars of
durum wheat and flax as affected by soil type. Plant
Physiology 182, 115–124.
Cimino, G. and Toscano, G. (1993) Effects of digested
sewage-sludge on yield and heavy metal accumulation
in horticultural species. Bioresource Technology 46,
217–220.
Dudka, S., Piotrowska, M. and Terelak, H. (1996) Transfer
of cadmium, lead and zinc from industrially
contaminated soil to crop plants: A field study.
Environmental Pollution 94, 181–188.
Ernst, W.H.O. (1996) Bioavailability of heavy metals and
decontamination of soils by plants. Applied Geochemistry
11, 163–167.
Friedland, A.J. (1990) The movement of metals through soils
and ecosystems. pp 7–20 in Shaw A.J. (Ed.) Heavy metal
tolerance in plants: Evolutionary aspects. Boca Raton, CRC
Press.
Gerendas, J. and Sattelmacher, B. (1997) Significance of Ni
supply for growth, urease activity and the
concentrations of urea, amino acids and mineral acids
and mineral nutrients of urea-grown plants. Plant and
Soil 190, 153–162.
Gupta, Y.P. (1983) Nutritive value of food legumes. pp
287–328 in Arora, S.K. (Ed.) Chemistry and biochemistry of
legumes. London, Edward Arnold.

Haghiri, F. (1973) Cadmium uptake by plants. Journal of
Environmental Quality 2, 93–96.
Halstead, R.L., Finn, B.J. and MacLean, A.J. (1969)
Extractability of nickel added to soils and its concentra-
tion in plants. Canadian Journal of Soil Science 49, 335–342.
Huang, C.Y., Bazzaz, F.A. and Vanderhoef, L.N. (1974) The
inhibition of soybean metabolism by cadmium and lead.
Plant Physiology 54, 122–124.
Leita, L., De Nobili, M., Mondini, C. and Baca Garcia, M.T.
(1993) Response of leguminosae to cadmium exposure.
Journal of Plant Nutrition 16, 2001–2012.
L’Huillier, L., d’Auzac, J., Durand, M. and Michaud-
Ferriere, N. (1996) Nickel effects on two maize (Zea
mays) cultivars: growth, structure, Ni concentration and
localization. Canadian Journal of Botany 74, 1547–1554.
Lund, R.E. (1981) Cadmium metallurgy. pp 122–123 in
McGraw-Hill encyclopedia of science and technology Vol. 3,
New York, Maidenhead, McGraw-Hill.
MacNicol, R.D. and Beckett, P.H.D. (1985) Critical tissue
concentrations of potentially toxic elements. Plant and
Soil 85, 107–129.
Marchiol, L., Leita, L., Martin, M., Parasite, A. and Exurbia,
G. (1996) Physiological responses of two soybean
cultivars to cadmium. Journal of Environmental Quality
25, 562–566.
Marschner, H. (1982) General introduction to the mineral
nutrition of plants. pp 37–60 in Lauchli, A.; Bieleski, R.L.
(Eds) Encyclopedia of plant physiology. New Series Vol.
15A. Berlin, Springer-Verlag.
McGrath, S.P. and Smith, S. (1990) Chromium and nickel.

pp 125–150 in Alloway, B.J. (Ed.) Heavy metals in soils.
Glasgow, Blackie.
McIlveen, W.D. and Negusanti, J.J. (1994) Nickel in the
terrestrial environment. Science of the Total Environment
148, 109–138.
Miles, D.F., Tekrony, D.M. and Egli, D.B. (1988) Changes in
viability, germination and respiration of freshly
harvested soybean seed during development. Crop
Science 28, 700–704.
Mishra, D. and Kar, M. (1974) Nickel in plant growth and
metabolism. Botanical Review 40, 395–452.
Moraghan, J.T. (1993) Accumulation of cadmium and
selected elements in flax seed grown on a calcareous
soil. Plant and Soil 150, 61–68.
Mossé, J. and Pernollet, J.C. (1983) Storage proteins of
legume seeds. pp 111–194 in Arora, S.K. (Ed.) Chemistry
and biochemistry of legumes. London, Edward Arnold.
Nellessen, J.E. and Fletcher, J.S. (1993) Assessment of
published literature on the uptake, accumulation and
translocation of heavy metals by vascular plants.
Chemosphere 27, 1669–1680.
Odendaal, W.A., Smith, G.A. and Smith, N. (1984) Die
sojaboon in die Republiek van Suid-Afrika: Bron van
protein in menslike and dierlike voeding. Tegniese
Mededeling No. 196. Dept. van Landbou en Water
Voorsiening.
Ouzounidou, G., Moustakas, M. and Eleftheriou, E.P.
(1997) Physiological and ultrastructural effects of cad-
mium on wheat (Triticum aestivum L.) leaves. Archives of
Environmental Contamination and Toxicology 32, 154–160.

Peterson, P.J. (1983) Adaptation to toxic metals. pp 51–69 in
Robb, D.A.; Pierpoint, W.S. (Eds) Metals and
micronutrients in plants. London, Academic Press Inc.
Petterson, D.S. and Harris, D.J. (1995) Cadmium and lead
content of lupin seed grown in Western-Australia.
Australian Journal of Experimental Agriculture 35, 403–
407.
Quaife, C. (1981) Cadmium disease. p 122 in McGraw-Hill
encyclopedia of science and technology Vol. 3, New York,
Maidenhead, McGraw-Hill Publishers.
Rascio, N., Dallavecchia, F., Ferretti, M., Merlo, L. and
Ghisi, R. (1993) Some effects of cadmium on maize
plants. Archives of Environmental Contamination and
Toxicology 25, 244–249.
Rauser, W.E. (1978) Early effects of phytotoxic burdens of
cadmium, cobalt, nickel and zinc in white beans.
Canadian Journal of Botany 56, 1744–1749.
Rauser, W.E. and Meuwly, P. (1995) Retention of cadmium
in roots of maize seedlings – role of complexation by
phytochelatins and related thiol peptides. Plant
Physiology 109, 195–202.
Sajwan, K.S., Ornes, W.H., Youngblood, T.V. and Alva,
A.K. (1996) Uptake of soil applied cadmium, nickel and
selenium by bush beans. Water, Air and Soil Pollution 91,
209–217.
Salim, R., Isa, M., Al-Subu, M.M., Sayrafi, S.A. and
Sayrafi, O. (1995) Effect of irrigation with lead and
cadmium on the growth and on the metal uptake of
cauliflower, spinach and parsley. Journal of Environmental
Science and Health Part A: Environmental Science and

Engineering & Toxic and Hazardous Substance Control 30,
831–849.
Siegel, S.M. and Siegel, B.Z. (1985) Differential elimination
of mercury during maturation of leguminous seeds.
Phytochemistry 24, 235–236.
452 H. L. Malan and J. M. Farrant
Singal, N., Gupta, K., Joshi, U.N. and Arora, S.K. (1995)
Phosphorus content and growth of fenugreek as affected
by cadmium application. Biologia Plantarum 37, 309–
313.
Stefanov, K., Seizova, K., Yanishlieva, N., Marinova, E. and
Popov, S. (1995) Accumulation of lead, zinc and
cadmium in plant seeds growing in metalliferous
habitats in Bulgaria. Food Chemistry 54, 311–313.
Steyn, C.E., van der Watt, H.v.H. and Claassens, A.S. (1996)
On the permissible nickel concentration for South
African soils. South African Journal of Science 92, 359–363.
Stock, W.D. and Lewis, O.A.M. (1986) Soil nitrogen and the
role of fire as a mineralising agent in a South African
coastal fynbos ecosystem. Journal of Ecology 74, 317–328.
Tjell, J.C. and Christensen, T.H. (1992) Sustainable
management of cadmium in Danish agriculture. pp
273–286 in Vernet, J P. (Ed.) Trace metals in the
environment. Vol. 2: Impact of heavy metals on the
environment. Amsterdam, London, Elsevier Press.
Verkleij, J.A.C. and Schat, H. (1990) Mechanisms of metal
tolerance in higher plants. pp 179–194 in Shaw, A.J. (Ed.)
Heavy metal tolerance in plants: Evolutionary aspects. Boca
Raton, CRC Press.
Welch, R.M. (1995) Micronutrient nutrition of plants. Critical

Reviews in Plant Sciences 14, 49–82.
Woolhouse, H.W. (1982) Toxicity and tolerance in the
responses of plants to metals. pp 246–300 in Lange, O.L.;
Nobel, P.S.; Osmond, C.B.; Ziegler, H. (Eds) Encyclopedia
of Plant Physiology. New Series Vol 12C. Berlin, Springer-
Verlag.
Xian, X. (1989) Response of kidney bean to concentration
and chemical form of cadmium, zinc and lead in
polluted soils. Environmental Pollution 57, 127–137.
Yoshida, S. (1986) Isolation and partial characterization of a
cadmium-binding protein in soybeans from a cadmium-
polluted field. Agricultural and Biological Chemistry 50,
2273–2278.
Received 22 January 1998,
accepted after revision 10 July 1998
© CAB INTERNATIONAL, 1998
Metal pollutants and soybean seed development 453