Effect of leaching on soil properties and lucerne growth following lime and gypsum amendments to a soil with an acid subsoil


Black, A.S.; Cameron, K.C.

New Zealand Journal of Agricultural Research 27(2): 195-200

1984


In pot trials in 1981-2, CaSO4 and Ca(OH)2 were applied at 0.86 or 2.58 g Ca2+/pot to A or B horizons of a Hororata very stony silt loam and either leached or left untreated before sowing lucerne cv. Rere in early Jan. Liming the A horizon had no effect on pH, exchangeable Al3+, the Al3+:Ca2+ ratio or root yield measured in mid-May in the B horizon. Liming the B horizon increased pH, reduced Al3+ and the Al3+:Ca2+ ratio and increased root growth in that layer. Effects of CaSO4 were small and root yields were unaffected. Shoot DM yields were significantly increased by soil amendment, the increase being significantly greater following application of Ca(OH)2 rather than CaSO4 to the B horizon.

New
Zealand
Journal
of
Agricultural
Research,
1984,
Vol.
27
:
195-200
195
0028-8233/84/2702-0195$2.50/0
©
Crown
copyright
1984
Effect
of
leaching
on
soil
properties
and
lucerne
growth
following
lime
and
gypsum
amendments
to
a
soil
with
an
acid
subsoil
A.
S.
BLACK
K.
C.
CAMERON
Department
of
Soil
Science
Lincoln
College,
Canterbury,
New
Zealand
Abstract
Amelioration
of
subsoil
acidity
was
studied
in
pots
containing
reconstructed
profiles.
Gypsum
or
lime
was
placed
in
the
A
or
B
horizon,
either
with
or
without
leaching,
before
growing
lucerne.
Liming'the
A
horizon
did
not
alter
the
pH,
exchangeable
Al',
Al
ratio,
or
root
yield
in
the
B
horizon.
The
high
lime
application
rate
gave
a
pH
of
7.8
in
the
A
horizon.
Lime
addition
to
the
B
horizon
increased
pH,
decreased
Al"
-
and
Al":
Ca'';
and
increased
root
growth
in
that
layer.
Although
gypsum
increased
exchangeable
Ca
1
',
the
pH
was
decreased,
and
Al"
-
:Ca"
only
decreased
slightly
in
the
B
horizon.
Gypsum
did
not
increase
B
horizon
root
yields,
with
or
without
leaching.
The
practical
implications
of
these
results
are
discussed.
Keywords
soil
chemistry;
soil
amendments;
lime;
gypsum;
leaching;
pH;
aluminium;
calcium;
lucerne;
root
growth
INTRODUCTION
Lucerne
grown
on
the
yellow-brown
stony
soils
associated
with
the
yellow-brown
and
yellow-grey
earths
of
the
upper
Canterbury
Plains
(Kear
et
al.
1967)
does
not
show
the
drought
resistance
of
lucerne
grown
on
similar
soils
lower
on
the
Plains.
Poor
subsoil
root
growth
in
the
upper
Plains'
soil
was
noted
in
the
field,
and
as
there
was
no
obvious
physical
barrier
it
was
thought
that
root
develop-
ment
was
limited
by
low
pH.
Lucerne
is
very
sensitive
to
pH
extremes
in
solu-
tion
culture
media
(Munns
1965;
Andrew
1978)
Received
25
October
1983;
revision
18
January
1984
especially
when
the
plants
depend
on
nitrogen
from
rhizobial
fixation
(Andrew
1978).
Similar
results
in
soils
may
be
related
to
AP'
and
Mn"
-
toxicities.
These
toxic
effects
can
be
reduced
by
several
pro-
cedures
including
liming
or,
as
found
in
solution
culture,
by
increasing
the
Ca"
concentration
(Sut-
ton
&
Hallsworth
1958;
Munns
1965).
Surface
applications
of
lime
have
little
effect
on
subsoil
pH
even
at
very
high
rates
(Reeve
&
Sum-
ner
1972).
The
lack
of
either
Ca''
movement
into
lower
horizons
or
a
pH
effect
may
be
related
to
the
low
solubility
of
lime
and/or
the
extra
cation
exchange
capacity
generated
by
liming,
which
would
retain
the
Ca''
against
leaching.
However,
Reeve
&
Sumner
(1972)
have
shown
that
gypsum
can
leach
into
lower
horizons
and
improve
the
Ca'
relative
to
the
Al''
status
of
the
soil.
Short-term
studies
(5
weeks)
by
Simpson
et
al.
(1979)
showed
that
mixing
gypsum
with
an
acid
subsoil
did
not
improve
root
development.
This
was
attributed
to
the
reduction
in
soil
pH
created
by
the
salt
addition.
Furthermore,
the
electrolyte
concentration
was
not
reduced
by
leaching
which
normally
occurs
in
the
field
during
the
winter
months.
The
objectives
of
the
present
study
were:
(1)
to
evaluate
the
hypothesis
that
on
a
yellow-brown
stony
soil
of
the
upper
Canterbury
Plains,
subsoil
acidity
restricted
root
development;
and
(2)
to
study
the
effectiveness
of
applying
gypsum
with
subse-
quent
leaching
as
a
method
for
improving
subsoil
root
development.
MATERIALS
AND
METHODS
Samples
of
a
Hororata
very
stony
silt
loam
(Kear
et
al.
1967)
(Dystrochrept;
Soil
Survey
Staff
1975)
were
collected
from
under
a
lucerne
stand
near
Oxford,
North
Canterbury
(grid
reference
NZMSI
S75/712875).
Soil
was
collected
from
4
locations
from
depth
intervals
of
0-100
mm
(top
of
the
A
horizon)
and
250-500
mm
(B
horizon).
After
air-
drying,
the
soil
was
passed
through
a
10
mm
screen
and
the
coarser
material
was
discarded.
Some
soil
properties
are
shown
in
Table
1.
Cylindrical
pots
(102
mm
diam.,
300
mm
high)
were
packed
with
1
910
g
of
air-dry
B
horizon
soil
196
New
Zealand
Journal
of
Agricultural
Research,
1984,
Vol.
27
followed
by
530
g
of
A
horizon
soil.
The
following
treatments
were
applied
to
the
pots:
2
Ca
sources
(Ca(OH),
and
CaSO
4
)
each
at
2
application
rates
(0.86
and
2.58
g
Ca"/pot);
2
horizon
placements
(the
Ca"
amendment
being
mixed
throughout
either
the
A
or
B
horizon
soil);
all
of
the
above
either
with
or
without
a
pre-
sowing
leaching
treatment.
In
the
leached
treat-
ments,
twice
the
volume
of
water
held
in
the
soil
against
drainage
was
passed
through
the
pots
over
7
days.
This
volume
was
an
approximation
of
the
leaching
volume
during
one
winter
season
in
the
field.
Unleached
pots
were
watered
up
to
the
same
weight
as
the
leached
pots
in
which
drainage
had
ceased.
The
electrical
conductivities
of
the
last
80
ml
of
leachate
in
the
gypsum
treatment
A
horizon
low
rate,
A
horizon
high
rate,
B
horizon
low
rate,
and
B
horizon
high
rate
were
0.9,
1.3,
0.3,
and
1.5
mS/cm
respectively.
In
all
lime
treatments
the
value
was
<
0.3
mS/cm.
The
treatments
were
combined
in
a
2
4
factorial
arrangement
with
leached
and
unleached
controls
(no
amendment).
The
design
was
replicated
3
times.
The
Ca"
treatments
were
applied
before
leaching,
but
all
pots
were
given
the
following
basal
fertiliser,
applied
after
leaching
at
the
following
rates
(mg/pot):
Ca(H,PO
4
),.H,0
=
249;
KCI
=
46.8;
MgSO
4
.711,0
=
251;
CuC1,.2H-,0
=
4.4;
ZnCI,
=
3.4;
(N11
4
)
6
Mo
7
0,
4
.4H,0
=
0.08;
Na,13
4
0
7
.10H,0
=3.7.
Six
seeds
of
Medicago
sativa
cv.
Rere
which
had
been
inoculated
with
Rhizobium
were
sown
in
each
pot
on
6
January
1982.
Each
pot
was
thinned
to
4
plants
following
establishment.
To
ensure
that
plants
used
water
from
the
B
horizons,
the
pots
were
allowed
to
dry
to
half
the
original
water
con-
tent
before
more
water
was
applied.
Tops
were
har-
vested
on
9
March
1982
and
15
May
1982,
and
the
weight
of
dry
matter
was
determined.
At
the
second
harvest,
roots
were
sieved
out
of
the
A
horizon
and
3
equal
depth
intervals
in
the
B
horizon
for
dry
matter
determination.
The
soil
collected
was
retained
for
analysis.
The
following
methods
of
soil
analysis
were
used:
pH
on
1:2.5
soil
water
extract;
exchangeable
Al"
with
IN
KCI
(McLean
1965),
exchangeable
bases
excluding
Na.
using
1N
Na-acetate
at
pH
8.5
to
reduce
dissolution
of
Ca
from
undissolved
lime
(Chapman
1965)
and
determined
by
atomic
absorption
spectrophotometry.
RESULTS
AND
DISCUSSION
Leaching
did
not
significantly
alter
the
soil
prop-
erties
or
plant
yields
in
the
control
pots,
therefore
only
the
mean
data
for
the
controls
are
reported.
Table
1
Some
properties
of
the
A
and
B
horizon
from
the
Hororata
soil
(
<
2
mm
fraction).
A
Horizon
B
Horizon
C.E.C.
-
pH7
(me%)
16.8
10.6
Exchangeable
cation&
(me%)
Ca'
9.0
1.5
Mg
,
.
0.74
0.30
K'
0.40
0.15
Na'
0.10
0.02
Al'•
0.16
2.39
H•
0.05
0.03
pH
5.80
5.40
Olsen
P
11
4
'
Ca'',
Mg'',
K
.
,
and
Na
with
1M
NH,
acetate
(pH
7),
and
Al'
and
H
with
1M
KCI.
Soil
pH
The
pH
of
5.37
in
the
B
horizon
of
the
control
(Table
2)
is
considerably
below
the
level
required
for
good
lucerne
growth
(Langer
1977).
Incorpo-
ration
of
lime
elevated
the
pH,
but
gypsum
addi-
tion
caused
a
significant
pH
decrease
relative
to
lime.
The
extent
of
the
pH
change
was
significantly
greater
when
the
materials
were
mixed
directly
into
a
horizon.
In
the
B
horizon
the
leaching
treatment
had
no
significant
effect.
There
were
several
significant
interactions
of
treatments
on
soil
pH
(Table
2).
The
data
for
the
significant
interaction
of
form,
rate,
and
placement
in
both
horizons
are
shown
in
Table
3.
Liming
the
A
horizon
at
the
high
rate
produced
a
pH
of
7.78
which
would
be
expected
to
depress
plant
growth.
Even
at
the
high
rate,
liming
the
A
horizon
did
not
alter
the
pH
of
the
B
horizon
compared
with
the
control,
suggesting
that
little
movement
of
lime
had
occurred.
However,
liming
of
the
B
horizon
increased
its
pH
to
5.73
and
6.50
at
the
low
and
high
rates
respectively.
Only
at
the
high
rate
was
a
pH
suitable
for
lucerne
root
growth
produced.
Application
of
gypsum
to
either
horizon
depressed
the
pH
compared
with
the
control,
regardless
of
where
the
gypsum
was
placed.
This
demonstrates
its
solubility
and
mobility
in
the
moisture
stream
during
routine
watering.
The
pH
decrease
resulted
from
an
increase
in
soluble
salts
which
displaced
Al":H`
into
the
soil
solution.
This
conclusion
is
supported
by
the
significant
form
of
leaching
interaction
which
occurred
in
both
hori-
zons
(Table
2).
For
example,
in
the
B
horizon
following
lime
addition,
the
pH
was
5.69
and
5.71
with
and
without
leaching
respectively.
For
gyp-
sum,
the
pH
values
were
4.88
and
4.74
respect-
ively.
Leaching
removed
much
of
the
dissolved
salt,
0
C
II
*
*
* :
Z:
M
000
*
N
W
C•;
WI
CZ)
00
00
Coi
*
* *
Z
CI)*
*
* * *
z:
ow,m-
Z*
*
Ta
ble
2
E
ffec
t
o
f
tre
a
tmen
ts
on
so
il
p
rop
e
r
t
ie
s
a
fte
r
t
he
fina
l
ha
rves
t.
;
'a
'
CO
U
co
al
OD
0
0
CO
0
0.
0
00
Black
&
Cameron-Effect
of
leaching
on
soil
properties
and
lucerne
growth
197
Table
3
pH
values
from
the
interaction
of
form,
rate,
and
placement
in
the
A
and
B,
horizons.
Form
Application
rate
Placement
horizon
Low
High
A
Horizon
Lime
A
B
6.76
5.87
7.78
5.95
Gypsum
A
5.35
5.23
B
5.67
5.20
SI
0.24
B,
Horizon
Lime
A
5.25
5.32
B
5.73
6.50
Gypsum
A
4.95
4.76
B
4.98
4.63
Si
0.12
N
N
N
N
*
*
co)*
(
Z
A
*
Z*
00
z
oo
O
0 *
C>C
3
06,
1
10
00000
0
*
ZZZZ
r-
gorDc?
oddo
*
Z*
**
co)
Z
*
* * *
*
*
*
*
'F.
so
0 1,
m
O
so
o
wiv
sci
e,
00
z
Tr
r-
W
0'
W
0
vi vi
Cr,
00
N
O'
,
Z;
0";
,
43
kr;
*
cf)
*
*
*
*
*
*
*
as
as
1:Val.rEt.
0
2
x
x
rx
a
o
cct_cgV,..
Exxxxx
wrx
a,
a
0.)
1.2.4.,
a,
4.,
4.,
2
a'Et
N
0
which
allowed
the
pH
to
rise
in
the
gypsum
treat-
0
.._.
g
ment.
The
finding
that
the
pH
values
were
low
c.3
especially
at
the
high
gypsum
rate
(Table
3)
relative
11
.,,,,
to
the
control
(Table
2)
suggested
that
the
excess
g
4
salt
was
not
removed.
This
was
supported
by
the
E
II
electrical
conductivity
measurements
reported
on
4.,
§
the
leachates
in
the
Materials
and
Methods
section.
4.r
E
In
no
treatment
combinations
with
gypsum
was
the
o
....
pH
in
the
B
horizon
suitable
for
lucerne
root
growth
.
o
cu
(i.e.,
approaching
pH
6).
o
E
N
....
.0
ttl
112
F_,
Exchangeable
A
1
3
+
I:0
'
..c
0
a
Only
very
low
levels
of
exchangeable
Al
3
+
were
.
5,
,....o
present
in
the
A
horizon
(Table
2).
In
the
B
hori-
0
al
o
7)
cu
zon,
however,
the
exchangeable
Al"
level
(2.5
me%)
. -
II
in
the
unamended
soil
would
be
sufficient
to
reduce
the
root
growth
of
certain
species
including
lucerne
1::
4
1:1
o
(Devine
et
al.
1976).
..
E
f,
Liming
reduced
the
AP+
levels
in
the
B
horizon
cu
co
whereas
gypsum
did
not
(Table
2).
The
application
T
-5
v
cit
S
of
amendments
to
the
B
horizon
depressed
=
II
exchangeable
Al"
compared
with
applications
to
o-
the
A
horizon.
Leaching
had
no
main
effect,
reflect-
ing
the
high
affinity
of
Al"
for
the
exchange
sites.
0
OA
=
.E
The
significant
form,
rate,
and
placement
inter-
n
1.1
action
on
AP*
in
the
B
2
horizon
is
shown
in
Table
P.'
.
1
.S
4.
The
only
treatment
which
substantially
reduced
,
t''
AP+
was
lime
added
to
the
B
horizon;
the
effect
o
c
O
.
'
0
increased
with
rate
of
application.
There
was
only
s
cq
slight
evidence
that
gypsum
reduced
Al"
in
this
o
0
x'
soil,
an
effect
reported
by
Reeve
&
Sumner
(1972).
r
ta
N
els
of
Al
3
+
to
levels
suitable
for
root
growth.
II
II
Only
liming
of
the
B
horizon
reduced
the
toxic
lev-
N
N
N
198
New
Zealand
Journal
of
Agricultural
Research,
1984,
Vol.
27
Table
4
Interaction
of
form,
rate,
and
placement
on
exchangeable
Al
Al":Ca"ratio,
and
Al"
saturation
percentage
in
the
B,
horizon.
Exchangeable
Al"
(me%)
AI"
:Ca"
ratio
Al"
saturation'
(%)
Placement
Form
horizon
Low
rate
High
rate
Lime
A
2.48
2.36
B
0.85
<
0.01
Gypsum
A
2.52
2.48
B
2.48
2.30
Low
rate
High
rate
Low
rate
High
rate
1.98
1.43
66
59
0.30
0.01
23
0.01
1.43
0.97
59
49
1.30
0.39
56
28
sx
0.06
0.11
100
Al"
Al"
-
-1-Ca"
Exchangeable
(plus
soluble)
Ca
2
+
Both
amendments
significantly
increased
the
Ca
2
+
status
of
both
horizons
(Table
2)
by
similar
amounts,
with
the
greater
increase
at
the
high
rate,
when
placed
in
the
particular
horizon
without
leaching.
The
significant
form,
placement,
and
leaching
interaction on
the
Ca
2
+
status
of
the
B,
horizon
is
shown
in
Table
5.
Leaching
had
little
effect
on
Ca
2
+
regardless
of
placement
of
lime,
reflecting
the
min-
imal
leaching
of
that
source.
Leaching
of
gypsum
placed
in
the
A
horizon
had
no
effect
on
the
Ca:
2
+
status
of
the
B2
horizon,
but
leaching
of
gypsum
applied
to
the
B
horizon
decreased
the
Ca
2
+
from
6.03
to
3.00
me%
because
of
the
removal
of
excess
dissolved
gypsum.
Where
the
excess
was
leached,
the
Ca
2
+
status
(3.00
me%)
was
only
about
half
that
of
the
limed
B
horizon
(5.63
me%).
These
data
sug-
gest
that,
in
the
field,
extra
Ca
2
+
from
gypsum
may
be
rapidly
leached
through
and
from
a
B
horizon
where
the
exchange
sites
are
dominated
by
Al
3
+
(Table
1).
Also
the
Ca
2
+
retained
from
gypsum
application
will
be
much
less
than
from
lime.
Fur-
thermore,
gypsum
application
to
the
surface
soil
will
not
effectively
alter
the
Ca
2
+
status
of
the
subsoil.
Al"
content
relative
to
other
cations
The
ratio
of
Al
3
+:Ca
2
+
extracted
from
the
B2
hori-
zon
is
shown
in
Table
2.
The
mean
of
treated
pots
was
less
than
the
control.
Both
Ca
2
+
sources
pro-
duced
similar
mean
ratios,
although
the
reduction
was
greatest
at
the
higher
rate
where
Ca
2
+
was
placed
in
the
B
horizon.
The
significant
form,
rate,
and
placement
inter-
action
on
Al
3
+:Ca
2
+
in
the
B
horizon
is
shown
in
Table
4.
Lime
or
gypsum
added
to
the
A
horizon
reduced
the
ratio,
and
the
effect
increased
with
increasing
rate.
However,
the
effect
was
greater
with
Table
5
Interaction
of
form,
placement,
and
leaching
on
the
Ca"
status
of
the
B,
horizon.
Form
Placement
horizon
Ca"
(me%)
Unleached
Leached
Lime
Gypsum
A
B
B
1.47
5.50
2.23
6.03
1.47
5.63
2.20
3.00
sic
0.34
placement
in
the
B
horizon,
and
lime
was
found
to
be
more
effective
than
gypsum.
Russell
(1978)
reported
that
for
white
clover
the
ratio
had
to
be
<
0.2
to
obtain
a
90%
relative
yield.
Helyar
&
Anderson
(1971)
suggested
that
the
Al
3
+
saturation
of
the
exchange
complex
had
to
be
<
20%
for
lucerne.
The
Hororata
B
horizon
contained
low
concentrations
of
K+
and
Mg
2
+
so
the
percentage
Al
3
+
saturation
can
be
estimated
by
(100
Al/Al+Ca),
as
shown
in
Table
4.
The
only
treatment
which
produced
such
a
low
ratio
or
saturation
was
lime
added
at
the
high
rate
to
the
B
horizon.
This
was
achieved
through
both
a
reduction
in
Al
3
+
and
an
increase
in
Ca
2
+.
At
the
low
lime
rate,
some
restriction
to
growth
would
still
be
expected
from
Al
3
+
toxicity.
Gypsum
did
not
achieve
an
A1
3
+:Ca
2
+
balance
suitable
for
plant
growth.
Dry
matter
yield
of
tops
The
dry
matter
yield
of
plant
tops
was
low,
con-
sidering
the
period
of
growth
(Table
6).
This
N
N
c0
it
N
00
oo
00
NN
ON
00,00,
*
Z
C1)
Z
r.O
N
N
r--
rsi
a•
o,
00
D
NS
rei
C/
as
a
a,
a
,
r••
e•
-
;
,
•0
as
a,
o
e•-;
c•i
N
Z
CA
Z
0000
0000
ZZZZ
et
-•
r•I
•—•
4444
X
00
00
*
Z Z Z
o
et
v1
N
tr)
r-
••,R
00
N
00
r--
DM
y
ie
ld
o
f
top
s
(g
/p
o
t)
*
*
Z
(1)
Z
*
00
Z
Z
CA CA
Z
00
CL
".
0
0
bx
.se,
Ta
b
le
6
E
ffe
c
t
o
f
trea
tmen
ts
on
top
a
n
d
ro
o
t
dry
ma
tter
y
ie
lds.
Black
&
Cameron—Effect
of
leaching
on
soil
properties
and
lucerne
growth
199
resulted
from
moisture
stress
which
developed
during
periods
when
the
plants
were
forced
to
use
water
stored
in
the
B
horizon.
The
total
yield
from
the
2
harvests
was
signifi-
cantly
greater
from
the
amended
treatments
com-
pared
to
the
control
(Table
6).
Lime
produced
a
significantly
higher
yield
than
gypsum
both
at
the
first
harvest
and
in
total.
There
were
no
other
sig-
nificant
main
effects.
There
was
a
significant
form
x
placement
inter-
action
(SR
=
0.15)
at
the
second
harvest
(Table
6).
The
yield
(g/pot)
was
similar
from
gypsum
(4.30)
and
lime
(4.22)
when
added
to
the
A
horizon,
but
lime
(4.50)
was
superior
to
gypsum
(4.01)
when
applied
to
the
B
horizon.
The
yields
from
pots
with
gypsum
added
to
the
B
horizon
were
similar
to
the
controls
(3.96).
These
results
clearly
demonstrate
the
beneficial
effects
of
lime
incorporation
into
the
B
horizon,
confirming
that
under
the
dry
condi-
tions
imposed
in
this
experiment,
the
subsoil
did
influence
lucerne
growth.
This
effect
is
unlikely
to
result
from
nutrient
supply
(as
all
pots
received
basal
nutrients,
so
the
results
reflect
the
greater
recovery
of
water
from
the
B
horizon
when
the
horizon
was
limed.
There
was
a
small
beneficial
effect
of
gypsum
application,
since
the
total
yield
(5.73
g/pot)
was
greater
than
in
the
control
(5.41
g/pot)
(Table
6).
This
trend
resulted
mainly
from
a
slight
improve-
ment
following
application
to
the
A
horizon.
The
cause
of
this
trend
was
related
to
greater
root
pro-
duction
in
the
A
horizon,
and
will
be
discussed
in
the
next
section.
Dry
matter
yield
of
roots
There
was
no
significant
difference
between
the
root
yields
in
the
control
and
the
mean
of
the
amended
pots
in
either
the
A
or
B
horizons
(Table
6).
In
the
A
horizon,
gypsum
produced
significantly
more
root
dry
matter
than
lime.
The
high
pH
following
lim-
ing
(7.28)
appears
to
have
been
more
detrimental
than
the
low
pH
(5.28)
after
gypsum
application.
The
greater
root
growth
in
the
A
horizon
was
reflected
in
greater
top
yields.
No
other
main
effects
or
interactions
were
evident
in
A
horizon
dry
matter
yields.
Root
yields
in
the
B
horizon
were
greater
when
the
amendments
were
added
to
the
B
horizon
(3.12
g/pot)
compared
to
the
A
horizon
amendment
(2.67
g/pot)
(Table
6),
but
the
main
effects
of
Ca"
form,
rate,
and
leaching
regime
were
not
significant.
There
was
a
significant
interaction
(SFc
=
0.23
g/pot)
of
form
and
placement
on
root
yields
in
the
B
horizon
(Table
6).
Placement
of
lime
in
the
B
horizon
rather
than
in
the
A
horizon
produced
an
increase
in
root
growth
from
2.55
to
3.55
g/pot
200
New
Zealand
Journal
of
Agricultural
Research,
1984,
Vol.
27
respectively,
whereas
the
gypsum
amendment
pro-
duced
similar
yields
regardless
of
placement
(2.90
and
2.79
g/pot
respectively).
This
is
consistent
with
the
data
in
Table
5,
where
only
liming
of
the
B
horizon
reduced
the
acidity
problem,
i.e.,
Al"
toxicity.
There
was
a
significant
interaction
(sx =
0.23
g/pot)
of
Ca"
placement
and
leaching
on
B
hori-
zon
root
yields
(Table
6).
When
the
A
horizon
was
amended,
leaching
caused
a
slight
increase
in
yield
from
2.47
to
2.77
g/pot.
In
contrast,
leaching
decreased
the
root
yields
from
3.40
to
2.85
g/pot
when
the
B
horizon
was
amended,
probably
because
the
Ca"
source
was
removed
through
leaching.
The
effect
was
principally
related
to
leaching
of
Ca"
from
gypsum
rather
than
lime
because
of
the
sol-
ubility
of
the
former.
This
is
consistent
with
the
exchangeable
Ca"
data
already
discussed.
Any
beneficial
effects
of
gypsum
were
short-lived.
CONCLUSIONS
Acidity
of
the
B
horizon
of
the
Hororata
soil
causes
poor
lucerne
growth.
Lime
added
directly
to
the
B
horizon
of
the
pots
improved
the
pH
and
exchangeable
Ca"
levels
and
reduced
the
exchangeable
A1
3
'
levels.
Heavy
liming
of
the
A
horizon
did
little
to
improve
the
subsoil
conditions
and
created
problems
in
the
A
horizon.
Therefore,
heavy
liming
of
surface
soil
is
not
a
practical
solu-
tion
to
the
problem.
At
establishment
of
a
lucerne
crop,
incorporation
of
lime
deeper
into
the
soil
would
improve
root
growth.
Gypsum
decreased
the
pH
and
did
not
decrease
exchangeable
Al"
levels.
Although
gypsum
did
increase
the
exchangable
ce+
levels
and
reduced
the
Al
3
+:Ca
2
+
ratio,
the
reduction
was
not
sufficient
to
minimise
the
effect
of
Al"
toxicity.
Much
of
the
Ca"
was
removed
by
leaching,
indicating
the
lack
of
displacement
of
Al"
by
Ca"
from
the
exchange
sites
in
an
acid
subsoil.
It
is
not
practical
to
use
higher
rates
of
gypsum
to
achieve
exchange,
as
the
high
rate
used
in
this
study
was
equivalent
to
13.7
t/ha.
ACKNOWLEDGMENTS
The
authors
thank
Mr
N.
P.
Smith
for
competent
tech-
nical
assistance,
and
the
Lincoln
College
Research
Fund
for
financial
support.
We
also
thank
Mr
M.
L.
Wensley
from
whose
property
the
soil
was
collected.
REFERENCES
Andrew,
C.
S.
1978:
Mineral
characterisation
of
tropical
forage
legumes.
In:
Andrew,
C.
S.;
Kamprath,
E.
J.
ed.,
Mineral
nutrition
of
legumes
in
tropical
and
subtropical
soils.
Melbourne.
C.S.I.R.O.
pp.
93-
111.
Chapman,
H.
D.
1965:
Cation-exchange
capacity.
In:
Black,
C.
A.
et
al.
ed.,
Methods
of
soil
analysis.
Madison.
American
Society
of
Agronomy.
pp.
899-
900.
Devine,
T.
E;
Foy,
G.
D.;
Fleming,
A.
L.;
Hanson,
C.
H.;
Campbell,
T.
A.;
McMurtrey,
J.
E.;
Schwartz,
J.
W.
1976:
Development
of
alfalfa
strains
with
dif-
ferential
tolerance
to
aluminium
toxicity.
Plant
and
soil
44
:
73-79.
Helyar,
K.
R.;
Anderson,
A.
J.
1971:
Effects
of
lime
on
growth
of
five
species,
on
aluminium
toxicity
and
on
phosphate
availability.
Australian
journal
of
agricultural
research
22
:
707-721.
Kear,
B.
S.;
Gibbs,
H.
S.;
Miller,
R.
B.
1967:
Soils
of
the
Downs
and
Plains
of
Canterbury
and
North
Otago,
New
Zealand.
New
Zealand
Department
of
Sci-
entific
and
Industrial
Research
Soil
Bureau
bulletin
14.
Langer,
R.
H.
M.
1977:
Lucerne.
In:
Langer,
R.
H.
M.
ed.,
Pastures
and
pasture
plants.
Wellington,
Reed
Ltd.
pp.
346-363.
McLean,
E.
0.
1965:
Aluminium.
In:
Black,
C.
A.
et
al.
ed.,
Methods
of
soil
analysis.
Madison,
American
Society
of
Agronomy.
pp.
985-988.
Munns,
D.
N.
1965:
Soil
acidity
and
growth
of
a
legume.
1.
Interactions
of
lime
with
nitrogen
and
phos-
phate
on
growth
of
Medicago
sativa
L.
and
Tri-
folium
subterraneum
L.
Australian
journal
of
agricultural
research
16
:
733-741.
Sutton,
C.
D.;
Hallsworth,
E.
G.
1958:
Studies
on
the
nutrition
of
forage
legumes.
1.
The
toxicity
of
the
low
pH
and
high
manganese
supply
to
lucerne
as
affected
by
climatic
factors
and
calcium
supply.
Plant
and
soil
9
:
305-517.
Simpson,
J.
R.;
Pinkerton,
A.;
Lazdovskis,
J.
1979:
Inter-
acting
effects
of
subsoil
acidity
and
water
on
the
root
behaviour
and
shoot
growth
of
some
geno-
types
of
lucerne
(Medicago
saliva
L.).
Australian
journal
of
agricultural
research
30:
609-619.
Soil
Survey
Staff.
1975:
Soil
taxonomy.
Washington
Soil
conservation
Service,
U.S.
Department
of
Agri-
culture.
Agriculture
handbook
number
436.
Reeve,
N.
G.;
Sumner,
M.
E.
1972:
Amelioration
of
sub-
soil
acidity
in
Natal
exisols
by
leaching
of
surface
applied
amendments.
Agrochemophysica
4:
1-6.
Russell,
J.
S.
1978:
Soil
factors
affecting
the
growth
of
legumes
on
low
fertility
soils
in
the
tropics
and
sub-tropics.
In:
Andrew,
C.
S.;
Kamprath,
E.
J.
ed.,
Mineral
nutrition
of
legumes
in
tropical
and
sub-
tropical
soils.
Melbourne.
C.S.I.R.O.
pp.
75-92.