meat bone meal and fox manure as p sources for ryegrass (lolium multiflorum) grown on a limed soil
TRANSCRIPT
RESEARCH ARTICLE
Meat bone meal and fox manure as P sources for ryegrass(Lolium multiflorum) grown on a limed soil
Kari Ylivainio Æ Risto Uusitalo Æ Eila Turtola
Received: 1 August 2007 / Accepted: 5 December 2007 / Published online: 20 December 2007
� Springer Science+Business Media B.V. 2007
Abstract Phosphorus (P)-rich by-products, such as
meat and bone meal (MBM) and fur animal manures,
are potential P sources in plant production systems.
However, the solubility of P and its availability to
plants in these forms has not been evaluated. We
characterized P solubility in MBM, fox manures
(FoxM) and dairy manure (DairyM) by Hedley
fractionation and assessed P availability for ryegrass
in a pot experiment. Up to 81% of P was water-soluble
in DairyM, but only about 3 and 5–28% was soluble in
MBM and FoxM products, respectively. Of the P in
MBM and FoxM, 90 and 65–89%, respectively, was
soluble only in 1 M HCl. Most of the P was inorganic;
DairyM contained the highest share (14%) of organic
P. Based on ryegrass yields and P uptake in a 3-year
pot experiment with three P levels (25, 50 and
100 mg kg-1), P availability was equal in the DairyM
and superphosphate (SP) treatments. Compared with
the availability of P in DairyM and SP, 19 and 35–
54% of the P in MBM and FoxM, respectively, was
immediately available to the plant; for the 3-year
period with ten ryegrass cuts, the respective P
availabilities increased to 63 and 69–87%. Additions
of the sparingly soluble P sources MBM and FoxM
increased the acid-soluble P concentrations in the
experimental soil, with MBM having the strongest
effect. However, the acid-soluble P fraction decreased
with time. Although the immediate bioavailability of
P in sparingly soluble P sources was lower than that in
DairyM and SP, our results suggest that their use as a
long-term P supply for perennial plants in non-
calcareous soils should be encouraged.
Keywords Dairy manure � Fox manure �Meat and bone meal � P availability �P fractionation � Ryegrass
Introduction
As a result of bovine spongiform encephalopathy
(BSE), commonly known as ‘‘mad cow disease’’, the
EU banned the use of meat and bone meal (MBM) as
a feed for ruminants in 1994 and, subsequently, as a
feed for all livestock in 2001. In addition, EC
Regulation No. 1774/2002 of the European Parlia-
ment and the Council further restricted the use of
MBM as a fertilizer in 2002. Alternative methods for
handling MBM, such as incineration and cement
production, have been explored; however, in the
spring of 2006, restrictions on MBM usage as a
fertilizer were lifted in all regions of the EU.
The EU produces about 3 million tons of MBM
annually. MBM contains about 5% phosphorus (P),
mostly as calcium phosphate. However, due to its
chemical nature, the P in MBM is classified as being
K. Ylivainio (&) � R. Uusitalo � E. Turtola
MTT Agrifood Research Finland, 31600 Jokioinen,
Finland
e-mail: [email protected]
123
Nutr Cycl Agroecosyst (2008) 81:267–278
DOI 10.1007/s10705-007-9162-y
sparingly soluble. Another similar type of high-P
by-product in Finland is fur animal manure, which is
related to MBM because much of the P in fur animal
diets derives from MBM. Due to their high P
concentrations, both MBM and fur animal manure
are potential P sources for agricultural systems,
especially for organic farms specialized in plant
production. The annual manure P produced by
Finnish fur animal farms amounts to about 2 million
kilograms, whereas the amount of P associated with
MBM, eligible for fertilizer use, is about half of that.
If used solely as fertilizer, P derived from MBM and
fur animal manure could, respectively, contribute
about 0.5 and 1.0 kg of total P per cultivated ha in
Finland. In addition to these sparingly soluble P
sources, cattle manure comprises the most significant
source of manure-based P in Finland, with an average
annual input of 5.6 kg P per cultivated ha. These
numbers can be compared with P input, in the form of
mineral fertilizers, of 11–15 kg ha-1 (MMM 2004;
Antikainen et al. 2005).
In addition to P, MBM contains a considerable
amount of N (8%). The availability of this N to cereals
has been found to be close to that of mineral fertilizers
(Jeng et al. 2004), whereas the availability of P in
MBM is largely unexplored. In Norway, Jeng et al.
(2006) found that the relative efficiency of P in MBM
was about 50% of that of mineral P fertilizer for the
first crop of barley and ryegrass and that MBM had
residual P effects the following year. Fur animal
manure, in turn, has a long history as a P fertilizer in
Finland, and 40% of the P in fur animal manure is
considered to be available for plants. Hence, when fur
animal manure is used as a P source in plant
production, surplus application rates of total P are
common. In order to ensure environmentally and
economically sustainable methods of using these
sparingly soluble P sources in plant production
systems, the availability of P for plants in these
products needs to be further clarified. The necessity
for such studies is supported by the results of a recent
study showing that long-term surplus P applications,
as in the case of fur animal manure, had increased
soluble P fractions in soils down to a depth of 60 cm
(Uusitalo et al. 2007), strongly suggesting that P
derived from this source may be partly converted into
a water-soluble form and leach into lower soil profiles.
In this study, we examined the solubility of P in
MBM, fox manure processed to various extents
(FoxM) and dairy manure (DairyM). We also deter-
mined the effect of all three P sources on the growth
and P uptake of ryegrass in a P-deficient soil in a
3-year pot experiment. The availability of P in these
products was compared with a soluble P source,
superphosphate (SP). The aim of the study was to
provide a scientific foundation for the use of MBM,
FoxM and DairyM as P fertilizers. Ryegrass was
chosen as the test plant because of its efficient dry
matter production and P uptake (Brink et al. 2001).
Material and methods
FoxM was obtained from a commercial composting
facility (Natural Compost, Kaustinen, Finland) and
used as received or processed to produce composted
FoxM (cFoxM) or pelletized cFoxM (pcFoxM). To
produce cFoxM, FoxM was mixed with peat (1/1,
v/v) and placed in a batch composter in which air was
injected from the bottom. The mixture reached 70�C
within 24 h and was maintained at this temperature
for another 24 h, after which the compost was dried
with suction. During drying, the moisture content
decreased from 70 to 50%. The total duration of the
composting phase was 1 week. During pelletizing, the
cFoxM was dried with hot air (325�C) and com-
pressed through a 6-mm sieve, producing pellets
about 1 cm in length.
Honkajoki Oy (Honkajoki), the main facility for
handling animal carcasses in Finland, provided MBM
for the experiment. DairyM was taken from a manure
heap from MTTs barn (Agrifood Research Finland,
Jokioinen, Finland), where it had been composted
with peat and stored for a few months on a paved
plate.
Manure analyses
Total P concentrations of the manures and MBM
were analyzed with an inductively coupled plasma–
atomic emission spectrometer (ICP-AES; Thermo
Jarrel Ash, Franklin, MA) after microwave digestion
with aqua-reqia-HF (Lamothe et al. 1986). Total N
and C concentrations were analyzed with on a CN-
2000 analyzer (LECO, St. Joseph, MI).
The solubility of P was assessed using the Hedley
fractionation scheme as modified by Sharpley and
268 Nutr Cycl Agroecosyst (2008) 81:267–278
123
Moyer (2000). In brief, air-dried samples of manure
were finely ground using a mortar, and 1-g subsam-
ples were extracted, twice with water at a 1:60 (w/v)
ratio, then once with each of the following sequen-
tially: 0.5 M NaHCO3, 0.1 M NaOH and 1 M HCl.
Extraction times were 16 h, except for the first water
extraction, which was 4 h. Following extraction, the
samples were centrifuged (3000 g, 15 min), and
inorganic P (Pi) [supernatants filtered through a 0.2-
lm nucleopore membrane (Whatman, Maidstone,
UK)] and total P (unfiltered supernatant digested at
120�C with sulfuric acid and peroxodisulfate) con-
centrations in the supernatant were determined using
molybdate colorimetry (Murphy and Riley 1962).
The difference between total P and Pi in the
supernatant was assumed to represent organic P (Po).
Soil analyses
The soil used as growth medium was retrieved from
the plough layer of a Histic Podzol [as tentatively
classified according to the FAO (1998) system],
which had a low content of plant-available P,
according to Finnish agronomic criteria (Table 1).
Soil texture was analyzed with a pipette method, as
described by Elonen (1971), C concentration was
analyzed with the LECO CN-2000 analyzer, and soil
pH was determined in a 1:2.5 (v/v) water suspension.
Plant available P, Ca, K and Mg concentrations were
analyzed according to the Finnish agronomic soil
testing protocol [acid ammonium acetate (AAAc);
Vuorinen and Makitie 1955], and P content was also
determined with 0.5 M NaHCO3 (Olsen-P). In addi-
tion, soil P was analyzed with a modified Hedley
fractionation scheme as described above.
Growth experiment
Ryegrass (Lolium multiflorum var. Turgo) was grown
under a glass roof outdoors at ambient air tempera-
ture on sandy soil for 3 consecutive years. Before
establishing the experiment, the soil was air-dried and
passed through a 6-mm sieve to remove coarse
fragments and root debris. The soil (6.5 kg) was then
limed with an amount of CaCO3 (6.85 g) that was
calculated to bring the pH to 6.5.
Phosphorus sources are listed in Table 2. Phos-
phorus application rates were 25, 50 and 100 mg kg-1
soil, corresponding to 162.5, 325 and 650 mg pot-1,
respectively. The P sources were applied only once, at
the beginning of the experiment, and mixed into the
whole soil volume. All the P sources, except SP and
MBM, were air dried and passed through a 6-mm
sieve before application. The control treatment did not
receive P. Treatments were replicated four times.
In the first year, all of the pots received the
following amounts of nutrients during sowing (mg
pot-1): 1000 N (as NH4NO3 and KNO3), 1000 K
(KNO3), 200 Mg (MgSO4), 20 Na (NaCl), 10 Fe
(FeSO4), 10 Zn (ZnSO4), 10 Mn (MnSO4), 5 Cu
(CuSO4), 1 B (H3BO3) and 1 Mo (Na2MoO4). In the
following years, the same amounts of nutrients were
applied during the sowing, with the exception that the
amounts of applied N and K were 1500 mg pot-1.Table 1 Characteristics of the soil used in the pot experiment
Soil texture (%)
0.2–2 mm 18
0.02–0.2 mm 67
0.002–0.02 mm 10
\0.002 mm 5
Total C (%) 2.1
Soil pH (water, 1:2.5, w/v) 5.9
Olsen P (mg kg-1 soil) 18.6
AAAc extractable nutrientsa (mg l-1)
P 3.2
Ca 815
K 169
Mg 57
a Acid ammonium acetate, pH 4.65
Table 2 Total phosphorus (P), nitrogen (N), calcium (Ca) and
carbon (C) concentrations and dry weight (DW) percentage
content of the air-dried amendments used in the pot experiment
P sourcea P
(mg g-1)
N
(mg g-1)
Ca
(mg g-1)
C
(mg g-1)
DW
(%)
MBM 62.7 71.2 141.5 338.0 96.5
FoxM 34.3 34.1 60.1 276.4 91.2
cFoxM 25.7 34.1 48.9 321.9 92.1
pcFoxM 24.1 31.6 44.2 278.4 91.0
DairyM 3.8 20.5 8.8 382.6 89.7
SP 90.8 – 200.9 7.1 97.9
a MBM, Meat and bone meal;; FoxM, fox manure; cFoxM,
composted FoxM; cpFoxM, pelletized cFoxM; DairyM, dairy
manure; SP, superphosphate (soluble P source)
Nutr Cycl Agroecosyst (2008) 81:267–278 269
123
Ryegrass was sown in the same pots three
consecutive springs, with a total of about 50 germi-
nating seeds per pot, and irrigated with deionized
water. The ryegrass was cut four times in the first
2 years, but only twice in the third year, due to poor
growth. Each year the final cuttings were carried out
after growth had ceased due to low temperature. Each
ryegrass stand was cut 2 cm above the soil surface,
washed briefly with detergent (0.1% Deconex), rinsed
three times with deionized water, blotted dry and
dried at 65�C in a forced-draught oven and ground
with a hammer mill. The P concentration in plant
material was analyzed after HNO3 (about 7 M)
digestion, with ICP-AES, and N was analyzed using
the Kjeldall method. In-house reference material was
included in the analyses.
After each cut during the growth period, supple-
mental doses of N, K and Mg were applied to all pots
to ensure their sufficiency for the following crop. In
the first year, the supplements per pot included
800 mg of N and K and 100 mg of Mg. In the second
year, 1500 mg of N and K and 100 mg of Mg were
applied per pot after the two first cuts, and 500 mg of
N and K and 50 mg of Mg was added after the third
cut. In the third year, 1000 mg of N, 500 mg of K and
50 mg of Mg were applied after the first cut. For the
second and the third year, the supplemental doses of
K and Mg were determined according to the AAAc
extractable concentrations of K and Mg in soil
samples taken from the pots at the end of the
previous years (after the fourth and eight ryegrass
cuts; data not shown). Additional liming was done
before the seeding in the third year due to a decline in
soil pH. Soils with a pH lower than 5.6 (the lowest
value was 5.1, obtained for the control treatment)
received 10 g of CaCO3, while the other soils
received 7 g. After the experiment was terminated,
soil pH varied from 6.0 (control) to 6.5 (cFoxM,
50 mg P kg-1).
After the last cut in each year, soil sample cores
were taken from each pot through the entire soil
column (about 0.5 l pot-1). The soil in each pot was
then sliced horizontally through the middle and the
top half turned upside down on top of the bottom half.
The soils were stored over winter at ambient air
temperature, covered with plastic under a glass roof.
In the spring, the soils were passed through a 6-mm
sieve and the ryegrass roots were placed at the bottom
of the pots.
To estimate P availabilities from different P
sources, data on ryegrass yields and P uptake for
the SP and DairyM treatments were used to solve the
following equation: y = A + B*(1-e- Cx), where A is
the minimum ryegrass yield or P uptake, B is the
maximum yield or P uptake response, C is a
coefficient and x the amount of applied P (mg kg-1
soil). This equation was then used with the data for
sparingly soluble P treatments to calculate the
amount of SP- or DairyM-based P required to
produce the corresponding yield or P uptake. These
values were then made proportional to the amount of
P applied.
Results and discussion
Phosphorus content and solubility of P sources
Of the by-products studied here, MBM had the
highest and DairyM the lowest total P concentration
(Table 2). The composting of FoxM decreased the
total P concentration due to the addition of peat prior
to composting, while pelletizing the compost had no
further effect on total P concentration (Table 2).
Phosphorus solubility varied widely between the
different P sources (Table 3). In DairyM, 81% of the
sum of the P fractions was water-soluble compared to
87% in SP; in contrast, water-soluble P was a minor
fraction (3–28%) in MBM and fox manure products,
with 65–90% of the P being in acid-soluble form
only. A high share of acid-soluble P in the FoxM is in
line with the diet of foxes, which at most may derive
30% of its P content from MBM; Baltic herring as
well as other dietary sources also contain P of low
solubility.
Composting of FoxM had no influence on P
solubility, but pelletizing cFoxM decreased both the
content of water-soluble P (Table 3) and total N
concentration (Table 2). During pelletizing, the com-
post was dried with hot air (325�C), which may alter
P solubility. The results are in agreement with the
study of O’Connor et al. (2004), who found that the
pelletizing of biosolids decreased water-soluble P
fractions and increased NH3 volatilization.
Most of the P in the studied by-products was
inorganic. Only in DairyM was a notable proportion
(14%) of P present as Po (Table 3), and most of it
was water- and NaHCO3-extractable, summing to
270 Nutr Cycl Agroecosyst (2008) 81:267–278
123
about 12% of the total P fractions. This results agrees
with that reported by Sharpley and Moyer (2000) who
found Po to be 14% of the total P extracted in water-
and NaHCO3-fractions from dairy manure.
Differences in P solubility in manures are related
to the diet and age of the animals, the bedding
material and the handling and storage of the manure.
However, the composting of FoxM had no influence
on P solubility, indicating that bone-derived P was
not affected by composting. Similar results were
obtained with poultry manure by Dao et al. (2001),
who argued that the solubility of dicalcium phos-
phate, added as dietary mineral supplements, is not
affected by composting. The DairyM in the present
study was mixed with peat and stored on a paved
plate for a few months. While the pile was rather
dense, the composting was probably minimal and
hence the manure can be regarded as stockpiled
manure.
Ryegrass yields in the pot experiment
All P sources increased the total sum of the ten
ryegrass yields significantly compared to the control
treatment (Table 4) with MBM having the least effect
(25 mg P kg soil-1, 27%) and DairyM have the
largest effect (100 mg P kg soil-1, 120%). Both
cFoxM and FoxM increased ryegrass yields equally,
and pcFoxM gave significantly lower yields than
FoxM only in the first and fourth cuts. Therefore, we
present yield data only for cFoxM.
Among the invidual cuttings, the most significant
differences in ryegrass yields were obtained in the
first cut, where DairyM gave the highest yields in all
P application levels, followed by SP [ cFoxM [MBM (Table 4). Thereafter, yield differences leveled
off. The lower yields of the SP treatment compared to
DairyM treatment were partly related to N availabil-
ity. All treatments received the same base level of
inorganic N, but during the first year it may have been
too low for maximum ryegrass growth, especially at
the highest P level (SP treatment), whereas organic
amendments provided N as well. This is supported by
the fact that among all the treatments in the first cut,
the lowest level of N was found in the 100 mg SP–
P kg-1 (19.1 mg N g-1 DW) treatment, which was
significantly lower than that in the 25 mg SP–P kg-1
(25.2 mg N g-1 DW) treatment. The same phenom-
enon was evident with other P sources as well, but
was least pronounced with DairyM (24.6 and
24.5 mg N g-1 DW in 25 and 100 mg DairyM–
P kg-1 treatments, respectively). Depressed N avail-
ability in SP treatments resulted in equal total N
uptakes for the 25 and 100 mg SP–P kg-1 treatments
(739 vs. 747 mg pot-1, respectively), whereas the P
uptake almost doubled (Table 5).
In the second experimental year, there was an
overall decline in the yields, which decreased more in
the two lower P application levels (25 and 50 mg P
kg-1 soil) than in the 100 mg P kg-1 treatments. At
the highest P application level, the yield depression
was most evident in the third year, indicating a
decrease in plant available P. This was supported by
the prolonged time to reach a given growth stage.
Most of the differences between the ryegrass
yields in the first 2 years were observed in the first
cuts of the growing seasons (Table 4, first and fifth
cuts). For example, with the highest P application
levels of MBM and DairyM, yield differences in cuts
Table 3 Inorganic (-i) and organic (-o) P concentrations in air-dried P sources according to the Hedley fractionation scheme
MBM (mg g-1) FoxM (mg g-1) cFoxM (mg g-1) pcFoxM (mg g-1) DairyM (mg g-1) SP (mg g-1)
Pw-i 1.0 6.9 5.6 0.5 3.2 84.8
Pw-o 0.8 1.5 1.4 0.7 0.3 –
PNaHCO3-i 2.3 1.7 0.8 0.7 0.2 0.6
PNaHCO3-o 0.1 0.1 0.3 0.2 0.2 –
PNaOH-i 0.4 0.3 0.2 0.4 0.1 5.3
PNaOH-o – 0.3 0.3 0.2 0.1 –
PHCl-i 51.3 19.2 17.5 22.6 0.2 7.1
RP-i 55.9 28.0 24.2 24.2 3.7 97.8
RP-o 0.9 1.7 1.9 1.1 0.6 –
Nutr Cycl Agroecosyst (2008) 81:267–278 271
123
Ta
ble
4D
rym
atte
ry
ield
(g)
of
ryeg
rass
inth
eth
ree
con
secu
tiv
ey
ears
Tre
atm
ent
(mg
Pk
g-
1)
Fir
sty
ear
Sec
on
dy
ear
Th
ird
yea
rR
cuts
1–10
1st
cut
2n
dcu
t3
rdcu
t4
thcu
tR
cuts
1–4
5th
cut
6th
cut
7th
cut
8th
cut
Rcuts
5–8
9th
cut
10
thcu
tR
cuts
9–10
Co
ntr
ol
18
.3a
18
.6a
13
.7a
19
.4a
70
.0a
9.1
a1
5.9
a1
3.2
a4
.8a
42
.9a
17
.0a
12
.3a
29
.3a
14
2.2
a
MB
M
25
21
.6a
21
.6b
c1
5.3
ab2
3.0
ab8
1.5
ab1
3.2
a1
9.9
ab1
8.6
ab7
.3b
c5
9.0
abcd
24
.8ab
16
.7ab
cd4
1.5
abc
18
2.0
b
50
24
.3b
c2
4.4
abcd
17
.6ab
c2
7.8
bcd
94
.1b
cd1
9.7
ab2
5.1
bc
26
.3cd
8.9
cde
80
.0cd
e2
4.4
ab1
6.9
abcd
41
.3ab
c2
15
.4b
cd
10
02
6.7
cd2
7.3
bcd
ef2
1.8
cde
29
.9cd
e1
07
.7cd
e1
6.7
a2
8.8
cd3
1.5
de
11
.5f
88
.4ef
26
.5ab
20
.7cd
e4
8.3
c2
41
.3d
e
cFo
xM
25
25
.4c
23
.0ab
c1
5.8
ab2
4.2
abc
88
.3ab
c1
3.0
a1
8.9
ab1
6.5
ab6
.5ab
54
.8ab
24
.2ab
17
.8b
cde
42
.0ab
c1
85
.2b
50
29
.2d
26
.1ab
cde
18
.3ab
c2
7.5
bcd
10
1.0
cde
19
.5ab
25
.1b
c2
5.5
cd8
.8cd
e7
8.9
de
22
.5ab
19
.6cd
e4
2.1
abc
22
2.0
cd
10
03
4.5
e2
9.4
cdef
22
.1cd
e3
0.0
cde
11
6.0
efg
29
.1b
c3
2.3
ef3
3.7
e9
.6d
e1
04
.7fg
27
.1ab
21
.9d
e4
9.0
c2
69
.7ef
Dai
ryM
25
33
.5e
28
.8b
cdef
20
.0b
cd2
5.4
bcd
10
7.8
def
15
.9a
20
.1ab
16
.2ab
6.5
ab5
8.8
abc
17
.7a
13
.7ab
31
.4ab
19
8.0
bc
50
40
.2f
30
.8d
ef2
2.6
efg
h3
1.2
de
12
4.8
fgh
19
.0ab
24
.4b
c2
1.6
bc
8.0
bcd
73
.0b
cde
30
.7b
15
.8ab
c4
6.4
bc
24
4.2
de
10
04
7.7
g3
4.1
f2
6.1
e3
5.0
e1
42
.8h
43
.5d
36
.4f
36
.0e
9.3
de
12
5.1
h2
7.1
ab1
8.6
bcd
e4
5.6
bc
31
3.6
g
SP
25
29
.3d
26
.5b
cdef
19
.3b
c2
5.0
abc
10
0.2
cde
11
.6a
20
.7ab
18
.0ab
6.8
b5
7.1
ab2
3.8
ab1
9.0
bcd
e4
2.7
abc
20
0.0
bc
50
34
.4e
29
.2cd
ef2
2.3
cde
27
.9b
cd1
13
.8ef
g1
8.3
a2
4.6
bc
21
.7b
c7
.4b
c7
2.0
bcd
e3
0.0
b1
9.2
cde
49
.2c
23
5.0
d
10
03
9.2
f3
2.2
ef2
4.5
de
31
.3d
e1
27
.2g
h3
8.5
cd3
7.3
f3
4.3
e1
0.4
ef1
20
.4g
h3
0.6
b2
2.5
e5
3.0
c3
00
.7fg
Wit
hin
aco
lum
n,
val
ues
foll
ow
edb
yth
esa
me
lett
erd
on
ot
dif
fer
sig
nifi
can
tly
(P[
0.0
5,
Tu
key
’ste
st)
272 Nutr Cycl Agroecosyst (2008) 81:267–278
123
Ta
ble
5T
ota
lP
(mg
po
t-1)
up
tak
eb
yry
egra
ssin
the
thre
eco
nse
cuti
ve
yea
rs
Tre
atm
ent
(mg
Pk
g-
1)
Fir
sty
ear
Sec
on
dy
ear
Th
ird
yea
rR
cuts
1–10
1st
cut
2n
dcu
t3
rdcu
t4
thcu
tR
cuts
1–4
5th
cut
6th
cut
7th
cut
8th
cut
Rcuts
5–8
9th
cut
10
thcu
tR
cuts
9–10
Co
ntr
ol
38
.8a
33
.5a
27
.2a
33
.2a
13
2.8
a1
3.0
a2
4.3
a1
5.5
a8
.9a
61
.6a
16
.4a
15
.0a
31
.5a
22
5.8
a
MB
M
25
46
.8ab
46
.5ab
33
.7ab
c4
1.2
abc
16
8.2
b1
6.3
a3
3.6
abc
26
.7b
cd1
7.5
bcd
94
.1b
cd2
4.7
abc
24
.2b
cd4
8.9
bc
31
1.2
b
50
51
.8b
cd5
6.3
bcd
e4
1.3
cde
51
.6d
e2
01
.0cd
23
.3ab
44
.9cd
e3
8.0
e2
4.1
ef1
30
.3e
24
.2ab
c2
6.5
bcd
ef5
0.8
bc
38
2.0
c
10
04
6.5
ab6
4.6
def
50
.8fg
58
.1ef
22
0.1
de
20
.4ab
54
.0ef
53
.3f
36
.4h
16
4.2
f2
7.3
abc
32
.1d
ef5
5.7
cd4
39
.9d
cFo
xM
25
47
.1ab
50
.7b
c3
4.0
abc
43
.4b
cd1
75
.2b
c1
6.7
a3
0.3
ab2
5.5
bc
16
.4b
c8
9.0
ab2
2.6
abc
25
.9b
cde
48
.5b
c3
12
.7b
50
47
.4ab
63
.6cd
ef4
2.5
de
52
.1d
e2
05
.6d
24
.4ab
42
.3cd
36
.5e
23
.7ef
12
6.8
e2
1.3
ab2
8.5
cdef
49
.8b
c3
82
.2c
10
06
1.2
cde
78
.7g
h5
6.9
g6
6.8
fg2
63
.6fg
33
.0b
c6
0.4
f5
6.1
f2
9.5
g1
79
.0f
28
.4ab
c3
2.9
ef6
1.3
cd5
04
.0e
Dai
ryM
25
48
.9ab
c6
4.7
def
36
.5b
cd4
1.5
abc
19
1.6
cd1
7.9
a3
0.4
ab2
2.8
ab1
5.8
b8
7.0
ab1
7.9
ab1
9.5
ab3
7.4
ab3
16
.0b
50
65
.9e
71
.0fg
47
.3ef
51
.7d
e2
35
.9ef
23
.2ab
40
.1b
cd3
2.2
cde
22
.1d
ef1
17
.6cd
e2
7.7
abc
20
.9ab
c4
8.6
bc
40
2.1
cd
10
09
6.0
g8
5.4
h6
5.7
h7
2.8
g3
19
.8h
43
.9c
63
.3f
53
.3f
26
.3fg
18
6.7
f2
7.6
abc
27
.8cd
ef5
5.3
cd5
61
.8f
SP
25
44
.5ab
52
.1b
cd3
3.6
ab3
7.8
ab1
68
.0b
14
.7a
34
.6ab
cd2
5.0
bc
16
.6b
c9
0.9
bc
24
.2ab
c2
4.5
bcd
48
.7b
c3
07
.6b
50
63
.3d
e6
6.1
efg
42
.6d
e4
7.9
cd2
19
.8d
e2
2.0
ab4
5.3
de
34
.5d
e2
1.0
cde
12
2.8
de
30
.0b
c2
8.3
cdef
58
.4cd
40
1.0
cd
10
08
0.4
f8
5.5
h5
5.9
g6
1.9
f2
83
.6g
40
.1c
65
.4f
54
.6f
30
.9g
19
1.1
f3
3.7
c3
4.3
f6
8.0
d5
42
.8ef
Val
ues
inco
lum
ns
wit
hth
esa
me
lett
erd
on
ot
dif
fer
sig
nifi
can
tly
(P[
0.0
5,
Tu
key
’ste
st)
Nutr Cycl Agroecosyst (2008) 81:267–278 273
123
number one and five accounted for 65% of the total
yield difference during the first two years. Compared
with the SP and DairyM treatments, however, cFoxM
and MBM treatments increased ryegrass yields as the
growing season progressed, especially in the second
year (Table 4).
The importance of P availability at the onset of
growth was evident when comparing the sparingly
soluble P sources with SP and DairyM in the first
ryegrass cuts of the first and second growing seasons.
At the lowest P application level, DairyM increased
ryegrass yields the most in the first cut (Table 4),
even though total P uptake was at the same level as
for the other P sources (Table 5). For example,
treatment with DairyM produced a 55% higher
ryegrass yield than treatment with MBM with almost
the same total P uptake. Total P uptake increased in
DairyM and SP treatments with increased P applica-
tion level, whereas it remained about the same at the
two lowest P application levels for the cFoxM and
MBM treatments (Table 5). However, ryegrass yields
increased significantly as the P application level of
MBM and cFoxM increased from 25 to 50 mg kg-1 soil
(Table 4). This suggests that as the P application level
increased, ryegrass P needs were better met at the
beginning of the growth period and the P enhanced dry
matter production. It has been suggested that early P
acquisition is critical for producing optimal yields, while
later application of P does not improve growth (Grant
et al. 2001). However, from the second cut onwards,
sparingly soluble P sources clearly increased both the P
uptake and the ryegrass yield. This result indicates that
the subsequent growth potential of ryegrass was not
depressed by the lower availability of P in the MBM and
cFoxM during the initial growth stage, as evidenced by
the increased ryegrass yields in the later cuts. Similar
results were obtained by Baker et al. (1989) with bone
meal, who observed that corn yield and P uptake
increased in consecutive growth periods and that from
the second harvest onwards, yield and P uptake were
equal to those obtained with Ca(H2PO4)2.
Ryegrass has a high root–shoot ratio and is
referred to as a P-efficient species with relatively
low external P requirements (Fohse et al. 1988).
Phosphorus uptake by ryegrass has been shown to
increase if the availability of P increases after a
period of deprivation, probably due to a deprivation-
induced increase in the number of P uptake sites in
the roots (Breeze et al. 1985). In addition, ryegrass
growth is enhanced by the presence of apatite in acid
soil (Junge and Werner 1989), and Chien and Menon
(1995) suggested that apatite is a suitable P source for
plants with a long growth period and a low external P
requirement. Therefore, MBM and the processed fox
manures may be convenient storage P fertilizers for
plants with the above characteristics. This would
reduce the need for surface application of P, especially
with perennial crops, and ultimately abate the problem
of P transport by runoff (Turtola and Yli-Halla 1999).
Availability of P in the different P sources
When DairyM was mixed in soil, P availability was
equal to that of SP, whether it was estimated
according to ryegrass yields or by P uptake. This is
in line with the high proportion of water-soluble P in
DairyM. In addition, manure may depress soil P
sorption strength (Holford et al. 1997) as manure-
based organic acids may compete for the same
binding sites as P (Øgaard 1996; Haynes and
Mokolobate 2001). This may have been the reason
that the highest water-extractable P concentration
was obtained for the 100 mg P kg-1 DairyM treat-
ment at the end of the experiment (Table 6), despite
the fact that this treatment gave the highest level of P
uptake. DairyM and SP can be both considered as
reference P sources when evaluating the P availabil-
ity in the sparingly soluble P sources, because they
had the same level of P availability. In addition, N
and C effects on P acquisition are within the range of
that in SP and DairyM. The application levels of P
correlated well with ryegrass yields in the SP and
DairyM treatments for the 3-year period (R2 [ 0.99).
Phosphorus availability from sparingly soluble
sources, when judged by yields, increased as the
experiment proceeded. According to the data from
the very first ryegrass cut, P availability was 19, 35,
43 and 54% in the MBM, pcFoxM, cFoxM and FoxM
treatments, respectively, where the availability of P in
the sparingly soluble P sources was taken as an
average of the three P application levels. However, in
the fourth cut, the respective values were 65, 50, 71
and 100%. For the sum of the ten ryegrass cuts, the
respective P availabilities were 63, 69, 74 and 87%.
When judging P availability based on total P uptake,
higher values were obtained (data not shown). This is
probably related to the low acquisition of P during the
274 Nutr Cycl Agroecosyst (2008) 81:267–278
123
early development of the ryegrass, which depressed
ryegrass growth.
Enhancement of ryegrass growth due to N derived
from the P amendments was most evident in the first
ryegrass cut. At the highest P application level, N
uptake in the DairyM treatment was 420 mg more
(total N uptake 1168 mg pot-1) than in the SP
treatment (total N uptake 747 mg pot-1). This prob-
ably stimulated both growth and P uptake, as shown by
Belanger et al. (2002). For the other organic amend-
ments, N derived from MBM was mostly in a plant-
available form (C:N ratio 4.7:1), as evidenced by the
approximately 250 mg greater N uptake from MBM
(100 mg P kg-1) than from SP (100 mg P kg-1) dur-
ing the first year. This difference in N uptake in the first
year accounted for 33% of the total N content in the
applied MBM. Moreover, the N uptake for MBM was
the second highest (after the DairyM treatment) in the
first ryegrass cut, even though MBM increased
ryegrass yield the least. This result is in line with
observations that N in MBM was equivalent to at least
80% of the N in mineral fertilizer (Jeng et al. 2004).
In the three succeeding years, P concentration in
the ryegrass stand varied between 1.46–2.69, 1.01–
1.88 and 0.90–1.12 mg g-1 dry weight (DW), respec-
tively, excluding the eighth and tenth cuts (data not
shown, values can be calculated by dividing the
values in Table 5 with the values in Table 4). Due to
the cessation of growth caused by low temperature,
the eighth and tenth ryegrass cuts were done at an
earlier growth stage, and thus P concentrations were
higher, varying between 1.87–3.18 and 1.21–1.57 mg
g-1 DW, respectively. Phosphorus concentrations of
ryegrass did not correlate with the respective yields
during the experiment (R2 = 0.03), indicating that P
concentration was a poor indicator of the physiolog-
ically available P content, due to the dilution effect
caused by enhanced growth.
In this study, ryegrass growth became depressed
before P concentration decreased below 1 mg g-1
DW, which is the concentration believed to indicate a
severe P deficiency (Yli-Halla 1991). The N: P ratio
of the crop may be a better indicator of the P status of
plants than P concentration. According to Gusewell
(2004), biomass production is depressed by P short-
age when the N: P ratio is[20. In this study, the N: P
ratio was up to 41 in the ninth cut (data not shown),
indicating a severe P deprivation.
Solubility of phosphorus in the experimental soil
After ten ryegrass cuts, those P fractions showing the
greatest decrease were the water- and bicarbonate-
Table 6 Hedley fractionation scheme for experimental soils after first year (four cuts) and third year (ten cuts) in 100 mg P kg-1
treatments
Pwater PNaHCO3 PNaOH PHCl RPi and Po
Pi Po Pi Po Pi Po Pi
Beginning of the exp. 4.4 7.8 37.1 30.0 105.9 112.5 176.5 474.2
Treatment
After 1st year
Control 2.3 a 7.7 ab 29.1 abc 42.7 d 96.7 b 108.8 a 173.1 ab 460.4 ab
MBM 5.4 bc 12.6 c 38.3 de 31.3 ab 103.0 bc 125.3 b 245.0 d 560.9 ef
cFoxM 5.7 c 12.4 c 42.4 ef 38.6 cd 118.0 de 140.7 cd 215.3 cd 572.9 f
DairyM 6.0 c 19.1 d 37.9 cde 32.8 abc 111.1 cd 122.0 ab 173.6 ab 502.5 bcd
SP 6.4 c 10.2 bc 47.6 f 38.0 bcd 129.8 e 127.9 bc 175.2 ab 535.2 cdef
After 3rd year
Control 0.9 a 6.4 a 21.8 a 29.6 a 79.0 a 121.1 ab 166.5 ab 425.3 a
MBM 2.4 a 10.5 bc 26.2 ab 28.8 a 92.7 ab 127.0 bc 228.0 cd 515.6 bcdef
cfoxM 2.3 a 10.5 bc 25.1 ab 32.6 abc 112.7 cd 119.8 ab 207.2 bc 510.2 bcde
DairyM 2.9 ab 12.7 c 25.1 ab 32.7 abc 89.2 ab 145.9 d 173.7 ab 482.3 abc
SP 2.0 a 11.7 c 30.7 bcd 34.8 abc 121.0 de 133.9 bcd 214.9 cd 549.1 def
Columns with the same letter do not differ significantly (P [ 0.05, Tukey’s test)
Nutr Cycl Agroecosyst (2008) 81:267–278 275
123
extractable Pi fractions – 80 and 41%, respectively –
in the control treatment (no P added) (Table 6).
However, these two fractions represented only 8% of
the sum of the Pi fractions. Quantitatively, the NaOH-
extractable Pi fraction decreased the most, by
26.9 mg kg-1 (25% decrease). Only the acid-soluble
P fraction was unaltered in the control treatment. As
for the P-amended soils, the highest P application
level (100 mg P kg-1) affected the concentrations of
the different P fractions most and, therefore, only the
results of 100 mg P kg-1 treatments are presented
(Table 6). In addition, the P solubility in soils after
the addition of the different fox manures (FoxM,
cFoxM, pcFoxM) was rather similar, and only the
results for cFoxM are presented.
After the first growth period, the water- and
NaHCO3-extractable Pi concentrations in the soils
were significantly higher due to P applications, but
thereafter the influence of the treatments on these
fractions was less evident. Throughout the treatment
period, the NaHCO3- and NaOH- extractable Pi
concentrations remained significantly higher only in
the SP treatments, and the NaOH-extractable Pi
concentrations remained significantly higher only in
the cFoxM treatment (Table 6). The acid-soluble Pi
fraction increased the most in the MBM and cFoxM
treatment. This is in line with the high share of acid-
soluble P in these P sources. During the two first
years, DairyM and SP had no influence on the acid-
soluble Pi fraction. During the last year, however, this
fraction increased somewhat in the MBM, cFoxM
and SP treatments. This may be due to the liming
with CaCO3 before the last seeding. A similar
increase in the acid-soluble P fractions after liming
was detected by Guo et al. (2000), who also found
that the Ca–P formed was not stable in highly
weathered soil.
Phosphorus in MBM is mostly in the form of
calcium phosphate, and its solubility is partly
governed by soil pH. After the first growth period,
about 70% of the added P in the MBM treatment
(100 mg P kg-1) was still in acid-soluble form. In
the succeeding years, the amount of this P fraction
decreased, but after 3 years about 60% of the added P
still remained in acid-soluble form. This result shows
that in our non-calcareous soils, MBM-derived P
gradually converted to a more soluble form. A recent
study has shown that a long history of fur animal
manure application on non-calcareous soils increased
soluble P fractions up to a depth of 60 cm (Uusitalo
et al 2007), suggesting that P derived from fur animal
manure gradually becomes water-soluble and leaches
down into the soil profile.
Of the P amendments used, DairyM increased
water-extractable Po concentrations of the soil the
most. After the first growth period, the water-
extractable Po concentrations in the DairyM treat-
ments increased from an initial value of 7.8 mg kg-1
to final values of 11.6 mg kg-1 (25 mg P kg-1) and
19.1 mg kg-1 (100 mg P kg-1). These values were
significantly higher than that found in the control
treatment (7.7 mg kg-1). The increased concentra-
tion of water-soluble Po with the DairyM treatment
(100 mg P kg-1) corresponded to the amount of
water + bicarbonate-extractable Po contents in the
added DairyM. In the succeeding years, the content of
water-soluble Po in the experimental soil declined,
indicating mineralization.
Variations in the NaHCO3- and NaOH-extractable
Po concentrations during the experiment were prob-
ably related to both the characteristics of the P
amendments and the amount of organic acids pro-
duced by ryegrass roots due to P deficiency and their
influence on microbial activity in the soil. While the
amount of C applied was the highest in the DairyM
treatment (Table 2), the mineralization of organic
matter may explain the steady increase in NaOH
extractable Po to its maximal level at the end of the
experiment in DairyM treatment. Other possible C
sources for soil microorganisms are organic acids,
which are excreted by roots after a period of P
deficiency (Jones 1998) and may increase the growth
of microorganisms and enhance microbial activity in
the rhizosphere (Toal et al. 2000). Increased micro-
bial activity in the rhizosphere may also increase Po in
the NaHCO3- (Helal and Sauerbeck 1984; Chen et al.
2002) and NaOH-extractable fractions (Zoysa et al.
1997, 1999). According to Guo et al. (2000),
NaHCO3–Po starts to contribute to the plant-available
P pool after the soil-available Pi pool is exhausted. In
the control treatment, there were indications of these
events, with an increase of NaHCO3–Po from 30 mg
kg-1 to 42.7 mg kg-1 during the first growth period,
and a subsequent decline to 32.1 mg kg-1 after the
second growth period.
Sparingly soluble P sources maintained the AAAc-
extractable P concentration at elevated levels
throughout the experiment (Table 7). However, lower
276 Nutr Cycl Agroecosyst (2008) 81:267–278
123
ryegrass yields in the MBM and cFoxM treatments
compared to the DairyM and SP treatments demon-
strate that the concentrations of AAAc-extractable P
were not in agreement with the plant availability of
P in soils supplemented with these sparingly soluble
P sources. The water-extractable P concentrations
were also lower in both sparingly soluble P sources
(Table 3) and in soils amended with these products,
although the difference was not significant when
compared with the respective values of the DairyM
treatment (Table 6). Overestimation of the concen-
tration of readily soluble P by the AAAc extraction in
soils amended with these sparingly soluble P sources
was probably due to the dissolution of calcium
phosphates by AAAc (pH 4.65). In the study of Jeng
et al. (2006), the amount of readily available P in
MBM was estimated to be 33–40% of the total P
when the estimation was based on the extraction of
MBM with ammonium lactate (pH 3.75). These
results suggest that extracting calcium phosphate
compounds, or soils amended with these compounds,
does not reflect their immediate P availability for
plants but rather gives an idea of their long-term
availability. In the long term, the concentrations of
water-extractable P may also increase in soils
amended with sparingly soluble P sources when the
amount of P added continuously exceeds the crop P
uptake (Uusitalo et al. 2007).
Conclusion
This study showed that MBM and fox manure may
represent good P storage fertilizers for plants which
have a long growing period and efficient P uptake
systems. The application of such fertilizers would
reduce the need for surface application of P for
perennial grasses, subsequently reducing P runoff.
Moreover, MBM and fox manure could be used
instead of inorganic fertilizers to improve the recy-
cling of P, and in organic farms lacking animal
production and their own manure source. Although
most P in MBM and fox manure was initially in an
acid-soluble form, P eventually converted to a plant-
available form and sustained ryegrass yields compa-
rable to those achieved with DairyM or SP. However,
plant growth as well as nutrient uptake is more
intense in pot experiments than in the field. There-
fore, field studies are underway to further evaluate the
P availability from these sparingly soluble P sources.
Acknowledgements We thank research assistant Pirkko
Maki and laboratory technician Anja Lehtonen for their
skillful technical assistance with the pot experiment and the
related analyses. The Ministry of Agriculture and Forestry,
the Finnish Fur Breeders’ Association and Honkajoki Oy are
gratefully acknowledged for funding.
References
Antikainen R, Lemola R, Nousiainen J, Sokka L, Esala M,
Huhtanen P, Rekolainen S (2005) Stocks and flows of
nitrogen and phosphorus in the Finnish food production
and consumption system. Agric Ecosyst Environ
107:287–305
Baker AM, Trimm JR, Sikora FJ (1989) Availability of phos-
phorus in bone meal. J Assoc Off Anal Chem 72:867–869
Belanger G, Bregard A, Michaud R (2002) Phosphorus uptake
and concentration of timothy genotypes under varying N
applications. Crop Sci 42:2044–2048
Breeze VG, Robson AD, Hopper MJ (1985) The uptake of
phosphate by plants from flowing nutrient solution III.
Effect of changed phosphate concentrations on the growth
and distribution of phosphate within plants of Loliumperenne L. J Exp Bot 36:725–733
Brink GE, Pederson GA, Sistani KR, Fairbrother TE (2001)
Uptake of selected nutrients by temperate grasses and
legumes. Agron J 93:887–890
Chen CR, Condron LM, Davis MR, Sherlock RR (2002)
Phosphorus dynamics in the rhizosphere of perennial
ryegrass (Lolium perenne L.) and radiata pine (Pinusradiata D. Don.). Soil Biol Biochem 34:487–499
Chien SH, Menon RG (1995) Factors affecting the agronomic
effectiveness of phosphate rock for direct application. Fert
Res 41:227–234
Dao TH, Sikora LJ, Hamasaki A, Chaney RL (2001) Manure
phosphorus extractability as affected by aluminum- and
iron by-products and aerobic composting. J Environ Qual
30:1693–1698
Table 7 Acid ammonium acetate (AAAc)-extractable P
(mg l-1) in the experimental soils after the first year (four
ryegrass cuts) and third year (ten ryegrass cuts) in 100 mg P
kg-1 treatments
P source AAAc
After first year After third year
Control 2.3 ab 1.7 a
MBM 9.6 e 4.8 cd
cFoxM 7.8 e 4.2 bcd
DairyM 5.1 d 2.6 ab
SP 5.4 d 2.9 abc
Averages with the same letter do not differ significantly
(P [ 0.05, Tukey’s test)
Nutr Cycl Agroecosyst (2008) 81:267–278 277
123
Elonen P (1971) Particle-size analysis of soil. Acta Agric Fenn
122:1–122
FAO (1998) World reference base for soil resources. World
Soil Resources Report 84. FAO, Rome, Italy
Fohse D, Claassen N, Jungk A (1988) Phosphorus efficiency of
plants I. External and internal P requirement and P uptake
efficiency of different plant species. Plant Soil 110:
101–109
Grant CA, Flaten DN, Tomasiewicz DJ, Sheppard SC (2001)
The importance of early season phosphorus nutrition. Can
J Plant Sci 81:211–224
Guo F, Yost RS, Hue NV, Evensen CI, Silva JA (2000)
Changes in phosphorus fractions in soils under intensive
plant growth. Soil Sci Soc Am J 64:1681–1689
Gusewell S (2004) N:P ratios in terrestrial plants: variation and
functional significance. New Phytol 164:243–266
Haynes RJ, Mokolobate MS (2001) Amelioration of Al toxicity
and P deficiency in acid soils by additions of organic
residues: a critical review of the phenomenon and the
mechanisms involved. Nutr Cycl Agroecosyst 59:47–63
Helal HM, Sauerbeck DR (1984) Influence of plant roots on C
and P metabolism in soil. Plant Soil 76:175–182
Holford ICR, Hird C, Lawrie R (1997) Effects of animal
effluents on the phosphorus sorption characteristics of
soils. Aust J Soil Res 35:365–373
Jeng A, Haraldsen TK, Vagstad N, Grønlund A (2004) Meat
and bone meal as nitrogen fertilizer to cereals in Norway.
Agric Food Sci 13:268–275
Jeng AS, Haraldsen TK, Grønlund A, Pedersen PA (2006)
Meat and bone meal as nitrogen and phosphorus fertilizer
to cereals and ryegrass. Nutr Cycl Agroecosyst 76:
183–191
Jones DL (1998) Organic acids in the rhizosphere—a critical
review. Plant Soil 205:25–44
Junge A, Werner W (1989) Investigations on interactions of
phosphorus compounds in partially acidulated phosphate
rock and fertilizer effectiveness. Fert Res 20:129–134
Lamothe PJ, Fries TL, Consul JJ (1986) Evaluation of a
microwave oven system for the dissolution of geologic
samples. Anal Chem 58:1881–1886
MMM (2004) Mid-term evaluation of the Horizontal Rural
Development Programme. (In Finnish with English sum-
mary). MMM:n julkaisuja 1/2004, Vammala
Murphy J, Riley JP (1962) A modified single solution method
for the determination of phosphate in natural waters. Anal
Chim Acta 27:31–36
O’Connor GA, Sarkar D, Brinton SR, Elliott HA, Martin FG
(2004) Phytoavailability of biosolids phosphorus. J Envi-
ron Qual 33:703–712
Øgaard AF (1996) Effect of fresh and composted cattle manure
on phosphate retention in soil. Acta Agric Scand Sect B
46:98–105
Sharpley AN, Moyer B (2000) Phosphorus forms in manure
and compost and their release during simulated rainfall.
J Environ Qual 29:1462–1469
Toal ME, Yeomans C, Killham K, Meharg AA (2000) A
review of rhizosphere carbon flow modelling. Plant Soil
222:263–281
Turtola E, Yli-Halla M (1999) Fate of phosphorus applied in
slurry and mineral fertilizer: accumulation in soil and
release into surface runoff water. Nutr Cycl Agroecosyst
55:165–174
Uusitalo R, Ylivainio K, Turtola E, Kangas A (2007) Accu-
mulation and translocation of sparsely soluble manure
phosphorus in different types of soils after long-term
excessive inputs. Agric Food Sci (in press)
Vuorinen J, Makitie O (1955) The method of soil testing in use
in Finland. Agrogeol Publ 63:1–44
Yli-Halla M (1991) Phosphorus supplying capacities of soils
previously fertilized with different rates of P. J Agric Sci
Finl 63:75–83
Zoysa AKN, Loganathan P, Hedley MJ (1997) A technique for
studying rhizosphere processes in tree crops: soil phos-
phorus depletion around camellia (Camellia japonica L.)
roots. Plant Soil 190:253–265
Zoysa AKN, Loganathan P, Hedley MJ (1999) Phosphorus
utilisation efficiency and depletion of phosphate fractions
in the rhizosphere of three tea (Camellia sinensis L.)
clones. Nutr Cycl Agroecosyst 53:189–201
278 Nutr Cycl Agroecosyst (2008) 81:267–278
123