Journal of Soil Science, 1988,39,493-504

Effect of liming and phosphate additions on sulphate

leaching in soils
N. S. BOLAN, J. K. SYERS*, R. W. T I L L M A N & D. R. S C O T T E R
Fertilizer and Lime Research Centre, Massey University, Pa[merston North, New Zealand

SUMMARY
The effect of lime (CaCO,) and phosphate additions on surface charge characteristics and
their effect on the leaching of sulphate were examined for two soils (Patua loam and
Tokomaru silt loam) which differed in their adsorption capacities for sulphate.
Incubation of soils with either CaCO, (0-600 mmol kg-') or phosphate (0-208 mmol
kg-') resulted in a two- to five-fold increase in the net negative charge and a similar
decrease in the adsorption of sulphate. The effect of either lime or phosphate addition on
both the surface charge and sulphate adsorption was more pronounced for the allophanic
Patua soil than for the Tokomaru soil containing mainly vermiculite.
In a column experiment, liming induced the leaching of sulphur either by the desorp-
tion of adsorbed sulphate or by the mineralization of organic sulphur. During a miscible
displacement study, addition of either CaCO, or phosphate resulted in an early break-
through of sulphate in the leachate. In a pulse experiment, in which soils were incubated
with sulphate (3.12 mmol kg-') for 1 week and subsequently leached with water, more
added sulphate was lost in the leachate of the soils previously incubated with either CaCO,
or phosphate.

INTRODUCTION
The movement of ions through a soil depends on, among other things, the surface-charge character-
istics of the soil (Wann & Uehara, 1978). The charge characteristics of variable charge soils, unlike
those with permanent change, can be manipulated by the addition of soil amendments, such as
phosphate fertilizers and liming materials (Sanchez & Salinas, 1981).
Addition of phosphate to soils has been shown to increase the adsorption ofcations such as CaZf
(Ryden & Syers, 1976) and K + (Wann & Uehara, 1978), and to decrease the adsorption of anions
such as SO:- (Ryden et al., 1987). Addition of lime, however, has a varying effect on the adsorption
of both anions and cations. An increase in pH due to liming increases the adsorption ofcations such
as Mgz+(Grove et al., 1981) and K + (Goedert et al., 1975), and decreases the adsorption of anions,
such as SO:- (Marsh et al., 1987). In contrast, an increase in CaZ+concentration due to liming can
decrease the adsorption of cations such as K+ (Edmeades, 1982) and increase the adsorption of
anions such as phosphate (Barrow, 1984). The net effect of liming on both cation and anion
adsorption depends on soil pH and on the amount of calcium added through liming.
Unlike phosphate, SO:- is weakly held at soil surfaces (Barrow, 1978) and is therefore more
susceptible to leaching in many soils. Liming both decreases the adsorption of SO:- (Marsh et al.,
1987) and increases the mineralization of organic sulphate (Korentager et ai., 1983). Similarly, PO:-
application appears to decrease SO:- adsorption, either by competing for adsorption sites or by
increasing the negative charge on the surface. Thus both lime and phosphate addition increase the
amount of SO:- in the soil solution. However, if the accumulated sulphate is not recovered by the
plant it is susceptible to leaching. Whereas the effect of liming and phosphate application on

*Present address: Department of Soil Science, The University, Newcastle upon Tyne NEl 7RU, UK.

493
494 N . S. Bolan et al.
sulphate adsorption has been investigated (Marsh et al., 1987), their effects on the subsequent
leaching of SO:- do not appear to have been examined in detail.
In this paper the effects of both liming and phosphate addition on the adsorption of sulphate
and its subsequent movement through soil are examined for two soils with contrasting sulphate
adsorption characteristics.

MATERIALS A N D METHODS
Soil
Two surface soils, Patua loam (a Hydric Dystrandept, with allophanic clay) and Tokomaru silt loam
(a Typic Fragiaqualf, with vermiculite), which vary widely in their origin, mineralogy, and anion
adsorption capacity (Bolan et al., 1986b), were used. The soils were sieved ( < 2 mm) and air-dried
before use.

Incubation
Varying amounts of either CaCO, or Ca(H,PO,),. H,O (Table 1) were mixed with air-dried soils and
distilled water was added to bring the gravimetric water content to 0.40 and 0.60 for the Tokomaru
and Patua soil, respectively;these water contents approximate 'field capacity'. The wet samples were
incubated in plastic bags at 20°C for 4 weeks. The moisture content was checked on alternate days
and water added if necessary. After incubation, the samples were air-dried, sieved (<2 mm) and
stored in air-tight plastic bags.

Sulphate adsorption
Sulphate adsorption was measured in 0.01 M NaCl solution containing varying amounts of SO:-
added as K,SO,. Two drops of chloroform were added to the samples which were shaken end-over-
end for 40 h at 20°C at a soi1:solution ratio of 1:8. At the end of the shaking period, suspension pH
was measured and the supernatant solution was separated by centrifugation and filtration. The

Table 1. Amounts (mmol kg-I) of CaCO, and Ca(H2P04)H20 added to soils prior to incu-
bation and values for pH, surface charge (kC kg-') and 0.01 M CaCI,-extractable S (mmol
kg-I) after incubation

Surface charge
Amount (mmol kg-') Extractable
added sulphate
Treatment (mmol kg-I) pH +ve -ve (mmol kg-')

Patua soil
CaCO, Lo - 4.7 11.9 53.9 0.33
L' 160 5.6 8.0 157.5 0.45
L2 320 6.4 3.9 236.2 0.58
LI 600 7.0 3.0 277.6 1 .02
Ca(H2P04)2H20 f' 26 4.7 9.9 72.5 0.28
p2 52 4.8 6.9 83.9 0.36
PI 104 5.0 2.0 96.3 0.39
Tokomaru soil
CaCO, Lo - 4.9 3.9 74.6 0.16
Ll 40 5.9 2.0 83.9 0.16
L2 80 6.5 1 .o 96.3 0.25
L3 160 6.8 NA 109.8 0.31
Ca(H,PO,),H,O PI 13 4.9 2.0 71.5 0.15
p2 26 5.1 NA 74.6 0.10
PI 52 5.0 NA 72.5 0.20

NA = negative adsorption of C1during charge measurements.

 Soil charge & sulphate leaching 495
concentration of SO:- was measured by the method of Johnson & Ulrich (1959) and the amount
adsorbed was estimated from the difference between the amount added and the amount remaining
in solution.

Surface charge
Surface charge was measured in 1.O M NaCl using the ion-retention method of Schofield (1949). A
soil sample (2 g) was shaken with 30 ml of electrolyte solution in a weighed centrifuge tube for
30 min, centrifuged, the solution decanted, and the soil retained for a further four washings with the
electrolyte. The final extract was used for chloride and sodium measurements to enable corrections
to be made for the entrained solution determined by weighing. Adsorbed Naf and CI- were then
extracted by shaking for 20 min with 5 x 20 ml 1.0 M NH,NO, solution. Chloride was determined
using the spectrophotometric method of Florence & Farrer (1971) and Na by atomic emission at
589 nm in the presence of 0.2% KNO, solution.

Breakthrough curves
PVC tubes, 300mm in length with an inner diameter of 51.2mm, were used for the column
experiments. A nylon mesh (60 pm holes) was sealed to the base of the column, minimising the dead-
end volume. Air-dried soil ( < 2 mm) was added in small increments into the columns, minimising
particle segregation. The columns were packed to a soil depth of 260 mm, giving bulk densities of
0.81 Mgrn-, for Patua loam and 1.05 Mgrn-, for Tokomaru silt loam. Breakthrough curves for the
Tokomaru soil were only measured for the samples incubated with the zero and highest rates of lime
or phosphate, since preliminary work indicated that the effects of both liming and phosphate
addition on the adsorption of sulphate were less than for the Patua soil.
Complete details of the column experiments were given in Bolan et al. (1986a). In brief, the soil
columns were initially saturated with water by leaching with 3 pore volumes of distilled water.
Sodium azide ( 0 . 0 0 1 5 ~ was
) added to all solutions to arrest microbial growth. A solution of
0.0031 M K,SO, (100 pg S ml-') (C,) labelled with 35S was next added to the columns in 50 ml
aliquots and the leachate was collected in 25 ml fractions. The effluent aliquots were analysed for 35S
(C,) by liquid scintillation counting. So long as the concentration of soil sulphate had been reduced
by preleaching with water, there was no difference in the breakthrough curves obtained by the
conventional analysis of 32Sor by the analysis of 35S(Bolan et ul.,1986a), and the use of ,'S simplified
considerably the analysis of sulphate.

Pulse experiments
In a separate experiment the movement of sulphate which had been incubated with soil was
examined for selected soil samples (Patua soil-Lo, L,, L2,L, and P,; Tokomaru soil-Lo,L, and P,).
Soils previously incubated with CaCO, and phosphate were further incubated with "S-K,SO,
solution at the rate of 3.12 mmol S kg-' soil for 1 week at field capacity. The soils were then air dried
and sieved to < 2 mm. Unincubated soils were packed to a depth of 200 mm in PVC columns, as
described above. The incubated soils were added to the top 60 mm of the columns and the soil
columns were saturated from the base. The columns were leached with 3 pore volumes (V,) of water
and the leachates were analysed for 35S.The columns were then cut into 20 mm sections and the soil
samples extracted with 0.4 M Ca(H,PO,), (Searle, 1979)and analysed for ,'S. In order to examine the
effect of leaching on the movement of calcium added as CaCO,, soil pH (1:2.5 soikwater) and the
amount of 1.O M KC1-extractable calcium were also measured for each section of columns of the
Patua soil incubated with varying amounts of lime.

Modelling the breakthrough curves

The experimental breakthrough curves were compared with those derived theoretically from the
convection-dispersion equation for miscible displacement of solutes during a steady flow through a
soil column (Lapidus & Amundson, 1952):
496 N . S. Bolan et al.
where Cis the concentration of the ionic species in solution (mol m-3), D is the dispersion coefficient
(m2SC'), v is the pore water velocity (m s-I), xis the distance ( m )in the direction of flow, t is the time
(s), p is the bulk density (kg rn-,), S, is the amount of the solute adsorbed per unit mass of the soil
(mol kg-') and 0 is the volumetric water content (m' m-').
If adsorption is instantaneous and completely reversible, as observed in a previous study (Bolan
et al., 1986a), and follows the Freundlich isotherm then:

S, = KCM

where K and Mare constants. Substitution of Equation ( 2 )in Equation (1) yields

In general, given a set of boundary conditions, only a numerical solution of Equation (3) is
possible. Here the numerical apporach of de Wit & van Kuelen (1975) was adopted, except that a
microcomputer and BASIC were used rather than a mainframe and Continuous System Modelling
Programme. There were two other differences. Firstly, a steady solute flux surface boundary con-
dition (CivO)was used. The justification for assuming a steady solute flux in the column experiments
was given by Bolan et al. (1986a). Secondly, for step-function experiments, flow in a column one
segment longer than that used experimentally was modelled, and the predicted C, in the break-
through curves was calculated as the mean of the soil solution concentration in the last two seg-
ments. This circumvented the problem of the model computing the midpoint concentration in a
segment, while C, is the concentration at the base of the column. For M = 1, that is a linear isotherm,
close agreement was found between the numerical solution and the analytical solution for a step-
function boundary condition (van Genuchten & Alves, 1982).

RESULTS A N D DISCUSSION
Sulphate adsorption isotherms
For both soils, the adsorption of SO:- decreased with increasing pH resulting from lime addition
(Fig. 1). A similar decrease in adsorption with increasing pH has been reported for anions, such as
PO:- (Barrow, 1984)and SO:- (Gebhardt & Coleman, 1974;Nodvin et al., 1986)and has frequently
been related to changes in surface charge with increasing pH (Marsh et al., 1987). Similarly, incu-
bation of soils with phosphate resulted in a decrease in the amount of SO:- adsorbed (Fig. 1).
Metson & Blakemore (1978) have suggested that for many weakly-weathered soils the addition of
PO:- results in a rapid desorption of adsorbed SO:- because of competition from the more strongly
adsorbed PO:- ion.
Adsorption isotherms for SO:- were adequately described by the Freundlich equation
(Equation (2)). While the addition of both CaCO, and phosphate decreased the value of K, only the
addition of phosphate caused a small increase in the value of M . From a simulation study for
competitive adsorption, Murali & Aylmore (1983) have suggested that the decrease in Kis related to
a decrease in the number of sites available for adsorption and the increase in M is related to a
decrease in the strength of adsorption. This suggests that the addition of CaCO, decreased SO:-
adsorption largely by decreasing the number of adsorption sites and that the addition of PO:-
decreased SO:- adsorption by decreasing both the number of sites available and the strength of
SO$- adsorption.

Surface charge
Incubation with CaCO, caused a small decrease in positive charge but a larger increase in negative
charge in the two soils (Table 1). This resulted in an increase in the net negative charge of both soils.
The increase in net negative charge with liming has often been related to an increase in pH (Haynes,
1985). Similar increases in net negative charge with increasing pH have been reported for variable-
charge soil components, such as hydrous iron and aluminium oxides (Hingston et al., 1972) and for
 Soil charge & su@hare leaching 497
Potuo

C (rnol rn-3)

Fig. 1. Sulphate adsorptionisothermsfor soils incubatedwith varying dressingsof lime (L) or phosphate(P). For
clarity, the adsorption isotherms for the Tokomaru soil Loand L, only are presented.

oxisols (van Raij & Peech, 1972). The effect of pH on surface charge has frequently been viewed as
the adsorption or desorption of H f or OH- at mineral surfaces (Breeuwsma & Lyklema, 1973).
Thus the surface charge will become increasingly negative as the pH (OH- activity) increases and
increasingly positive as the pH decreases. Similarly, an increase in negative charge due to the
dissociation with increasing pH of Hf from COOH groups will occur in soils with a large amount of
organic matter, such as the Patua loam (Wada & Okamura, 1980).
Incubation with PO:- decreased the positive charge of both soils, but increased negative charge
only in the case of the Patua loam. The negative adsorption of C1- in the case of the Tokomaru silt
loam at higher amounts of PO:- incubation indicates that only negligible amounts of positive
charge were present after PO:- addition. This resulted in an increase in the net negative charge of
both soils. A similar increase in net negative charge with the addition of PO:- has been observed
both for hydrous iron and aluminium oxides (Ryden & Syers, 1977) and for variable charge soils
(Schalscha eta]., 1974). It has been shown that PO:- is initially adsorbed onto the positive sites and
thereby decreases positive charge. With increased surface coverage, PO:- is adsorbed onto both
neutral and negative sites resulting in an increase in negative charge (Hingston et al., 1972; Ryden &
Syers, 1977).

Leaching of soil sulphur

The addition of both CaCO, and phosphate increased the concentration of 0.01 M CaC1,-extractable
sulphate (Table 1). The effect was more pronounced for the Patua soil than for the Tokomaru soil,
which may partly be related to the higher organic matter content of the former. Also the effect was
more pronounced with liming than with phosphate addition. Similar increases in extractable SO:-
with increasing pH have been observed in other studies (David et al., 1982; Probert, 1976) and
several reasons have been given for this effect (Freney & Stevenson, 1966; Korentager et al., 1983).
498 N . S. Bolan et al.
These include: (i) SO:- mineralized from soil organic matter by microorganisms growing in a more
favourable pH environment; (ii) SO:- released from organic matter by chemical hydrolysis; (iii)
adsorbed SO:- released from the soil surface; or (iv) SO:- released from sparingly soluble iron and
aluminium hydroxy sulphates which becomes more soluble at higher pH. Similarly, the increase in
CaC1,-extractable SO:- resulting from PO:- addition may be related to the desorption of the
adsorbed SO:-.

0.07-

0.06- A.
1.
L3 'L-
.-A---
A*\
: 0.05- .A
*
I .-A-.
E
- --A
0
\
-E
0.04 -
\*
.-
C
._
e ...... \
L, ...........................
e
$ o.03 @. ... \
0
....
0
-. .. \
v)
'._
- a... \
0.02
....
...
*... '.
\
-
0.01
*-*-*,
L1 \
A<.

o-I
*-+--*-
*-+-
m-----.- --a- --.---a---a---,__,__-~---~
Lo
*........_
..._
'*\

I I

Pore volumes
Fig. 2. Concentration of sulphate in the leachate during a preliminary leaching with distilled water of the Patua
soil incubated with varying amounts of lime.

0 2 4 6 0
Pore volumes
Fig. 3. Breakthrough curves for S for the soils incubated with varying rates of (a) lime or (b) P. The curve for P, in
the Patua soil is not presented. (points = measured; lines = calculated).
 Soil charge & sulphate leaching 499
Initial leaching of the soil columns with distilled water removed most of the SO:- accumulated
during incubation (Fig. 2). The initial concentration of SO$- in the effluent increased with increasing
quantities of CaCO, and phosphate addition. This effect was more pronounced with the Patua soil
than with the Tokomaru soil (data not presented). Korentager et al., (1983) have suggested that
liming induces the leaching of soil S because of an increased mineralization of organic sulphur.
Alternatively, this may be caused by decreased adsorption of SO:- with increasing pH (Probert,
1976). With increasing leaching, the concentration of S in the effluent decreased to a very low level
for all the soil columns.

Table 2. Physical data for the soil columns and the parameters used in Equation (3) to calculate the breakthrough
curves

P VP V e co D
Treatment (Mgm-’) (ml) (mms-I) (m3m-)) (molm-’) (mm*s-l) K M

Patua soil
LO 1.05 247 0.030 0.480 3.12 0.67 3.71 0.50
L, 1.05 256 0.029 0.497 3.12 0.67 2.01 0.56
L, 1.05 250 0.030 0.486 3.12 0.67 1.78 0.53
L3 1.05 249 0.028 0.484 3.12 0.67 1.36 0.52
PI 1.05 253 0.027 0.491 3.12 0.67 3.59 0.53
p, 1.05 256 0.025 0.497 3.12 0.67 3.18 0.57
p, 1.05 250 0.026 0.486 3.12 0.67 2.18 0.65
Tokomaru soil
LO 0.81 203 0.034 0.394 3.12 0.17 0.19 0.78
L3 0.81 209 0.033 0.406 3.12 0.17 0.16 0.76
p3 0.81 200 0.029 0.388 3.12 0.17 0.18 0.81

-
- ‘1
8

/
/
/
/
/
/
/

i_
0 //A A A 2 4 6 8

Fig. 4. Relationship between SO:- adsorption measured in column and batch experiments for Patua ( 0 )and
Tokornaru (A)soils which had been incubated with different amounts of lime or phosphate. The dotted line is
for a 1: 1 relationship.
10
 500 N . S. Bolan et al.
Percentage of added S
(a) Tokornaru
0 20 40 60 80 0 20 40 60 80 100 0 20 40 60 8 0

____I ----I

-E 80

---
E I20
f
2 160
I
240

Leachate
Lo L3 p3

( b ) Potua

Leachate I I I I I
Lo L, L2 L3 p3
Fig. 5. Distribution of phosphate-extractable )%O,within the soil column and in the leachate after leaching with
3 pore volumes of water. The dotted line represents the initial (prior to leaching) distribution of 35Sin the top
60 mm of the soil column.

Breakthrough curves
The breakthrough curves for various treatments are presented in Fig. 3. Added SO:- appeared
earlier in the effluent of the Tokomaru soil than of the Patua soil, consistent with the fact that the
latter adsorbs appreciably more SO:- than the former. With the addition of either CaCO, or
phosphate, applied SO:- appeared earlier in the effluent compared to the untreated soils. The effect
on the breakthrough curves of both CaCO, and phosphate addition was more pronounced with the
Patua soil than with the Tokomaru soil, as would be expected from the adsorption data (Fig. 1).
The breakthrough curves obtained for the various treatments were calculated, assuming instan-
taneous and reversible adsorption in Equation (3), and are also shown in Fig. 3. Physical data for the
soil columns and the parameters used to predict the breakthrough curves are given in Table 2.
Values of D were chosen which optimised the agreement between measured and simulated data of
the breakthrough curves for the untreated soil columns. Values of Mwere obtained from adsorption
isotherms and the values of K were calculated from Equation (2) using the value of M from the
isotherm and the C,,and &values from the column experiments. It has been observed that K, but not
M , is influenced by experimental conditions (Bolan et al., 1986a), and so it seems reasonable to use
the data for M , but not K, from the batch experiments in the column simulations.
Considering the wide range of breakthrough curves obtained there was generally good agree-
ment between measured and calculated data (Fig. 3). However, in the case of the Patua loam, the
calculated curves lagged behind the experimental data at the initial stages of the breakthrough,
except for the untreated column. Weaver et al. (1985) also obtained a good agreement between
experimental data and a simulated breakthrough curve using Equation (3) for an acid forest soil and
suggested that, for a limited range ofchemical changes, adsorption isotherms can be used to describe
the retention and movement of SO:- in soils.
 Soil charge & sulphate leaching 50 1
By measuring the area under the experimental breakthrough curves and subtracting Vp, to
account for SO:- in the soil solution, the amount of SO:- adsorbed can be calculated and this was
compared with that measured in the batch experiments (Fig. 4).Batch experiments gave a slightly
higher SO:- adsorption than the column experiments. In most other leaching studies (Fluhler et al.,
1982)adsorption has been found to be higher in batch experiments than in column experiments and
this has been related either to a greater contact of solute soil due to vigorous shaking in the batch
experiments, or to a difference in the ionic strength between the two systems.

Pulse experiment
The distribution of added SO:- in the column and in the leachate after leaching with 3 pore volumes
of water is shown in Fig. 5. In the case of the untreated soil, only 2% of the added SO:- was
recovered in the leachate from the Patua soil, whereas more than 60% of the added SO:- was
recovered in the leachate from the Tokomaru soil. For both soils the amount of SO:- recovered
from the soil column and in the leachate accounted for more than 98% of added SO$-.This suggests
that there was little or no net loss of sulphate by immobilization into soil organic matter.
When either CaCO, and phosphate were added, increased amounts of added SO:- were
recovered in the leachate. This was associated with the movement of increasing amounts of SO:- to
lower depths in the Patua soil. Because most of the SO:- was lost in the leachate from the Tokomaru

Labelled S (rnrnol kq-')

I 2 3

E
-E

f
a
U

Fig. 6. Observed distributionof phosphate-extractable"SO:- after leaching with 3 pore volumes of water in the
Patua soil incubated with lime ( A - - - A , La)and without (0-0, Lo).Correspondinglines without points
show predicted distribution.
502 N . S. Bolan et al.
PH Calcium [mmol kg-'1
0 10 20 30 40 50
I I I I I

- .';
I i
.b

Fig. 7. Distributionwith depth ofpH and 1 M KC1-extractableCa in Patua soil columns after leachingwith 3 pore
volumes of water. Soil incubated with different amounts of lime (LoW , L, +,
L, 0,L, A ) placed in 60 mm layer
on column.

soil, there was little difference in the distribution within the soil column. The distribution of SO:- in
columns of the Patua soil was predicted using Equation (3) and the predicted distribution was
compared with the observed one (Fig. 6). There was a discontinuity in the predicted distribution of
SO:- for the columns containing the incubated soil in the top 60 mm (incubated with either lime or
phosphate) caused by the heterogeneity of the two soil layers in relation to their SO:- adsorption
characteristics, as suggested by Selim et al. (1977). The observed distributions1are different from
those predicted. The model underestimated the amount of SO:- remaining in the incubated soil
occupying the top 60 mm of the soil column and also underestimated the depth to which the added
SO:- moved. Various reasons have been suggested for similar differences observed in a previous
study with unlimed soil (Bolan et al., 1986a). These include the non-uniformity of solute flow during
leaching, immobilization of added SO:- to organic S during incubation, or a change in the nature of
SO:- adsorption with time during incubation. The more extensive movement of SO:- in the CaC0,-
incubated soil in this study may also be due to a downward movement of CaCO, during leaching.
Data for the distribution of Ca within the column for the CaC0,-incubated Patua soil (Fig. 7)
suggest that there was a considerable movement of lime by leaching, which resulted in an increase in
soil pH at lower depths. This increase in pH would have decreased the adsorption capacity for SO:-
and enhanced its movement.

CONCLUSIONS
Differences between soils in the amount of SO:- lost by leaching have frequently been related to the
SO:- adsorption capacities of the soils (Chao et al., 1962; Metson & Blakemore, 1978), but there is
little direct evidence on this point in the literature. Comparison of leaching losses between the two
soils in the pulse experiment strongly suggests that the SO:- adsorption capacity of the soil is the
major factor determining the leaching losses of SO:-. The Patua soil, which contains short-
range order allophanic clay, adsorbed appreciably more SO:- than the vermicullite-dominated
Tokomaru soil and, as a consequence, the latter soil lost much more SO:- than the former.
 Soil charge & sulphate leaching 503
Addition of either CaCO, or phosphate decreased the adsorption of inorganic SO:- by increas-
ing the net negative charge of the soil, and increased the mineralization of organic S either by
increasing the microbial activity or by chemical hydrolysis. The combined effects of decreased
adsorption and increased mineralization produced an increase in CaCl, extractable sulphate. After
liming an acid soil, Elkins & Ensminger (1 971) observed an increase in SO:- concentration in the soil
solution with a corresponding increase in the S content of soybean plants grown in that soil.
Similarly, the addition of SO:- in the presence of PO:- (such as in superphosphate) increases the
amount of SO:- in the soil solution (Muller & McSweeny, 1974). Thus it has been suggested that
both liming and phosphate addition increases the availability of sulphate to plants. However, the
results from the present experiment suggest that in the absence of active uptake by plants, any
sulphate released into the soil solution by liming and phosphate addition is susceptible to leaching
and may be lost to the subsoil horizons. This is likely to be ofparticular significancein soils such as
the Tokomaru silt loam, which have a very low capacity to adsorb sulphate.

REFERENCES
BARROW, N.J. 1978. Inorganic reactions of phos- FLUHLER,
H., POLOMSKI,
J. & BLASER,P. 1982.
phorus, sulphur and molybdenum in soil. In Retention and movement of fluoride in soils.
Mineral Nutrition of Legumes in Tropical and Sub- Journal of Environmental Quality 11,461468.
tropical soils. (eds C. S. Andrew & J. Kamprath). FKENEY,J.R. & STEVENSON, F.J. 1966. Inorganic
pp. 19Ck206. CSIRO Australia. sulfur transformations in soils. Soil Science 101,
BARROW, N.J. 1984. Modelling the effects of pH on 307-3 16.
phosphate sorption by soils. Journal of Soil Science GEBHARDT, H. &COLEMAN, N.T. 1974. Anion adsorp-
35,283-297. tion by allophanic tropical soils: 11. Sulphate
BOLAN, N.S., SCOTTER, D.R., SYERS, J.K. & TILLMAN, adsorption. Soil Science Society of America Pro-
R.W. 1986a. The effect of adsorption on sulphate ceedings 38,259-262.
leaching. Soil Science Society of America Journal GOEDERT, W.J. COREY, R.B. & SYERS,J.K. 1975. Lime
50,1419-1424. effects on potassium equilibria in soils of Rio
BOLAN,N.S., SYERS,J.K. & TILLMAN, R.W. 1986b. Grande Do Sul, Brazil. Soil Science 120,107-1 11.
Ionic strength effects on surface charge and adsorp- GROVE, J.H. SUMNER, M.E. & SYERS,J.K. 1981. Effect
tion of phosphate and sulphate by soils. Journal of of lime on exchangeable magnesium in variable sur-
Soil Science 37,379-388. face charge soils. Soil Science Society of America
BREEUWSMA, A. & LYKLEMA, J. 1973. Physical and Journal 45,497-500.
chemical adsorption of ions in the electrical double HAYNES,R.J. 1985. Lime and phosphate in the soil-
layer on haematite (aFe,O,). Journal of Colloid and plant system. Advances in Agronomy 37,249-315.
Interface Science 43,437-448. HINGSTON, F.J. POSNER, A.M. & QUIRK,J.P. 1972.
CHAO,T.T., HARWARD, M.E. & FANG,S.C. 1962. Anion adsorption by goethite and gibbsite. I. The
Movement of 35Stagged sulphate through soil role of protons in determining the adsorption
columns. Soil Science Society of America Proceed- envelopes. Journal of Soil Science 23, 175-192.
ings 26,27-3 1. JOHNSON,C.M. & ULRICH,A. 1959. Analytical
DAVID,M.B., MITCHELL, M.J. & NAKAS,J.P. 1982. methods for use in plant analysis. California
Organic and inorganic sulfur constituents of a Agriculture Experimental Station Bulletin 776,
forest soil and their relationship to microbial 54-58.
activity. Soil Science Society of America Journal 46, KORENTAGER, L., BYRNES,H.B. & HELLUMS, T.D.
847-852. 1983. The effect of liming and leaching on sulfur
DE WIT,C.T. & VAN KEULEN, H. 1975. Simulation of supplying capacity of soils. Soil Science Society of
Transport Process in Soils. Centre for Agricultural America Journal 47,525-530.
Publishing and Documentation, Wageningen. LAPIDUS, L. & AMLJNDSON, N.R. 1952. Mathematics
EDMEADES, D.C. 1982. Effects of lime on effective of adsorption in beds. VI. The effect of longitudinal
cation exchange capacity and exchangeable cations diffusion in ion exchange and chromotographic
in a range of New Zealand soils. New Zealand columns. Journal of Physical Chemistry 56,
Journal of Agricultural Research 25,27-33. 98&988.
ELKINS,D.M. & ENSMINGER, L.E. 1971. Effect of MARSH,K.B., SYERS,J.K. & TILLMAN, R.W. 1987.
soil pH on the availability of adsorbed sulphate. Charge relationships of sulphate sorption in soils.
Soil Science Society of America Proceedings 35, Soil Science Society of American Journal 51,
931-934. 318-323.
FLORENCE, T.M. & FARRER, Y.J. 1971. Spectrophoto- METSON,A.J. & BLAKEMORE, L.C. 1978. Sulphate
metric determination of chloride at the parts-per- retention by New Zealand soils in relation to the
billion level by the mercury (11) thiocyanate competitive effect of phosphate. New Zealand
method. Analytica Chimica Acta 54,373-377. Journal of Agriculturai Research 21,243-253.
504 N . S. Bolun et ul.
MULLER,F.B. & MCSWEENY,G. 1974. Effect of SCHOFIELD, R.K. 1949. Effect ofpH on electric charges
granulation of superphosphate on movement of carried by clay particles. Journal of Soil Science 1,
phosphorus and sulfur in two soils. New Zealand 1-8.
Journal of Experimental Agriculture 2,53362. SEARLE,P.L. 1979. Measurement of adsorbed sul-
MURALI, V . & AYLMORE, L.A.G. 1983. Competitive phate in soils-effects of varying soi1:extractants
adsorption during solute transport in soils: 2. Simu- ratios and methods of measurement. New Zealand
lations o f competitive adsorption. Soil Science 135, Journal of Agricultural Research 22,287-290.
203-2 13. SELIM,H.M., DAVIDSON, J.M. & RAO, P.S.C. 1977.
NODVIN,S.C., DRISCOLL, C.T. & LIKENS,G.E. 1986. Transport of reactive solutes through multilayered
The effect of pH on sulphate adsorption by a forest soil. Soil Science Society of America Journal 41,
soil. Soil Science 142.69-75. 3-10.
PROBERT, M.E. 1976. Studies of ‘available’ and iso- VAN GENUCHTEN, W.T% & ALVES,W.J. 1982. Analyti-
topically exchangeable sulphur in some north cal solutions of the one-dimensional convective-
Queensland soils. Plant and Soil45,461475. dispersive solute transport equation. Agricultural
RYDEN,J.C. & SYERS,J.K. 1976. Calcium retention Research Service, USDA Technical Bulletin No.
in response to phosphate sorption by soils. Soil 1661.
Science Society of America Journal 40,845-846. VAN RAIJ, B. & PEECH, M. 1972. Electrochemical
RYDEN,J.C. & SYERS, J.K. 1977. Charge relationships properties of some oxisols and alfisols of the tropics.
of phosphate sorption. Nature 25551-53. Soil Science Society of America Proceedings 36,
RYDEN,J.C., SYERS,J.K. & TILLMAN, R.W. 1987. 587-593.
Inorganic anion sorption and interactions with WADA,K. & OKAMURA, Y. 1980. Electric charge
phosphate sorption by hydrous ferric oxide gel. characteristics of Ando A, horizon soils. Journal of
Journalof SailScience 38,211-217. Soil Science 31,307-314.
SANCHES, P.A. & SALINAS, J.G. 1981. Low-input tech- WANN,S.S. & UEHARA,G. 1978. Surface charge
nology for managing oxisols and ultisols in tropical manipulation in constant potential soil colloids: I.
America. Advances in Agronomy 34,279-306. Effect of solute transport. Soil Science Society of
SCHALSCHA, E.D., PRATT,P.F. & SOTO,D. 1974. America Journal 42,886-888.
Effect of phosphate adsorption on the cation WEAVER,G.T. KHANNA,P.K. & BEESE,F. 1985.
exchange capacity of volcanic ash soils. Soil Science Retention and transport of sulphate in a slightly
Society of America Proceedings 38,539-540. acid forest soil. Soil Science Society of America
Journal49,746750.

(Received 16 March 1987)