Geochemical Journal, Vol. 40, pp. 57 to 67, 2006
Role of fine roots in the plant-induced weathering of andesite for
several plant species
AKTER MEHERUNA and T ASUKU AKAGI*
United Graduate School of Agricultural Science, Tokyo University of Agriculture and Technology,
3-5-8 Saiwai-cho, Fuchu, Tokyo 183-8509, Japan
(Received November 16, 2004; Accepted July 12, 2005)
The present work aims at providing experimental evidence for weathering as a direct consequence of plant physiology,
and the importance of the proximity of fine roots to rock in the weathering process. Discussion is based on the release of
different elements from andesite rock particles by the three crop species: rice, maize, and soybean. We designed two types
of hydroponic crop pots, in which fine roots were allowed (or not allowed) to contact rock particles by using coarse (or
fine) net bags. A plant-free control was also run. Experiments were carried out in a controlled glasshouse for 35–38 days.
The pH in the media of all the planted pots decreased by about 1–1.5 units from the value of blank pots during the
experimental period, but it did not differ significantly between the coarse and fine net pots. The release of elements in the
presence of the plants was calculated by subtracting the depletion of media from the amount absorbed by the plants. We
observed a positive effect of plants on the release of elements from the rock particles. The amounts of Ca, Mg and Mn
released increased by a factor of 4–12, 4–28, and 4–7, respectively, except for Ca for rice. The amount of Si released was
also higher in all the planted pots. Between the coarse and fine net pots the released amounts of Ca and Mn differed
significantly for maize (p < 0.05) and that of Si for soybean (p < 0.05). In the case of Fe and Al, the fine net pots gave a
greater release than did the coarse ones. A significant difference was found only in the rice pots (p < 0.05 for Fe and p <
0.001 for Al). Maize showed the greatest growth of the three species and significantly higher release of some elements in
the coarse net pot than in the fine net pot. This implies that weathering may be caused partially by the direct contact of fine
roots with rock particles, together with the alteration of rhizospheric conditions by the roots.
Keywords: andesite, direct contact, fine roots, plants, weathering
centration of the surrounding liquid phase, which is directly responsible for substantial changes in the
rhizospheric pH (Römheld, 1991; Marschner, 1995;
Hinsinger, 1998). Higher plants also have an effect on
the redox conditions in the rhizosphere, which determine
the dynamics of Fe- and Mn-bearing minerals (Uren,
1981; Hinsinger, 1998). These rhizospheric processes
provide H+ for the chemical attack of silicate minerals,
and some acids such as oxalic acid can also chelate and
make soluble otherwise insoluble elements such as iron
and aluminum (Boyle and Voight, 1973; Graustein, 1981).
Some characteristic silicate weathering reactions, using
the anorthite component of plagioclase as an example,
are (Holland, 1978; Drever, 1994; Hinsinger et al., 2001):
INTRODUCTION
Nutrients required for plant growth, other than nitrogen and sometimes sulfur, are initially supplied by the
chemical dissolution of primary minerals in the process
known as weathering. Vascular plants should accelerate
weathering more than the activity of any likely pre-existing primitive terrestrial organisms, such as algae and lichens, because of the much greater contact area between
minerals and the huge mass of fine roots of the higher
plants and because of plants’ much faster growth and internal storage of rock-weathering-derived cations (Berner,
1992, 1995).
An important role of plants in rock weathering is their
production of organic and inorganic acids through root
exudates and root respiration, respectively, which enhance
the weathering rate. The very high growth rates of higher
plants are associated with a large uptake of water and
nutrients. This can modify considerably the ionic con-
2CO2 + 3H2O + CaAl 2Si2O8 = Al2Si2O5(OH)4
+ Ca2+ + 2HCO3–
(i)
H2C2O4 + H2O + CaAl2Si2O8 = Al2Si2O5(OH)4 + CaC2O4.
(ii)
*Corresponding author (e-mail: [email protected])
The possibility of enhanced weathering of phosphates
(Grinsted et al., 1982; Hinsinger and Gilkes, 1997), car-
Copyright © 2006 by The Geochemical Society of Japan.
57
bonates (Jaillard, 1987) and silicates (Robert and
Berthelin, 1986; Drever, 1994; Hinsinger et al., 2001) in
the rhizospheres of higher plants is now well established;
however, the rate of mineral weathering, and any change
to that rate, are extremely important factors influencing
the processes that maintain soil fertility and long-term
productivity.
Darrah (1993) reviewed the role of the rhizosphere in
plant nutrition and was particularly concerned with quantifying the root-mediated changes to the chemical, physical, and biological properties and processes of soil. He
concluded that, although many changes have been demonstrated experimentally, the quantitative significance of
the change is very difficult to define because many of the
mechanisms operate in parallel. From an ecological or
agronomic perspective, it is necessary to disentangle these
processes and quantify their individual contributions to
the total uptake as a prelude to understanding the competition or complementarities, and prior to making any reinforcement or selection of such attributes in programs
to fit crops to their environment. The nutrient requirements of plant species and their ability to alter chemical
conditions in the rhizosphere differ widely (Marschner,
1995). The differences are both qualitative, as implied
mechanisms, and quantitative, as resulting fluxes.
Plant nutrition is often investigated by nutrient balance on a field scale. Weathering, especially biological,
has been considered negligible with regard to plant nutrition, because the direct effects of higher plants on the
weathering of minerals and rocks have rarely been clearly
demonstrated in situ (Gobran et al., 1999) and the extent
to which such processes contribute to plant nutrition remains still a question for debate (Drever, 1994; Jackson,
1996). Field experiments do not generally make it possible to quantify the sources of the nutrients taken up by
the crops. Bedrock silicate rocks have been investigated
for their potential to provide plant nutrients, mainly in
studies of degraded soil originating within a temperate
climate environment (Coroneos et al., 1996). However,
very little is known about the possible use of silicate rocks
in tropical agriculture as alternative slow-release fertilizers (Barak et al., 1983) or amendments (Gillman, 1980).
Harley and Gilkes (2000) reviewed the various factors
that influence the release of plant nutrients from silicate
rock. To assess its potential use as a fertilizer, more investigations are needed on how the plants can affect the
kinetics of dissolution of silicate rocks and minerals.
In recognition of potential difficulties that might arise
during experimentation under actual field conditions, we
carried out our experiments hydroponically in a glasshouse where plants were allowed to be in direct contact
with rock particles via their fine roots in one treatment.
Conversely, they were not in contact in a second treatment in which two nets of different mesh sizes were employed as screens. The purpose of this study was: i) to
reconfirm earlier published results that the weathering rate
is enhanced by plants, ii) to evaluate the role of contact
of fine roots with silicate rock in terms of the release of
elements, and iii) to understand if weathering is occurring as a direct consequence of plant physiology.
MATERIALS AND METHODS
Experimental setup
A culture experiment was carried out hydroponically
in a glasshouse controlled at 25°C day and 18°C night
and with 55% to 70% relative humidity under natural light.
An experimental pot was designed with a large, white
plastic exterior box measuring 30 × 15 × 15 cm, which
contained a smaller plastic box with a grating at the bottom with the dimensions 13 × 10 × 8 cm (Fig. 1). We
used nets of two mesh sizes, one of which, at 32 µm, being finer than the diameter of fine roots, the other, at 190
µm being coarser than the fine root diameter.
Plant
Air connecting tube
silicon-rubber sealant
Aluminum foil
Small box
Large box
Net bag with rock grain
Air bubble
Roots
Air pump
10 g of rock
particles
sized 1- 2 mm
nylon net
(32 or 192 µm
pore)
Fig. 1. Schematic diagram of an experimental plant pot. The enlarged diagram of the net bag is shown in the right side.
58
Akter Meheruna and T. Akagi
A nutrient solution with the following composition was
used: KH 2 PO 4 1.93 × 10 –3 M; KNO 3 5.77 × 10 –3 M;
MgSO4·7H2O 2.08 × 10 –3 M; MnSO4·5H2O 2.53 × 10 –6
M; H 3 BO 3 0.028 × 10 –3 M; CuSO 4 2.44 × 10 –6 M;
(NH4)6Mo7O24·4H2O 0.30 × 10–6 M; ZnSO 4·7H2O 1.53 ×
10 –6 M; Ca(NO3)2·4H2O 4.30 × 10–3 M and Fe(III) EDTA
0.19 × 10–3 M. This solution had a conductivity of 1.131
mS/cm at a temperature of 24.7°C with pH 5.8.
At the beginning of the experiment, only 200 ml of
the nutrient solution was added to each of the experimental pots. A further 100 ml of the solution was added only
in the maize pots after 15 days. We deliberately supplied
an insufficient amount of nutrient solution to the plants
with the anticipation that a greater contribution of elements would be released from the rock particles if nutrients were not available in solution. The volume of the
hydroponic solution was 3L. The water level in the experimental pots was maintained at 10 cm depth by adding de-ionized water from time to time during the whole
period of the experiment. All the experimental pots were
covered with aluminum foil to avoid algal growth and
the foil was pierced in a number of places so that the corresponding number of seedlings could be planted as described below. The absence of algae in the experimental
media was confirmed by analyzing suspended matter in
the solutions at the end of the experiment. Each pot was
aerated continuously using an air pump at a rate of 730
ml/min to circulate the medium and keep it well mixed.
The experiment was carried out with four replications for
35 to 38 days.
Preparation of rock particles
Andesite collected in Manazuru, Kanagawa Prefecture, Japan, was used. After crushing the rock, particles
were prepared by dry sieving, washing with tap water
several times to remove immediately dissolvable salts on
the surfaces, drying at 90°C for 3 h, followed again by
dry sieving. We separated rock particles sized 1–2 mm
by sieving. Two types of net bags (32 µm and 190 µm)
were prepared by taking approximately 10 g rock particles within the bags and sealing all the sides with siliconrubber sealant (Fig. 1). The net bags allowed the roots to
grow only around the fine net bag but both around and
into the coarse net bag.
Plant preparation
Two monocots, rice (Oryza sativa) and maize (Zea
mays), and one dicot, soybean (Glycine max L.) were used.
After washing the seeds with a 2 M H2O2 solution for 20
min and then several times with deionized water, they
were germinated on cotton in a separate plastic box. After 10 days of germination, rice and maize seedlings were
transferred into a vessel containing the nutrient solution
and cultivated until they were >6 cm in height. Soybean
seedlings were allowed to grow on cotton without the
nutrient solution. Finally, four maize, two soybean, and
10 rice seedlings per pot were transferred onto the net
bag of the experimental pots through the holes in the
aluminum foil. The same number of seedlings as those
planted was kept aside to determine the initial content of
elements in the plants.
Sampling and analysis
At the beginning of the experiment, we collected the
solution 30 min after setting up the experimental pots with
water and the nutrient solutions in the glass house. At the
end of the experiment, solution and all of the plant material in each pot were collected. Some wilted leaves of
plants dropped off after some time because of nutrient
deficiency. For each pot, we collected these leaves separately.
The sampling solutions were filtered using Nuclepore
filters with a pore size of 0.4 µm, then acidified with HCl
solution and stored for ICP-AES analysis. The fine roots
of all the plants entered into the coarse net bags, but not
into the fine net bags. In the case of soybean and maize,
the fine roots were easily recovered from the coarse net
bags without losing them; however, the fine roots of rice
adhered strongly to the rock particles. After air-drying
the andesite particles, we meticulously collected the detached roots from the rice coarse net bags using plastic
tweezers.
The plant samples were washed and divided into shoot
and root portions. They were digested separately using
the following procedure described by Akiyama et al.
(2004). The plant samples were dried in an oven at 90°C
for three days, weighed, and then stored in airtight
polyethylene bags. After the samples were ground, 0.1 to
0.5 g of each sample was ashed in a nickel crucible in a
muffle furnace at the stepwise-increasing temperatures
of 300, 550, and 700°C. The ash was transferred into a
Teflon beaker using milli Q water and was digested with
Na2CO3 (0.25 to 0.3 g for a 0.15 g sample) on a hotplate
at 250°C for 3 hours. After cooling the beaker, 10 g of
6 M HCl was added, and it was heated again at 250°C on
a hotplate for 7 h. Finally, it was made up to a volume of
100 ml in a plastic bottle with 0.1 M HCl.
Rock particles were dissolved using an acid digestion
method. Finely ground rock powder (0.1 g) was placed in
a Teflon beaker and 10 ml concentrated HNO3 was added.
It was then kept with a lid overnight at 120°C on a hotplate. The beaker was heated at 250°C about 4 h and the
solution was then evaporated to dryness. This digestion
step was repeated twice and supernatant solutions were
collected into a clean plastic bottle. The residue in the
beaker was transferred into a platinum crucible, and
heated in a muffle furnace at 700°C. After cooling, the
residue was treated with concentrated HF on a hotplate.
Andesite weathering by the fine roots of plant
59
The amount of Si was determined gravimetrically from
the difference in weights before and after the HF treatment. The final residue was dissolved with concentrated
HNO3 and the solution was combined with the collected
supernatant solution and was made up to a volume of
about 100 ml with 0.1N HNO 3.
The concentrations of Si, Al, Ca, Mg, Fe and Mn in
the solution, plant, and rock samples were determined with
ICP-AES.
Calculation of the net release of cations
The design of the experimental pots used in the experiments allows the determination of the mass-balance
of elements in its two compartments, i.e., in the whole
plants (sink for elements) and in the rock substrate (source
for elements), which itself consists of a liquid phase (solution) and a solid phase (rock particles). The mass-balance of a measured element in the system (Hinsinger et
al., 1993) can be thus written as
M total = M rock + Msolution + Mplant.
(1)
Because the system is closed, M total remains constant.
Therefore,
Mtotal = Mi rock + Mi solution + M i plant
= M f rock + Mf solution + M f plant,
(2)
where i and f represent initial and final values, respectively. Then,
(Mi rock – Mf rock) = (Mf plant – Mi plant)
– (Mi solution – Mf solution).
(3)
This equation can be simply written as
Mrelease = M uptake – Mdepletion.
(4)
We calculated the release of elements from rock particles
in each pot using Eq. (4).
RESULTS AND DISCUSSION
pH variation of experimental media
Throughout this experiment, we measured the pH of
the experimental media at seven-day intervals (Table 1).
After seven days, the pH of all the media except that in
the soybean pots increased, probably because of a surface exchange reaction of rock particles with solution and/
or rapid initial dissolution of rock particles. It is now generally accepted that pH changes in the rhizosphere essentially originate from the differential rates of cation and
anion uptake causing an imbalance of positive and negative charges entering the root. This imbalance is compensated by releasing hydrogen and hydroxyl ions (Nye,
1981; Haynes, 1990).
Low pH in the soybean media indicates that the rate
of nutrient uptake by soybean might be higher than that
by the other plants from the onset of growth. Legumes
relying on symbiotic N 2 fixation have been shown to
acidify their rhizosphere due to an excess uptake of cations (Jarvis and Robson, 1983; Römheld, 1986). There was
obviously a trend of decreasing pH in the presence of the
plants. After a while, the pH of all the planted pots became constant, possibly because plant growth was restricted by a nutrient deficiency. The pH differed significantly between the blank and planted pots during the
whole experimental period (from t-test). The pH in the
planted pots decreased by 1–1.5 units compared to that
in the blank pots (Table 1). The pH differences among
the pots with different crop species were highly significant (p < 0.01) but the different mesh sizes did not result
in any significant pH differences (from two way ANOVA).
Differences in both the rate of plant growth and the
dissolution of rock powder were thought to be mainly
Table 1. pH variation of media during the experimental period
Days
Coarse Net Pot
Fine Net Pot
Blank
Rice
Maize
Soybean
7
14
21
28
6.15(±0.10)Ca
6.13(±0.10)Ca
6.33(±0.30)Aa
6.35(±0.31)Aa
7.15(±0.06)Aa*
7.10(±0.14)Aa**
6.68(±0.22)Ab
5.68(±0.15)Bc
6.50(±0.00)Ba**
6.55(±0.13)Ba*
4.98(±0.10)Bb*
4.80(±0.12)Cbc
4.93(±0.06)Da
4.73(±0.06)Db
4.87(±0.06)Ba
4.70(±0.10)Cb
35
38
6.25(±0.31)Aa
6.33(±0.31)Aa
5.23(±0.21)Bd
4.75(±0.06)Cc
4.58(±0.05)Bd*
4.70(±0.00)Cb
Blank
Rice
Maize
Soybean
6.13(±0.05)Ca 7.28(±0.05)Aa*
6.20(±0.14)Ba 6.80(±0.08)Ab**
6.35(±0.13)Ba 6.63(±0.13)Ab
6.33(±0.10)Aa 5.58(±0.26)Bc
6.28(±0.05)Bb**
6.38(±0.05)Ba*
4.78(±0.10)Cd*
4.90(±0.00)Cc
5.10(±0.10)Da
4.77(±0.12)Cb
4.80(±0.00)Cb
4.57(±0.06)Dc
6.23(±0.10)Aa 5.03(±0.10)Bd
6.35(±0.06)Aa
4.70(±0.00)Cd
4.50(±0.00)Be*
4.73(±0.06)Cb
Deviations in one sigma ( σ) are in parentheses.
Statistic Analyses) t-test: The mean values did not significantly differ when followed by the same letter in a given column and by the same capital
letter in the same row (p = 0.05 to <0.001) within the coarse or fine net pots. **,*significant difference at p < 0.01 and p < 0.05 level,
respectively, between the coarse and fine net pots. two-ways ANOVA: Significant difference at p < 0.01 level among the species with time but no
significant difference within the species between the two types of pots.
60
Akter Meheruna and T. Akagi
Andesite weathering by the fine roots of plant
61
Si
6.57 (±2.26)
Ca
25.20 (±0.24)
Mg
9.53 (±0.08)
Al
0.62 (±0.01)
Fe
1.55 (±0.04)
Mn
0.22 (±0.002)
Dry wt. (g)
Root: Shoot ratio
Si
Maize
Soybean
6.37 (±2.41)
Ca
17.97 (±0.30)
Mg
7.36 (±0.06)
Al
0.42 (±0.002)
Fe
1.01 (±0.01)
Mn
0.14 (±0.003)
Dry wt. (g)
Root: Shoot ratio
Si
8.63 (±2.60)
Ca
17.02 (±0.23)
Mg
6.29 (±0.14)
Al
0.41 (±0.01)
Fe
1.09 (±0.05)
Mn
0.14 (±0.07)
Dry wt. (g)
Root: Shoot ratio
Rice
0.25 (±0.02)*
0.42 (±0.02)
0.02 (±0.001)
0.50 (±0.01)**
0.001 (±0.003)
8.86 (±0.12)**
0.74 (±0.04)
0.04 (±0.01)
0.25 (±0.05)
0.67 (±0.08)
0.017 (±0.003)
5.74 (±0.11)
1.22 (±0.09)
0.06 (±0.01)
0.24 (±0.01)
0.48 (±0.04)
0.011 (±0.002)*
1.70 (±0.25)
0.26 (±0.05)
0.27 (±0.07)
6.36 (±0.14)
0.33 (±0.009)
6.02 (±0.13)
0.39 (±0.06)
0.97 (±0.05)
0.12 (±0.01)
6.31 (±2.60)
17.42 (±0.23)
Solution
6.37 (±2.40)
16.56 (±0.14)
Initiala)
0.36 (±0.03)
0.56 (±0.06)
3.15 (±0.29)
4.59 (±0.36)
0.17 (±0.05)
0.11 (±0.01)
0.84 (±0.04)
16.16 (±1.79)
1.29 (±0.13)
0.37 (±0.04)
3.30 (±0.28)
7.38 (±0.52)
0.26 (±0.09)
0.14 (±0.04)
3.65 (±0.43)
24.43 (±1.54)
0.93 (±0.07)
1.52(±0.15)
0.59 (±0.04)
4.32 (±0.25)
0.18 (±0.05)
0.08 (±0.02)
3.64 (±0.67)
1.70 (±0.32)
0.21 (±0.04)
0.22 (±0.04)
0.18 (±0.07)
1.13 (±0.25)
0.29 (±0.11)
6.29 (±0.51)
0.38 (±0.07)
0.33 (±0.05)
0.74 (±0.10)
4.28 (±0.41)
1.13 (±0.15)
19.80 (±1.13)
7.56 (±0.42)
0.52 (±0.14)
1.01 (±0.03)*
0.56 (±0.03)
0.3 (±0.04)
0.38 (±0.12)
0.64 (±0.04)
4.26 (±0.27)
7.24 (±0.53)
3.56 (±0.25)
4.29 (±0.46)
31.67 (±1.52)*
10.94 (±0.38)
6.75 (±0.34)
0.42 (±0.08)**
0.49 (±0.25)*
0.81 (±0.03)
2.94 (±0.32)
11.25 (±1.46)
13.26 (±0.66)






Plant total
0.64 (±0.18)
2.66 (±0.42)
2.43 (±0.19)
0.24 (±0.05)
0.41 (±0.26)
0.22 (±0.02)
1.42 (±0.19)
2.24 (±0.27)












9.01 (±1.25)
10.60 (±0.64)
Root
Coarse Pot (Final)
Shoot
0.30 (±0.02)*
0.43 (±0.04)
0.02 (±0.001)
0.46 (±0.01)**
0.007 (±0.01)
8.11 (±0.26)**
0.05 (±0.01)
0.37 (±0.11)
0.72 (±0.07)
0.015 (±0.001)
5.92 (±0.43)
0.72 (±0.06)
1.60 (±0.25)
1.11 (±0.17)
0.07 (±0.01)
0.22 (±0.03)
0.46 (±0.05)
0.014 (±0.002)*
6.05 (±0.35)
0.37 (±0.004)
0.29 (±0.08)
0.21 (±0.10)
6.02 (±1.93)
16.90 (±0.58)
Solution

4.73 (±0.01)
0.20 (±0.04)
0.17 (±0.02)
0.56 (±0.02)
2.98 (±0.13)
0.37 (±0.03)
0.98 (±0.06)
17.26 (±0.16)
6.89 (±0.53)
0.29 (±0.05)
0.13 (±0.01)
0.31 (±0.04)
3.28 (±0.25)
1.23 (±0.09)
3.30 (±0.09)
21.82 (±2.25)
0.54 (±0.04)
1.44 (±0.12)
1.00 (±0.03)
4.29 (±0.15)
0.14 (±0.07)
0.12 (±0.03)
5.60 (±2.06)
10.34 (±0.43)





2.55 (±0.69)
1.66 (±0.11)
0.21 (±0.07)
0.17 (±0.02)
0.19 (±0.03)
1.10 (±0.06)
0.15 (±0.14)
6.73 (±0.55)
3.45 (±0.33)
0.25 (±0.04)
0.41 (±0.05)
0.61 (±0.06)
4.01 (±0.20)
0.53 (±0.11)
3.02 (±0.23)
2.60 (±0.19)
0.67 (±0.02)
0.83 (±0.17)
0.26 (±0.02)
1.45 (±0.14)
3.92 (±0.22)






Root
Fine Pot (Final)
Shoot
Deviations in one sigma ( σ) are in parentheses.
Initial content is the sum of the measured content of the de-ionized water, additional nutrient solution and seedlings content in the plants pot (except for soybean).
Statistic Analyses) t-test: *,**Significant difference at p < 0.05 and p < 0.01 level, respectively, between the coarse and fine net pots.
a)
Si
Blank
Ca
Mg
Al
Fe
Mn
Element
Sample
Table 2. Amount of elements (mg/pot) in the experiment of 35–38 days plant/andesite system
6.39 (±0.11)
0.41 (±0.10)
0.34 (±0.01)
0.75 (±0.01)
4.08 (±0.09)
1.13 (±0.13)
19.81 (±0.85)
0.54 (±0.03)
0.54 (±0.06)
0.92 (±0.06)*
7.29 (±0.39)
10.34 (±0.52)
3.83 (±0.17)
28.55 (±2.12)*
6.89 (±0.31)
0.81 (±0.07)**
0.95 (±0.17)*
0.80 (±0.05)
2.89 (±0.26)
9.52 (±2.11)
13.36 (±0.37)






Plant total
15
1.25
Si
1
10
Al
0.75
0.5
5
0.25
0
0
Amount / mg
40
1.25
Ca
1
30
Mn
0.75
20
0.5
10
0.25
0
0
12.5
10
2
Mg
Fe
1.5
7.5
1
5
Blank
Soybean
Rice
Solution
Maize
Root
Blank
Shoot
Soybean
Rice
Fine
Coarse
Fine
Coarse
Fine
Coarse
Fine
F ine
C oarse
F ine
C oarse
F ine
C oarse
F ine
0
C oarse
0
Coarse
0.5
2.5
Maize
Initial a)
Fig. 2. Mass balance of elements in the andesite/plant hydroponic system at the end of the cultivation experiment. Deviation bars
stand for one sigma of the total contents. a) The amount in the media + seedlings (rice and maize) at the beginning of the
experiment.
responsible for the pH differences, which were, however,
not systematic. For all the plants, the pH at the end of the
experiment was almost the same in both the coarse and
fine net pots. The decreasing pH in all the media of the
planted pots presumably enhanced the release of elements
from the rock particles.
The amount of elements in solution
Although the nutrient solution was not prepared to
contain Si and Al, the notable presence of these elements
at the beginning of the experiment (Table 2) indicates that
some contamination occurred from the de-ionized water
and the circulated air in the glasshouse, before the experimental pots were covered with aluminum foil. We
observed the depletion of Ca, Mg and Mn in the solution
of all the planted pots at the end of the experiment, which
is in marked contrast to the concentration in the blank
62
Akter Meheruna and T. Akagi
pots (Table 2). This indicates a large requirement for these
elements by the plants. Amongst the three species studied in this experiment, the depletion of Ca, Mg, and Mn
was observed in the decreasing order of maize > soybean
> rice pots, which is consistent with the order of the dry
weight.
At the end of the experiment, the greatest amount of
Si was present in the soybean media, which indicates its
lower requirement by soybean than by the other plants. A
distinct Si depletion was found in the rice pots (Fig. 2).
Silicon is a nonessential element, which can, however,
be taken up in large amounts by many, but especially
graminaceous species (Epstein, 1999); its upake varies
within the graminaceous species depending on the root
transportation system (Marschner, 1995). For example,
rice can take up Si actively, whereas maize takes it up
passively (Marschner, 1995).
Table 3. Release of different elements (mg/10g) in the plant/andesite system
Elements
Si
Ca
Mg
Mn
Fe
Al
Coarse Net Pot
Fine Net Pot
Blank
Rice
Maize
Soybean
Blank
Rice
Maize
Soybean
−0.06(±0.37)
0.86(±0.16)
0.34(±0.10)
0.15(±0.08)
−0.71(±0.10)
0.15(±0.07)
4.33(±1.70)a
−2.54(±0.64)c
0.53(±0.33)b
0.68(±0.003)b
−0.11(±0.27)a*
0.24(±0.08)a**
3.45(±0.51)a
7.21(±1.53)a*
1.45(±0.38)a
0.80(±0.03)a*
−0.35(±0.22)a
0.19(±0.06)a
3.62(±0.04)a*
3.74(±1.14)b
0.28(±0.51)b
0.63(±0.10)b
−0.22(±0.06)a
0.22(±0.07)a
−0.35(±0.61)
0.34(±0.84)
0.03(±0.38)
0.09(±0.29)
−0.68(±0.07)
0.09(±0.04)
2.81(±2.03)a
−2.55(±0.26)c
0.67(±0.30)a
0.67(±0.005)a
0.32(±0.18)*
0.62(±0.08)a**
3.19(±0.34)a
4.07(±2.07)a*
0.86(±0.52)a
0.71(±0.06)a*
−0.29(±0.06)
0.29(±0.09)b
2.87(±0.35)a*
3.71(±0.86)b
0.38(±0.10)a
0.64(±0.001)a
−0.21(±0.04)
0.29(±0.13)b
Elements release was calculated using the equation: M release = M uptake – M depletion, where, Muptake = Final Plant Content – Initial Plant Content
and Mdepletion = Initial Water Content – Final Water Content.
Deviations in one sigma ( σ) of sample distribution are in parentheses.
Statistic Analyses) t-test: The mean values did not significantly differ when followed by the same small letter in a given row within the coarse or
fine net pots of plants.
**,*Significant difference at p < 0.01 and p < 0.05 level, respectively, between the coarse and fine net pots.
ANOVA: The released amount of Ca differed significantly (p < 0.01) among the species of each types of pots.
The comparatively smaller depletion of Fe in the media compared with that of the other essential elements
(Table 2) in all of the planted pots indicates a low requirement and/or its dissolution from rock particles. Furthermore, a proportion of the Fe and Al might have been
precipitated or adsorbed on the surfaces of the rock particles as oxides in all of the experimental pots, which is
inferred from the decrease of these elements in solution
of the blank pots.
When comparing the coarse and fine net pots, the only
significant differences in the elemental amounts of solution were for Si, Ca (p < 0.01) and Al (p < 0.05) in the
soybean pot and for Mn (p < 0.05) in the rice pots.
Plant growth
The total dry weights and root/shoot ratios of the plant
species are given in Table 2. Their dry weights differed
widely by the end of the experiment, with maize achieving the greatest growth, as was also noted in an earlier
report (Hinsinger et al., 2001). The growth of rice, maize
and soybean ceased after 15, 25 and 22 days, respectively,
owing to nutrient deficiency, which had been expected
(see Section “Materials and Methods”). The average shoot
and root dry weights of all the plants did not differ significantly between the coarse and fine net pots. The root/
shoot weight ratios depend on the nutrient uptake of a
plant (Marschner, 1995). The high root/shoot ratio observed for maize was due to its huge root mass.
Total amounts of elements in plant
The total amounts of elements in the plants are shown
in Table 2 and Fig. 2. The total amount of Ca present
decreased in order of maize > soybean > rice. The highest amount of Mg was found in maize. Although the total
dry matter of soybean was higher than that of rice, the
total Mg content of rice was a little higher than that of
soybean. The rate of Mg uptake can be strongly depressed
by other cations, such as K+ Ca2+, and Mn2+ (Heenan and
Campbell, 1981), as well as by H+, that is, by a low pH.
In the study reported here, greater Ca uptake and lower
pH in the soybean pots might have inhibited Mg uptake.
Rice, as a Si accumulating plant, accumulated this element more than the other plants. In all the plants, the
amount of Mn was higher than Fe, reflecting the greater
requirement for Mn than for Fe.
Ions with no apparent physiological role, such as various forms of Al, may also be taken up if they are present
at the root surface. Root uptake of Al is metabolically
regulated by exclusion mechanisms and by active excretion (Andersson, 1988). Fe and Al were deposited mainly
in the root portion of all plants. Aluminum is a nonessential and potentially highly toxic element. It is thus not
surprising that plants exhibited much lower concentrations of Al than Si.
Release of elements from rock particles in the presence
of plants
We calculated the release of elements from the rock
particles in the presence of plants by subtracting the depletion of the elements in the media from the uptake by
the plants using Eq. (4). The calculated amounts of the
elements released during the experimental period are
given in Table 3.
At the end of the experiments, the amounts of Ca, Mg,
Mn and Al released in the blank pots were small and did
not differ significantly between the coarse and fine net
pots. In the presence of the plants, the release of elements
was increased by a factor of 12–28 for Mg and 4–7 for
Mn. Maize and soybean enhanced Ca release by a factor
of 8–12 and 4–11, respectively. It was not possible to calculate the multiplying factors of Si and Fe release in the
presence of the plants due to the negative values of the
Andesite weathering by the fine roots of plant
63
elements in the blank pots. The amount of Si released in
the presence of the plants calculated using Eq. (4) was 2–
4 mg, which is comparable to those of Ca and Mg. The
derived quantities of Fe and Al were also higher in the
presence of plants. Of the three species, maize showed
the highest releases of Ca, Mg and Mn, which probably
reflects its higher requirement for these elements but only
the amounts of Ca released differed significantly (p <
0.01) among the species (Table 3). The ability to extract
particular elements varied from plant to plant depending
on their growth and uptake. The negative value of Fe release (Table 3) possibly indicates its adsorption as oxides or organometallic complexes on the surfaces of the
rock particles in all the pots and/or precipitation on the
root surface in the planted pots.
The measured ratio of major elements in the andesite
was Si:Ca:Mg:Al = 1:0.06:0.01:0.21 by weight. Compared
with their abundance in the rock, Ca, Mg and Mn were
preferentially released with respect to Si in the presence
of the plants, excluding the rice, whereas Fe and Al were
more likely to be retained in the solid andesite. The released Ca/Si ratio in the maize and soybean pots was 21–
34 times and 17–22 times that of their weight ratios in
the andesite, respectively. The results suggest the incongruent dissolution of the andesite and/or preferential dissolution of plagioclase feldspar. However for rice, the lack
of Ca release may indicate no dissolution of plagioclase
feldspar. Rice is a typical Si-accumulating plant and is
able to accumulate this element sufficiently for it to constitute up to 10% of the shoot dry weight, depending upon
the Si concentration in soil solutions (Ma et al., 2001). In
the case of the rice in the experiment, the initial Ca level
in the media may have been sufficient, or its high Si
amount might have inhibited Ca uptake. Ma and Takahashi
(1993) found that the addition of Si to the nutrient solution restricted the uptake of Ca by rice.
The released Mg/Si ratios were also higher in all the
planted pots than they were in the andesite. This suggests
that some ferromagnesian minerals within the andesite
might have been dissolved. Eick et al. (1996a, b) showed
a systematic and preferential release of Mg relative to Si.
In our experiment, the order of elements released from
the andesite was: Ca > Si > Mg in weight, both in the
maize and soybean pots. A higher Mn requirement, which
exceeded the amount supplied initially, accelerated Mn
uptake by the plants through two different mechanisms
possibly from Mn oxide or silicate minerals such as biotite
and hornblende as discussed below.
Notwithstanding a low requirement for Fe, as well as
an adequate initial amount in all the planted pots, the
plants caused a release of Fe from the rock particles. The
release of Fe and Mn by the plants may take place as two
different mechanisms. Graminaceous species, rice and
maize, secrete phytosiderophores which exhibit strong
64
Akter Meheruna and T. Akagi
complexing properties with respect to ferric Fe (Takagi
et al., 1984). The phytosiderophores can complex with
other micronutrients such as Mn (Treeby et al., 1989).
The graminaceous species thus can transport the nutrients by a specific uptake system on the root surface
(Römheld, 1987). On the other hand, dicotyledon plants
such as soybean can take up the micronutrients through a
reduction process by the reducing activity of the roots
(Römheld, 1991). In addition, it has been shown that the
exudation of organic acids by the plant roots increases as
a response to nutrient deficiencies (Jones and Darrah,
1995), which might suggest their possible implication in
the acquisition of mineral nutrients from the rhizosphere.
Many plants have been reported to release exudates that
can potentially form complexes with Fe and Mn, such as
citrate and malate (e.g., Jones et al., 1996; Dinkelaker et
al., 1989). Although we could not explain clearly the negative Fe release in the presence of the plants, the Mn release might be through the process of reduction and organic complexation in the planted pots of our experiment.
The amount of Al released was higher in all the fine
planted pots which differed significantly (p < 0.001) only
for rice from the coarse net pot. The variable solubility
of Al is illustrated by the well-known relationship between
pH and the activity of soluble monomeric Al (Lindsay,
1979). In our experiment, we observed the highest amount
of Al release in the fine net pots of rice, but the pH of rice
pots was higher than that of other planted pots. At around
pH 4.5, aqueous Al is mostly in ionic Al3+ form and, at
pH values above 4.5, the activity of ionic Al3+ decreases
with increasing pH. The activity of hydroxy-Al as well
as Al complexed with organic compounds is increased at
higher pH. The organic anions such as oxalate and citrate,
which are known Al-chelating agents (Kochian, 1995),
might have been produced in larger amounts in the fine
net pots by the plant roots. Furthermore, aqueous Al may
represent only a part of the mobilized Al (Rufyikiri et al.,
2004). Apart from root absorption, some of it might have
been sorbed on the exchange sites of the rock particles.
Effect of contact with fine roots on release or uptake of
elements from rock particles
Only in maize, which showed the greatest growth of
the three species, were the amounts of Ca and Mn released in the coarse net pots significantly higher than those
in the fine net pots (p < 0.05); the differences in the
amounts of Si and Mg released were less significant (p <
0.1). On the other hand, the differences in the release rates
of the elements between the two mesh sizes were statistically insignificant, except for Si (p < 0.05) in the soybean
pot and, in the opposite sense, for Fe (p < 0.05) and Al (p
< 0.001) in the rice pots.
The much greater extraction of elements by maize in
the coarse, when compared with the fine net pots may be
due to the active role of fine roots. Two possible mechanisms for the enhancement of elemental release by the
direct contact of fine roots are cation exchange mechanism, in which cations may be exchanged directly between
the minerals of rock particles and fine roots at a time of
high nutrient demand, and a surface trapping mechanism,
whereby some mineral particles from the rock may became trapped on the external surface of the root and then
incorporated into the root cortex with growth (Fu et al.,
2002). In our experiment, we observed increasing contact of fine roots with rock particles in the coarser net
pots in the order: soybean < maize < rice. Although nutrient depletion and deficiency occurred after two weeks
of the experiment, the high nutrient demand of maize
possibly accelerated the uptake of nutrient elements in
the coarse net pot by the direct contact of fine roots with
rock particles through the mechanisms discussed above.
Weathering rate enhancement by plants
In our experiment, the presence of plants resulted in a
substantial increase in the release of elements from the
andesite, with a major proportion being accumulated in
the plant tissues (Fig. 2). The varying release patterns of
several elements might be due to the difference in the
types of minerals in the rock particles attacked by the
plants or to the specific effects of the root exudates, as
mentioned in the previous sections.
The positive effects of plants on weathering have been
quantified in the laboratory as well as in field experiments.
In one of numerous studies, Hinsinger et al. (1993) grew
plants with phologopite mica as the sole source for both
K and Mg, and found that the plants were able to increase
the release of K and Mg by a factor of two to four after
four days. In a similar experiment using phosphate rock
and alumina sand as the mineral substrate, dissolution was
significantly enhanced by the action of the roots
(Hinsinger and Gilkes, 1997).
On a field scale, Moulton and Berner (1998) measured
the release of Ca and Mg into streams and growing trees
in areas of both vegetated and barren basaltic rocks in
Iceland. The rate of release was two to five times higher
in the area with vegetation. According to Hinsinger et al.
(2001), the amounts of Si, Ca, Mg and Na released from
basalt under leaching conditions in a laboratory increased
by a factor ranging from one to five in the presence of
crop plants. They also observed that the increase in the
amount of Fe released from basalt reached a maximum
of about 100-to 500-fold when it was associated with
banana and maize. The calculated dissolution rates of
ferromagnesian silicates (Taylor and Velbel, 1991) and
basalt (Benedetti et al., 1994) are reported to increase
when the botanical uptake was taken into account in calculation.
From our controlled glasshouse experiment, we ob-
tained not only the enhancement of the release of nutrient elements from the andesite particles by the crop plants
but also comparatively higher release by the direct contact of fine roots related to a high growth requirement.
CONCLUSION
The positive effect of plants on the release of elements
from andesite rock was confirmed by this study. The
amount released was found to vary depending on the plant
species. A marked increase was observed for Si (rice), Ca
(maize and soybean), Mn (maize, soybean and rice) in
the presence of the plants. Species-dependent variation
is caused by multiple factors such as the effect of specific root exudates (e.g., phytosiderophores), different
chemical conditions in the rhizospheres, different nutrient requirement and growth rates of different species etc.
The significant differences between the two mesh sizes
were observed only in maize pots, although the differences were smaller than the species-dependent contrasts.
No systematic difference in the pH and dissolved concentration of elements of the media was seen between
the two types of pots, except in the early stage. The difference between the two mesh sizes pots, therefore, implies: i) the proximity of the fine roots to the andesite
particles caused a variation in the release of elements from
the rock particles and ultimately in the uptake of elements
by the plants, ii) direct contact of fine roots on the rock
particles could be one of the important processes to absorb some of nutrients such as Si and Ca, and iii) weathering may be one of the consequences of the physiological activity of plants.
The contribution to plant nutrition of the direct contact between the fine roots and rock particles was, however, not explicitly answered in this study because of the
artificial growing conditions and some unexpected contamination. The contribution might have been underestimated because of restricted contact of roots with rock
particles by the nets and of severe nutrient deficiency
during the early stage of growth. Additionally, the amounts
of elements released from 10 g of andesite, an intermediate rock, the short period of the experiment, and use of
young plants probably led to an underestimate of the plant
effect. Further studies considering the micro-scale chemical, mineralogical, and physiological aspects are needed
to refine our results and to confirm the contribution of
fine roots to weathering and plant nutrition. To verify the
amount of elements released from the rock particles by
the fine roots, future experiments should, therefore, be
performed with sufficient nutrient, since the role of fine
roots tended to be important in growing plants without
any nutrient deficiency. This work may ultimately lead
to the suggestion that rock particles be used as slow release fertilizers for field crops.
Andesite weathering by the fine roots of plant
65
Acknowledgments—The authors thank Professor Matsumura,
Department of Crop Science, Tokyo University of Agriculture
and Technology for supplying plant seeds. We also gratefully
acknowledge the helpful reviews of James I. Drever and anonymous reviewer.
REFERENCES
Akiyama, K., Nakanishi, T., Ohnishi, R. and Akagi, T. (2004)
Dissolution method for the determination of silicon in plant
samples with ICP-AES. Bunseki Kagaku 53(11), 1229–1232.
Andersson, M. (1988) Toxicity and tolerance of aluminum in
vascular plants: a literature review. Water Air Soil Pollut.
39, 439–462.
Barak, P., Chen, Y. and Singer, A. (1983) Ground basalt and
tuff as iron fertilizer for calcareous soils. Plant Soil 73, 155–
158.
Benedetti, M. F., Menard, O., Noack, Y., Carvalho, A. and
Nahon, D. (1994) Water-rock interactions in tropical catchments: Field rates of weathering and biomass impact. Chem.
Geol. 118, 203–220.
Berner, R. A. (1992) Weathering, plants, and the long-term carbon cycle. Geochim. Cosmochim. Acta 56, 3225–3231.
Berner, R. A. (1995) Chemical weathering and its effect on atmospheric CO2 and climate. Chemical Weathering Rates of
Silicate Minerals (White, A. F. and Brantley, S. L., eds.),
Reviews in Mineralogy 31, 565–583.
Boyle, J. R. and Voight, G. K. (1973) Biological weathering of
silicate minerals: Implications for tree nutrition and soil
genesis. Plant and Soil 38, 191–201.
Coroneos, C., Hinsinger, P. and Gilkes, R. J. (1996) Granite
powder as a source of potassium for plants: A glasshouse
bioassay comparing two pasture species. Fert. Res. 45, 143–
152.
Darrah, P. R. (1993) The rhizosphere and plant nutrition: a quantitative approach. Plant Soil 155/156, 1–20.
Dinkelaker, B., Römheld, V. and Marschner, H. (1989) Citric
acid excretion and precipitation of calcium citrate in the
rhizosphere of white lupin (Lupinus albus L.). Plant Cell
Envir. 12, 285–292.
Drever, J. I. (1994) The effect of land plants on weathering
rates of silicate minerals. Geochim. Cosmochim. Acta 58,
2325–2332.
Eick, M. J., Grossl, P. R., Golden, D. C., Sparks, D. L. and
Ming, D. W. (1996a) Dissolution of a lunar basalt stimulant as affected by pH and organic anions. Geoderma 74,
139–160.
Eick, M. J., Grossl, P. R., Golden, D. C., Sparks, D. L. and
Ming, D. W. (1996b) Dissolution kinetics of a lunar basalt
stimulant at 298 K. The effect of pH and organic acids.
Geochim. Cosmochim. Acta 56, 157–170.
Epstein, E. (1999) Silicon. Annu. Rev. Plant Physiol. Plant Mol.
Biol. 50, 641–664.
Fu, F., Akagi, T. and Yabuki, S. (2002) Origin of silica particles found in the cortex of Matteuccia roots. Soil Sci. Soc.
Am. J. 66, 1265–1271.
Gillman, G. P. (1980) The effect of crushed basalt scoria on the
cation exchange properties of a highly weathered soil. Soil
Sci. Soc. Am. J. 44, 465–468.
66
Akter Meheruna and T. Akagi
Gobran, G. R., Clegg, S. and Courchesne, F. (1999) The
rhizosphere and trace element acquisition in soils. Fate and
Transport of Heavy Metals in the Vadose Zone (Selim, H.
M. and Iskander, A., eds.), 225–250, CRC Press.
Graustein, W. C. (1981) The effects of forest vegetation on solute acquisition and chemical weathering: A study of the
Tesuque watersheds near Santa Fe, New Mexico. Ph.D. dissertation, Yale University, New Haven, U.S.A.
Grinsted, M. J., Hedley, M. J., White, R. E. and Nye, P. H. (1982)
Plant induced changes in the rhizosphere of rape (Brassica
napus var. Emerald) seedlings. I. pH change and the increase
in P concentration in the soil solution. New Phytol. 91, 19–
29.
Harley, A. D. and Gilkes, R. J. (2000) Factors influencing the
release of plant nutrient elements from silicate rock powders: A geochemical overview. Nutr. Cycling. Agroecosyst.
56, 11–36.
Haynes, R. J. (1990) Active ion uptake and maintenance of
cation-anion balance: A critical examination of their role in
regulating rhizosphere pH. Plant Soil 126, 247–264.
Heenan, D. P. and Campbell, L. C. (1981). Influence of potassium and manganese on growth and uptake of magnesium
by soybeans (Glycine max (L.) Merr. cv Bragg). Plant Soil
61, 447–456.
Hinsinger, P. (1998) How do plant roots acquire minerals nutrients? Chemical processes involved in the rhizosphere.
Adv. Agron. 64, 225–265.
Hinsinger, P. and Gilkes, R. J. (1997) Dissolution of phosphate
rock in the rhizosphere of five plant species grown in an
acid, P-fixing mineral substrate. Geoderma 75, 231–249.
Hinsinger, P., Elsass, F., Jaillard, B. and Robert, M. (1993) Rootinduced irreversible transformation of a trioctahedral mica
in the rhizosphere of rape. J. Soil Sci. 44, 535–545.
Hinsinger, P., Barros, O. N. F., Benedetti, M. F., Noack, Y. and
Callot, G. (2001) Plant-induced weathering of a basaltic
rock: Experimental evidence. Geochim. Cosmochim. Acta
65(1), 137–152.
Holland, H. D. (1978) The Chemistry of the Atmosphere and
Oceans. Wiley, New York, 351 pp.
Jackson, T. A. (1996) Comment on “The effect of land plants
on weathering rates of silicate minerals” by James I. Drever.
Geochim. Cosmochim. Acta 60, 723–724.
Jaillard, B. (1987) Techniques for studying the ionic environment at the soil/root interface. Proceedings of the 20th International Potash Institute Colloquium, 23–25 June 1987,
231–245, Baden bei Wien, Austria.
Jarvis, S. C. and Robson, A. D. (1983) The effects of nitrogen
nutrition of plants on the development of acidity in Western Australia soils. I. Effects with subterranean clover grown
under leaching conditions. Aust. J. Agric. Res. 34, 341–353.
Jones, D. L. and Darrah, P. R. (1995) Influx and efflux of organic acids across the soil-root interface of Zea mays L.
and its implications in rhizosphere C flow. Plant Soil 173,
103–109.
Jones, D. L., Darrah, P. R. and Kochian, L. V. (1996) Critical
evaluation of organic acid mediated dissolution in the
rhizosphere and its potential role in root iron uptake. Plant
Soil 180, 57–66.
Kochian, L. V. (1995) Cellular mechanisms of aluminum tox-
icity and resistance in plants. Annu. Rev. Plant Physiol. Plant
Mol. Biol. 46, 237–260.
Lindsay, W. L. (1979) Aluminum. Chemical Equilibria in Soils,
34–49, Wiley, New York.
Ma, J. F. and Takahashi, E. (1993) Interaction between calcium
and silicon in water-cultured rice plants. Plant Soil 148,
107–113.
Ma, J. F., Miyake, Y. and Takahashi, E. (2001) Silicon as a beneficial element for crop plants. Silicon in Agriculture
(Datnoff, L. E., Snyder, G. H. and Korndorfer, G. H., eds.),
17–39, Elsevier Science, Netherland.
Marschner, H. (1995) Mineral Nutrition of Higher Plants. 2nd
ed., 537–594, Academic Press, London.
Moulton, K. L. and Berner, R. A. (1998) Quantification of the
effect of plants on weathering: studies in Iceland. Geology
26, 895–98.
Nye, P. H. (1981) Changes of pH across the rhizosphere induced by roots. Plant Soil 61, 7–26.
Robert, M. and Berthelin, J. (1986) Role of biological and biochemical factors in soil mineral weathering. Interactions of
Soil Minerals with Natural Organics and Microbes (Huang,
P. M. and Schnitzer, M., eds.), SSSA Spec. Publ. 17, 453–
495, ASA, CSSA, and SSSA, Madison, WI.
Römheld, V. (1986) pH-Veränderungen in der Rhizosphäre
verschiedener Kulturpfla-nzenarten in Abhängigkeit vom
Nährstoffangebot. Potash Rev. 55, 1–8.
Römheld, V. (1987) Existence of two different strategies for
the acquisition of iron in higher plants. Iron Trasport in
Animals, Plants, and Microorganisms (Winkelmann, G., Van
der Helm, D. and Neilands, J. B., eds.), 353–374, VCH
Chemie, Weinheim, FRG.
Römheld, V. (1991) The role of phytosiderophores in acquisition of iron and other micronutrients in graminaceous species: An ecological approach. Plant Soil 130, 127–134.
Rufyikiri, G., Nootens, D., Dufey, J. E. and Delvaux, B. (2004)
Mobilization of aluminium and magnesium by roots of banana (Musa spp.) from kaolinite and smectite clay minerals. Applied Geochemistry 19, 633–643.
Takagi, S., Nomoto, K. and Takemoto, T. (1984) Physiological
aspects of mugineic acid, a possible phytosiderophore of
graminaceous plants. J. Plant Nutr. 7, 469–477.
Taylor, A. B. and Velbel, M. A. (1991) Geochemical mass balances and weathering rates in forested watersheds of the
Southern Blue Ridge II. Effects of botanical uptake terms.
Geoderma 51, 29–50.
Treeby, M., Marschner, H. and Römheld, V. (1989) Mobilization of iron and other micronutrient cations from a calcareous soil by plant-borne, microbial, and synthetic metal
chelators. Plant Soil 114, 217–226.
Uren, N. C. (1981) Chemical reduction of an insoluble higher
oxide of manganese by plant roots. J. Plant Nutr. 4, 65–71.
Andesite weathering by the fine roots of plant
67