Phosphorus-efficient pastures: legume root traits for improved nutrient foraging

Rebecca Haling1, Zongjian Yang1, Natalie Shadwell1, Richard Simpson1, Richard Culvenor1, Adam Stefanski1, Graeme Sandral2, Daniel Kidd3, Hans Lambers3 and Megan Ryan3
CSIRO Agriculture Flagship, GPO Box 1600 Canberra, ACT, 2601, rebecca.haling@csiro.au
NSW Department of Primary Industries, Pine Gully Road, Wagga Wagga, NSW 2650, graeme.sandral@dpi.nsw.gov.au
3
University of Western Australia, School of Plant Biology & Institute of Agriculture, Perth, 6009, megan.ryan@uwa.edu.au
1
2

Abstract
The critical soil phosphorus (P) requirement (P supply for 90% maximum yield) of many temperate pastures is
determined by the high P requirements of key pasture legumes (commonly Trifolium or Medicago spp.). Legumes
that yield well with a lower critical P requirement could reduce the fertiliser requirement of these pastures.
Pot experiments were used to: (i) identify legumes with root traits likely to confer P acquisition efficiency,
and (ii) test the impact of these traits on critical P requirement. An initial screen of the root hair length (RHL)
and specific root length (SRL) of 13 legumes and two grasses was undertaken. Growth and root morphology
of five legumes (Trifolium subterraneum, T. hirtum, Ornithopus sativus, O. compressus, Biserrula pelecinus)
and one grass (Dactylis glomerata) were subsequently compared in detail using a sandy loam soil (8.3 mg/kg
Colwell P) that was amended by applying P to the top 5 cm of the soil profile (0, 15, 30, 70, 135, 250 mg P/
kg). Shoot and root growth were assessed after six weeks.
Significant variation in RHL (0.12-0.75 mm) and SRL (98-603 m/g) was found among the legumes, with
most being substantially shorter (RHL) or lower (SRL) than the grasses. In the P-response experiment,
Ornithopus species (the only legumes with RHL and SRL approaching that of the grasses) had critical P
requirements that were less than half that of T. subterraneum.
Selecting legumes that maximise root foraging via long, thin roots with long root hairs may reduce the
critical P requirement of pastures.
Introduction
The critical external phosphorus (P) requirement (P required for 90% maximum yield) of many temperate
pastures is determined by the high P requirements of key pasture legumes (commonly Trifolium or Medicago
spp.; Ozanne et al. 1969). Selecting legumes with lower external requirements for P could reduce the amounts
of P fertiliser that need to be applied to pastures because soil managed at lower soil test P concentrations are
likely to sorb less P (Simpson et al. 2014). The features of plants that allow them to acquire P more effectively
are longer, finer roots with longer root hairs and improved ability to forage for P (Richardson et al. 2011).
This work aimed to: (i) identify potentially useful variation in root morphology associated with nutrient
foraging among a range of novel legume species being developed in Australia (Nichols et al. 2007), and (ii)
test the impact of these traits on critical external P requirement.
Method
Variation in root hair length and specific root length
The root hair length (RHL) and specific root length (SRL) of 14 legumes and two grasses (Table 1) were assessed
in separate experiments. In the first experiment for RHL, a sandy loam soil was steam pasteurised, sieved (<5 mm)
and mixed with lime to raise pH (CaCl2) to 5.5. A complete nutrient solution was applied to the soil (extractable P
33 mg/kg; Colwell 1963). Pots (90 mm diam.; 200 mm height) were filled with 1.33 kg of the soil. Two plants were
established per pot and inoculated with an appropriate strain of Rhizobium. Soil was maintained at approximately
75% field capacity. Plants were grown in a controlled environment cabinet (15-20C; photon flux density 600 mol/
m2/s1; 12 h light/ dark). Four replicates were grown per species. Roots were washed from the soil four weeks after
seedling emergence. Root hairs were imaged using a fluorescence microscope fitted with a camera. RHL of ten root
hairs was measured on ten images per replicate using Image J (Rasband 1997-2014).
In the second experiment, SRL was assessed on three- and six-week old plants. SRL of the legumes
was correlated between harvests (R2=0.75) and results are only presented for the six-week old plants.
2015 Building Productive, Diverse and Sustainable Landscapes
Proceedings of the 17th ASA Conference, 20 24 September 2015, Hobart, Australia. Web site www.agronomy2015.com.au

Pots (90 mm diam.; 400 mm height) were filled with 2.746 kg of recycled potting mix containing added
superphosphate (extractable P 55 mg/kg; Colwell 1963). One plant was established per pot and inoculated
with an appropriate strain of Rhizobium. Soil was maintained at approximately 75% field capacity by
watering with a P-free nutrient solution. The plants were grown in a glasshouse (15-20C) under natural
lighting in May to July 2013 in Canberra, Australia. Five replicates were grown per species. Plants were
harvested six weeks after sowing. Roots were washed from the soil, scanned to determine root length using
WinRHIZO (Regent Instruments Inc.) and dry mass determined after drying at 70C. SRL was calculated as
length per unit root mass. Data from both experiments were analysed using ANOVA in R (R Core Team).
Growth in response to phosphorus application
In the third experiment, a subset of the species (Figure 1) was selected to determine growth in response to
P application and critical external P requirement. Soil was collected, pasteurised, sieved and limed as per
the screen for RHL, and amended with a P-free nutrient solution. Pots (90 mm diam.; 200 mm height) were
filled with a bottom layer of 1.0 kg of low P soil followed by a top layer of 0.333 kg of the same soil (11%
moisture) that had been amended by mixing with KH2PO4 at rates of 0, 15, 30, 70, 135 and 250 mg P/kg
(oven dry soil) to establish six P application treatments (n=5 replicates). This topsoil-subsoil arrangement
was used to mimic the stratification of P that occurs in soil under pastures. Plants were grown in a controlled
environment cabinet (15-20C; photon flux density 900 mol/m2/s1; 12 h light/dark) as microswards with
reflective sheets fitted around each pot and raised to equal plant height to reproduce the light conditions in
a pasture. Soil was maintained at approximately 75% field capacity. Shoots and roots from the fertilised
topsoil and from the subsoil layer were harvested six weeks after sowing. Roots from the topsoil were
scanned to determine root length using WinRHIZO (Regent Instruments Inc.). Dry mass of roots and shoots
was determined after drying at 70C. Total P uptake by the plants was determined by ashing root and shoot
dry matter at 550C, dissolving the ash in HCl and determining the P concentration of the solution using
malachite green. A Mitscherlich response curve [y = a - b*(e-cx)] was fitted to the shoot dry mass data in
R (R Core Team). Critical external P requirement was determined as the P application rate corresponding
with 90% of maximum shoot yield. Root mass fraction (root mass in each soil layer as a proportion of total
plant mass) was determined. Root mass fraction and total P uptake per unit root length were analysed using
ANOVA in GenStat 16th Ed (VSN International).
Results
RHL of the legumes ranged 6-fold from 0.12 to 0.75 mm (Table 1). SRL of the legumes ranged 3-fold from
98 to 320 m/g (Table 1). The key legume species used in temperate pastures in Australia, T. subterraneum,
had relatively short root hairs and low SRL. The Ornithopus spp. and B. pelecinus had RHLs and SRLs that
approached that of the grasses.
Table 1. Root hair length and specific root length of 12 legume and two grass species.
Scientific name
Bituminaria bituminosa
Trifolium tumens
Trifolium incarnatum
Trifolium subterraneum
Trifolium spumosum
Trifolium ambiguum
Trifolium purpureum
Trifolium hirtum
Medicago sativa
Lotus corniculatus
Biserrula pelecinus
Ornithopus sativus
Ornithopus compressus
Phalaris aquatica
Dactylis glomerata
LSD (P=0.05)

Common name
tedera
talish clover
crimson clover
subterranean clover
bladder clover
Caucasian clover
purple clover
rose clover
lucerne
birdsfoot trefoil
biserrula
French serradella
yellow serradella
phalaris
cocksfoot

Root hair length

(mm)
0.12
0.21
0.23
0.23
0.25
0.28
0.29
0.30
0.37
0.44
0.56
0.73
0.75
0.86
1.02
0.06

Specific root length

(m/g)
98
nd
259
159
239
187
177
290
209
205
299
320
307
371
603
45

2015 Building Productive, Diverse and Sustainable Landscapes

Proceedings of the 17th ASA Conference, 20 24 September 2015, Hobart, Australia. Web site www.agronomy2015.com.au

Trifolium hirtum was also notable in that it had the longest root hairs and highest SRL amongst the Trifolium
species. It was surmised that the root traits of these species might confer an advantage for P acquisition
species.
It relative
was surmised
that the root traits
these of
species
conferinanresponse
advantage
P acquisition
efficiency
to T. subterraneum.
The of
growth
these might
five species
to Pfor
was
compared using
efficiency
relative
to T. subterraneum.
The agrowth
of these
five species in response to P was compared using
D. glomerata
as a benchmark
species with
low critical
P requirement.
D. glomerata as a benchmark species with a low critical P requirement.
Shoot growth of all species increased in response to addition of P. Dactylis glomerata had the highest
Shoot growth of all species increased in response to addition of P. Dactylis glomerata had the highest
maximum yield and lowest critical external P requirement (Table 2). The Ornithopus spp. yielded as well as
maximum yield and lowest critical external P requirement (Table 2). The Ornithopus spp. yielded as well as
T. subterraneum but achieved this with less than half the amount of applied P (Table 2). Biserrula pelecinus
T. subterraneum but achieved this with less than half the amount of applied P (Table 2). Biserrula pelecinus
and
T. hirtum
hirtum had
had critical
critical external
external P
P requirements
and T.
requirements that
that were
were intermediate
intermediate to
to that
that of
of T.
T. subterraneum
subterraneum but
but also
also
had significantly
significantly lower
had
lower yields.
yields.
-Cx
Table 2.
Table
2. Critical
Critical external
external P
P requirement
requirement and
and parameters
parameters for
for aa Mitscherlich
Mitscherlich curve
curve [y
[y == A
A -- B*(e
B*(e-Cx)])]in
inresponse
response to
to
addition
of
P
for
five
legumes
and
one
grass
species.
Different
lowercase
letters
denote
significant
differences
addition of P for five legumes and one grass species. Different lowercase letters denote significant differences
(P<0.05; n=5).
n=5).
(P<0.05;
Species
Critical
external
Intercept
A
B
C
Critical
Intercept
A
B
C
P
Yield
at
P0
Max.
yield
Species
external P
Yield at P0
Max. yield
(mg P/pot)
P/pot)
(g/pot)
(g/pot)
(mg
(g/pot)
(g/pot)

Dactylis glomerata
Dactylis glomerata
Ornithopus
Ornithopus compressus
compressus
Ornithopus sativus
Trifolium subterraneum
subterraneum
Trifolium
Biserrula pelecinus
Biserrula
Trifolium pelecinus
hirtum
Trifolium hirtum

6.6 0.6a
6.6 0.6a
7.6 0.5a
0.5a
7.6
11.3 0.5b
26.7
26.7 1.3c
1.3c
17.3 1.0d
17.3
1.5e
1.0d
21.1
21.1 1.5e

2.22 0.07a
2.22 0.07a
1.12 0.05b
0.05b
1.12
0.83 0.04c
0.41
0.41 0.05d
0.05d
0.36 0.05d
0.36 0.06d
0.05d
0.36
0.36 0.06d

3.69 0.03a
3.69 0.03a
2.87 0.03b
0.03b
2.87
2.70 0.03c
2.68
2.68 0.05c
0.05c
2.04 0.04e
2.04 0.04e
0.04e
2.03
2.03 0.04e

0.4

Topsoil root mass fraction

(a)

Dactylis glomerata
Cocksfoot
Ornithopus
compressus
Yellow
serradella

Ornithopus
sativus
French
serradella

0.3

Biserrula
Biserrula pelecinus
Rose
Clover
Trifolium
hirtum
Sub
clover subterraneum
Trifolium

0.2

0.1

Subsoil root mass fraction

0
(b) 0

5
4.5

0.3

9
21
41
P fertilised applied (mg pot-1)

75

0.2

-1.47 0.06
-1.47 0.06
-1.76 0.06
0.06
-1.76
-1.87 0.06
-2.27
-2.27 0.06
0.06
-1.69 0.06
-1.69 0.06
0.06
-1.67
-1.67 0.06

0.811 0.017
0.811 0.017
0.788 0.016
0.016
0.788
0.841 0.011
0.923
0.923 0.005
0.005
0.885 0.010
0.885 0.008
0.010
0.905
0.905 0.008

All All
species
increased
theirtheir
species
increased
proportional
allocation
of dry
proportional allocation
of dry
matter
to
roots
(i.e.
root
mass
matter to roots (i.e. root mass
fraction [RMF]) in response to
fraction [RMF]) in response
decreasing supply of P; the effect
to decreasing supply of P; the
was most pronounced in the Peffect topsoil
was most
pronounced
in
amended
(Figure
1). Species
the
P-amended
topsoil
(Figure
with higher critical external
1). Speciesfor
with
highertocritical
requirements
P tended
external
requirementsmore
for P tended
allocate
proportionately
to
allocate
proportionately
more
biomass to roots and demonstrated
to roots and
demonstrated
the biomass
largest increases
in RMF
in
response
to lowincreases
P supply.
the largest
inIncreases
RMF in
in RMF
weretomost
at
response
low pronounced
P supply. Increases
levels
of
P
supply
at,
or
just
below
in RMF were most pronounced at
the levels
criticalofexternal
P requirement
P supply
at, or just below
of each
species.
the critical external P requirement
of each species.

0.1

0.0

10
9.0

23
21.0

45
40.5

83
75.0

P a(a)
pplied
(mg and
pot-1(b)
) subsoil roots for five
Figure 1. Root mass fraction for
topsoil
legumes
and one
grass
grown
sixtopsoil
rates ofand
P applied
to the
topsoil.
Bar
Figure
1. Root
mass
fraction
forat(a)
(b) subsoil
roots
for five
shows LSD
interaction
(P<0.05;
n=5). to the topsoil. Bar
legumes
and for
onetwo-way
grass grown
at six rates
of P applied
shows LSD for two-way interaction (P<0.05; n=5).

2015 Building Productive, Diverse and Sustainable Landscapes

2015 "Building Productive, Diverse and Sustainable Landscapes "
Proceedings of the 17th ASA Conference, 20 24 September 2015, Hobart, Australia. Web site www.agronomy2015.com.au
Proceedings of the 17th ASA Conference, 20 24 September 2015, Hobart, Australia. Web site www.agronomy2015.com.au

Dactylis glomerata

100

Ornithopus compressus

60

Trifolium subterraneum

pqr
pq

qrs
l

lm

no
mn

no
no

op

Trifolium hirtum

qrs

Biserrula pelecinus

80

rs

Ornithopus sativus

P uptake
perper
unitunit
root
mass
in in
P uptake
root
mass
thethe
fertilised
topsoil
increased
in
fertilised topsoil
response
to addition
of Pto(Figure
increased
in response
2).addition
The Ornithopus
spp. 2).
generally
of P (Figure
The
spp. generally
hadOrnithopus
higher P uptake
per unit root
hadthan
higher
uptake
per unit
mass
the Pother
legumes.
root mass than the other
legumes.

f
cd

cd

g
g

20

hi

ij

ij

k
k

40

ef
de
c
b
a

P uptake per unit topsoil root mass (mg g-1)

120

4.5
5

9.0
9

21.0
21

40.5
41

75.0
75

P applied (mg pot-1)

Figure 2. P uptake per unit mass of roots in the P-fertilised topsoil for five legumes and one grass grown at six
Figure 2. P uptake per unit mass of roots in the P-fertilised topsoil for five legumes and one grass grown at six
rates of P applied to the topsoil. Different letters denote differences for two-way interaction (P<0.05; n=5).
rates of P applied to the topsoil. Different letters denote differences for two-way interaction (P<0.05; n=5).

Discussion
Discussion
Trifolium subterraneum
hairs and
and low
low SRL
SRL relative
Trifolium
subterraneum and
and M.
M. sativa
sativa have
have short
short root
root hairs
relative to
to some
some of
of the
the grasses
grasses with
with which
theycommonly
are commonly
grown.
This limits
the ability
of these
species
to forage
and contributes
which
they are
grown.
This limits
the ability
of these
species
to forage
for P for
andPcontributes
to
to their
high
external
P requirement.
However,
significant
variation
root
traits
associated
with
nutrient
their
high
external
P requirement.
However,
significant
variation
in in
root
traits
associated
with
nutrient
foraging was found among a range of alternative legumes that are being developed for temperate pastures in
Australia. Most notably, O. compressus, O. sativus and B. pelecinus had RHLs and SRLs approaching that of
D. glomerata and P. aquatica.
Ornithopus
Ornithopus compressus
compressus and
and O.
O. sativus
sativus had
had aa critical
critical external
external requirement
requirement more
more similar
similar to
to that
that of
of D.
D.
glomerata
but
yielded
as
well
as
T.
subterraneum.
These
species
achieved
this
despite
allocating
glomerata but yielded as well as T. subterraneum. These species achieved this despite allocating
proportionately
Relatively high
high rates
rates of
of P
P uptake
uptake per
per unit
unit
proportionately less
less biomass
biomass to
to roots
roots in
in response
response to
to low
low PP supply.
supply. Relatively
root
mass
in
the
Ornithopus
spp.
supported
the
hypothesis
that
long
root
hairs
and
high
SRL
allows
these
root mass in the Ornithopus spp. supported the hypothesis that long root hairs and high SRL allows these
species to efficiently explore soil with concomitant advantages for P acquisition that contribute to their low
species to efficiently explore soil with concomitant advantages for P acquisition that contribute to their low
critical external requirement for P.
critical external requirement for P.
Conclusion
Conclusion
Selecting legumes that can maximise nutrient foraging via long, thin roots with long root hairs may reduce
Selecting legumes that can maximise nutrient foraging via long, thin roots with long root hairs may reduce
the critical P requirement of pastures.
the critical P requirement of pastures.
Acknowledgements
Acknowledgements
The work was funded by Meat and Livestock Australia and Australian Wool Innovation Limited and by a
The work was funded by Meat and Livestock Australia and Australian Wool Innovation Limited and by a
CSIRO Summer Studentship to NS.
CSIRO Summer Studentship to NS.
References
References
Colwell JD (1963) The estimation of the phosphorus fertiliser requirements of wheat in southern New South
Colwell Wales
JD (1963)
Theanalysis.
estimation
of the phosphorus
requirements
of wheat
in southern New South
by soil
Australian
Journal of fertiliser
Experimental
Agriculture,
3, 190-197.
Wales
by
soil
analysis.
Australian
Journal
of
Experimental
Agriculture,
3,
190-197.
Nichols PGH, Loi A, Nutt BJ et al. (2007) New annual and short-lived perennial pasture legumes for
Nichols Australian
PGH, Loi A,
Nutt BJ et
al. years
(2007)ofNew
annual Field
and short-lived
perennial
pasture
agriculture
- 15
revolution.
Crops Research,
104,
10-23.legumes for
Australian
agriculture
15
years
of
revolution.
Field
Crops
Research,
104,
10-23.
Ozanne PG, Keay J and Biddiscombe EF (1969) Comparative applied phosphate requirements of eight
Ozanne annual
PG, Keay
J andspecies.
Biddiscombe
EF (1969)
applied
phosphate
requirements of eight
pasture
Australian
JournalComparative
of Agricultural
Research,
20, 809-818.
annual
pasture
species.
Journal
of Agricultural
Research,
20, 809-818.
R Core
Team
(2013)
R: AAustralian
language and
environment
for statistical
computing:
R Foundation for Statistical
Computing,
Austria.and environment for statistical computing: R Foundation for Statistical
R Core Team
(2013)Vienna,
R: A language
Rasband
WS (1997-2014)
ImageJ: U. S. National Institutes of Health, Bethesda, Maryland, USA.
Computing,
Vienna, Austria.
Richardson
AE,
Lynch
JP,
Ryan PR
(2011)Institutes
Plant andofmicrobial
strategiesMaryland,
to improveUSA.
the phosphorus
Rasband WS (1997-2014) ImageJ:
U. et
S. al.
National
Health, Bethesda,
efficiency
of agriculture.
Plant
and(2011)
Soil, 349,
Richardson
AE, Lynch
JP, Ryan PR
et al.
Plant121-156.
and microbial strategies to improve the phosphorus
Simpson
RJ,
Richardson
AE,
Nichols
SN
and
Crush
JR
efficiency of agriculture. Plant and Soil, 349, 121-156.(2014) Pasture plants and soil fertility management
improve
the efficiency
phosphorus
in temperate
systems.
Crop &
SimpsontoRJ,
Richardson
AE, NicholsofSN
and Crush fertiliser
JR (2014)use
Pasture
plants andgrassland
soil fertility
management
Pasture
Science,
65,
556-575.
to improve the efficiency of phosphorus fertiliser use in temperate grassland systems. Crop & Pasture
2015 "Building Productive, Diverse and Sustainable Landscapes "
4
Science, 65, 556-575.
Proceedings of the 17th ASA Conference, 20 24 September 2015, Hobart, Australia. Web site www.agronomy2015.com.au
2015 Building Productive, Diverse and Sustainable Landscapes
Proceedings of the 17th ASA Conference, 20 24 September 2015, Hobart, Australia. Web site www.agronomy2015.com.au