Applied Soil Ecology 35 (2007) 441–448

www.elsevier.com/locate/apsoil

Nodulation of Medicago truncatula and Medicago

polymorpha in two pastures of contrasting
soil pH and rhizobial populations
Matthew D. Denton a,b,c,*, Christopher R. Hill a, William D. Bellotti a,
David R. Coventry a
a
The University of Adelaide, Roseworthy Campus, SA 5371, Australia
b
School of Plant Biology, The University of Western Australia, 35 Stirling Hwy, Crawley, WA 6009, Australia
c
Primary Industries Research Victoria (PIRVic), Department of Primary Industries, Rutherglen Centre,
RMB 1145 Chiltern Valley Road, Rutherglen Victoria 3685, Australia
Received 17 August 2005; received in revised form 19 February 2006; accepted 7 April 2006

Abstract
The size of the background rhizobial population can often determine the success of field nodulation and persistence of inoculant
rhizobia. Field experiments were conducted to determine the nodulation response of annual medics (Medicago spp.) in a pasture-
wheat-pasture rotation when grown in soils of contrasting pH and rhizobial populations. Medicago truncatula Gaertn. and M.
polymorpha L. were inoculated with one of three different strains of Sinorhizobium medicae (WSM540, WSM688) or S. meliloti
(NA39) or left uninoculated and sown in two fields of pH (CaCl2) 5.9 and 7.2 of differing soil rhizobial backgrounds (11 and
7.1  104 cells/g soil, respectively). Nodulation was assessed in years 1 and 3 of the rotation. At the site with a small rhizobial
background, M. polymorpha nodulated poorly when inoculated with the acid-sensitive strain NA39 but nodulated well when
inoculated with acid-tolerant strains WSM688 and WSM540. M. truncatula had a similar extent of nodulation with each of the
rhizobial inoculants. At the site with a large rhizobial background all treatments had greater than 85% of plants nodulated. Nodule
occupancies, assessed by PCR, provided further insight: at the site with a small rhizobial background both medic species
successfully nodulated with the acid-tolerant strains WSM540 and WSM688 and these strains persisted to year 3. However, at the
site with large rhizobial background, only one strain, WSM688, was identified from M. truncatula nodules in year 3. This study
highlights the importance of edaphic constraints and plant–rhizobia interactions to the successful development of nodulation in a
field environment.
# 2006 Published by Elsevier B.V.

Keywords: Legumes; Medicago truncatula; Medicago polymorpha; Nodule occupancy; Polymerase chain reaction (PCR); Rhizobia;
Sinorhizobium meliloti

1. Introduction

Annual medic (Medicago spp.) pastures contribute

substantially to the nitrogen (N) economy of cereal
* Corresponding author. Tel.: +61 2 6030 4559;
cropping systems in low rainfall (<400 mm) environ-
fax: +61 2 6030 4600. ments of southern Australia (Peoples and Baldock, 2001).
E-mail address: [email protected] (M.D. Denton). In addition to N contributions, the medic component

0929-1393/$ – see front matter # 2006 Published by Elsevier B.V.

doi:10.1016/j.apsoil.2006.08.001
442 M.D. Denton et al. / Applied Soil Ecology 35 (2007) 441–448

enhances soil carbon accumulation and provides oppor- annual medic pastures, the aim of this study was to track
tunities for animal-based enterprises and weed manage- the success of inoculant rhizobia in forming nodules on
ment (Reeves and Ewing, 1993). In southern Australia, annual medics over three seasons. The experiments
annual medics occur on 20 million ha of mainly neutral to comprised pasture-wheat-pasture rotations at two field
alkaline soils (Hill and Donald, 1998). sites with contrasting rhizobial population sizes. We
In long-term crop/medic rotations, the pasture legume investigated the extent to which (i) the legume species
component typically regenerates during the pasture and (ii) the size of the background rhizobial population
phase from seed banks in the soil (Ewing, 1999). To affected the success of inoculants in the pasture phases.
maximise N2 fixation, it is essential that elite rhizobia, M. truncatula and M. polymorpha were inoculated with
introduced as legume pasture inoculants, are able to either an acid-sensitive (NA39) or acid-tolerant rhizobia
survive in the absence of the legume host (Brockwell (WSM688, WSM540) and nodule occupancy of the
et al., 1991). Rhizobia may survive in nodule fragments, medics was assessed in years 1 and 3 of the experiment.
by colonising the rhizospheres of non-hosts, or as free-
living components of the soil microflora. Survival of the 2. Materials and methods
rhizobia in these circumstances must be sufficient to act
as an effective source of inoculation of the regenerating 2.1. Site characteristics and management
legume (Howieson, 1995).
Medicago species vary in their specificity for Two sites with red-brown earths (chromosols) were
appropriate rhizobia, Sinorhizobium meliloti and S. chosen in the Upper-North region of South Australia,
medicae. While Medicago truncatula nodulates and near Pekina (32846.593S, 138833.728E) and Black
fixes N2 with a relatively wide range of rhizobial Rock (32849.323S, 138842.789E). The Pekina site had a
genotypes (Brockwell and Hely, 1966; Brockwell, 2001), clay soil that contained 39 mg/kg (Colwell) P, 0.65%
M. polymorpha is more specific (Brockwell and Hely, organic C, 332 mg/kg K. The Black Rock had a clay soil
1966; Ballard and Charman, 2000; Bena et al., 2005). that contained 27 mg/kg P, 1.15% organic C, 771 mg/kg
Sinorhizobium spp. typically have a strong pre- K prior to sowing. The previous crop was field peas at
ference for neutral to alkaline soils. Survival, persis- Pekina and pasture at Black Rock. The sites were sown
tence, and nodulation of medic rhizobia are all restricted with annual Medicago pastures in the first year of the
under acid conditions (Howieson et al., 1991; Watkins experiment. The experimental design consisted of two
et al., 2003). Poor adaptation to acid soils leads to small cultivars: Medicago truncatula Gaertn. cv. Caliph and
populations of medic rhizobia in low pH soils M. polymorpha L. cv. Santiago and four inoculation
(Brockwell et al., 1991) which can be further reduced treatments: inoculation with either S. meliloti strain
by tillage (Coventry and Hirth, 1992). Below approxi- NA39, S. medicae WSM540 or S. medicae WSM688 or
mately pH 6.5 rhizobial populations are adversely no inoculation. Each of these strains has been used in
affected (Brockwell, 2001). The selection of rhizobia Australian commercial inoculants at one time or
from acid soils has resulted in strains with increased another. NA39 is intolerant of acid soils; strains
acid tolerance (Dilworth et al., 2001). Strain WSM688, WSM540 and WSM688 are tolerant of soils with a
for example, was isolated from a Sardinian soil of pH pH of 4.7 (Howieson et al., 1991). The rhizobia were
4.2 (Howieson et al., 1991). applied as peat inoculants and seed was lime-pelleted
The size of the background rhizobial population also with Plastaid1 in order to enhance rhizobial survival.
influences the success of nodulation and persistence of Following preparation, the pelleted seed was sown at
the inoculant rhizobia (Dowling and Broughton, 1986; 20 kg/ha by broadcasting and covered by light raking.
Howieson and Ballard, 2004). Survival of inoculant Triple super phosphate was applied at 20 kg/ha of P
rhizobia in a small background population (<100 cells/ prior to sowing. The plot design was a randomised
g soil) will depend primarily upon rhizobial tolerance of complete block design with 5 m  1.5 m plots sepa-
the edaphic environment. Conversely, survival of rated by 1.1 m spaces from other plots and blocks
rhizobia introduced into soil with a large background separated by 5 m.
population will depend upon the relative abilities of The sites were sprayed as necessary to control insect
background and introduced strains to compete for the and weed pests. In the second year wheat (Triticum
formation of nodules (Vlassak and Vanderleyden, 1997; aestivum cv. Barunga), was sown at 75 kg/ha with
Denton et al., 2003). 20 kg P/ha as triple super phosphate. In the third year
Since rhizobial persistence is crucial to the successful pastures were allowed to regenerate from the soil seed
establishment, N2 fixation and continuing productivity of bank. Annual rainfall at Pekina was 455, 348 and
 M.D. Denton et al. / Applied Soil Ecology 35 (2007) 441–448 443

456 mm for years 1, 2 and 3 and April–October growing the taproot as this nodule usually forms early and is
season rainfall was 326, 272 and 194 mm. At Black thought to provide a large proportion of nitrogen
Rock, the annual rainfall was 375, 277 and 343 mm fixation (Corbin et al., 1977). In addition, some plants
with April–October rainfall 271, 227, 155 mm over the had just a single very large nodule on the excavated root
3 years of the experiment. system. Nodules were surface sterilised in 95% ethanol
for 30 s; then 3% sodium hypochlorite for 1.5–3 min
2.2. Soil and plant sampling (depending upon nodule size) and washed eight times in
sterile water. Nodules were then crushed, streaked on
Soil cores were taken at three locations across the YEM (Vincent, 1970) and incubated at 28 8C for 4 days.
field sites at depths of 0–10, 10–20 and 20–50 cm to To obtain a pure colony, a single isolate from each
determine pH (1:5 in H2O and CaCl2; Rayment and nodule was re-streaked and used for subsequent PCR
Higginson, 1992). Estimations of most probable analysis. A PCR amplification method (Richardson
numbers (MPN) of Sinorhizobium spp. in the soil were et al., 1995) using the RP05 primer, was chosen for the
made at 0–10 cm depth, using a plant infection method identification of rhizobia. A number of nodule isolates
(Brockwell, 1963) with Medicago sativa L. as the test were confirmed to be rhizobia by re-inoculating M.
species. At Pekina the soil pH (CaCl2) was slightly sativa under aseptic conditions (Vincent, 1970).
acidic but increased with soil depth while the Black
Rock soil had a neutral pH (Table 1). Pekina had a small 2.3. Statistical analyses
rhizobial population while Black Rock had a large soil
surface population (Table 1). Statistical analyses (ANOVA) were performed using
In year 1 medic plants with entire root systems were Genstat 7.1. Where significant differences were
harvested at 15 weeks (early spring), to ensure observed in ANOVA, means were tested using a Tukey
nodulation had occurred and that nodules were still test.
fresh enough to isolate rhizobia effectively. The
percentage of plants nodulated was assessed on 100 3. Results
plants (20 samples/plot) and nodule occupancy was
assessed in 20 nodules from each medic  strain 3.1. Nodulation
treatment using nodules selected from all five replicate
plots. In year 3, plants with entire root systems were The medics responded differently at each field site
excavated in early spring. Rhizobia were isolated from (Fig. 1). At the site with a small rhizobial population
nodules (Vincent, 1970) sampled from field plots in both medic species had poor nodulation when sown
years 1 and 3 of the experiments. Isolates were obtained uninoculated (Fig. 1a). The nodules that did form on
from 20 nodules from each medic  strain treatment inoculated plants at this site were typically large. Only
using nodules from all five replicate plots. Insufficient half or less of the M. truncatula plants at the site with the
rhizobia were successfully isolated from M. polymor- small rhizobial background had nodules (Fig. 1a). In
pha nodulated with strain NA39 in year 1 of the contrast, M. polymorpha nodulated well with WSM540
experiment for PCR analysis. One nodule per plant was or WSM688 and poorly with the acid-sensitive strain
taken from the top 5 cm of root that was on, or close to, NA39. At the site with a large rhizobial background

Table 1
Soil pH (n = 3) and Sinorhizobium meliloti population sizes (derived from most probable number estimates) at the field sites prior to sowing medic
species
Site Soil depth (cm) Soil pHa (H2O) Soil pH (CaCl2) Rhizobial population
(rhizobia/g soil)b
Pekina 0–10 6.4  0.14 a 5.9  0.09 a 11
10–20 6.6  0.06 a 6.3  0.17 b
20–50 7.7  0.07 b 7.2  0.05 c
Black Rock 0–10 7.7  0.03 a 7.2  0.02 a 7.1  104
10–20 7.8  0.02 a 7.4  0.08 b
20–50 8.1  0.09 b 7.4  0.06 b
a
Values within a column group that are followed by different letters are significantly different at P < 0.05 (Tukey).
b
Rhizobial population size estimated using the most probable number technique (Brockwell, 1963).
444 M.D. Denton et al. / Applied Soil Ecology 35 (2007) 441–448

Fig. 1. The percentage of M. truncatula and M. polymorpha nodulated at sites with a low (a) and high (b) background soil rhizobial population in
year 1 of the experiments. The frequency of nodulation (nodulated vs. non-nodulated plants) was assessed using 20 plants per replicate (100 plants/
treatment). Columns that do not share the same letter are significantly different (P  0.05).

more than 85% of plants were nodulated regardless of

inoculation (Fig. 1b). Across all treatments, M.
truncatula had significantly (P < 0.05) greater nodula-
tion than M. polymorpha.

3.2. Nodule occupancy

Rhizobial isolates were identified from nodules

harvested in years 1 and 3 of the field experiments using
PCR with the random primer RP05 (Fig. 2). Banding
patterns of some rhizobial isolates clearly matched the
banding patterns of the commercial inoculants (lanes 3,
4, 6 and 10–15 in Fig. 2) and were considered to be
inoculant strains. Other isolates produced PCR banding
patterns that were clearly dissimilar to the commercial
strains (lanes 1, 2, 5, 7–9; Fig. 2). All nodule isolates Fig. 2. An example of PCR profiles used to identify rhizobial isolates
that were re-inoculated onto M. sativa did form nodules, from the two field sites, amplified using the RP05 primer. Nodule
isolates were collected from M. truncatula inoculated with strain
which indicated that banding profiles were from WSM688 (lanes 1–9; isolates collected in year 3 of the trial) and from
rhizobia. Inoculant strains were not recovered from M. polymorpha inoculated with strain WSM540 (lanes 10–15; isolates
plots that were not inoculated, indicating that cross were collected in year 1 of the trial). Amplification profiles for
contamination of strains was not a factor. commercial strains Sinorhizobium spp. strains WSM688 and
Successful nodule occupancy, based on PCR WSM540 are indicated in lanes on the left of the gel. Matching
amplification profiles for inoculant strain WSM688 are shown in lanes
profiling, was determined for all inoculant strains in 3, 4 and 6; matching amplifications for strain WSM540 are shown in
M. truncatula nodules in year 1 at the low rhizobia site lanes 10–15. SPP1 bacteriophage digested with EcoRI was used as the
and for M. polymorpha inoculated with WSM540 and molecular weight marker.
 M.D. Denton et al. / Applied Soil Ecology 35 (2007) 441–448 445

Fig. 3. Recovery of inoculant rhizobia from nodules sampled in the field at sites with a small (a and b) and large (c and d) background rhizobial
population. Nodule occupancies of M. truncatula and M. polymorpha are shown for years 1 (a and c) and 3 (b and d) of the experiment. Insufficient
nodules prohibited a determination for M. polymorpha nodulated with strain NA39 in year 1 at the site with a small rhizobial population.

WSM688 (Fig. 3a and b). In year 3 of the trial, nodule medics, in turn affecting the establishment of rhizobia
occupancy was only confirmed in M. truncatula and M. populations (Denton et al., 2000). When acid-tolerant
polymorpha for the two acid-tolerant strains, WSM540 rhizobia (WSM688, WSM540) were used to inoculate
and WSM688 (Fig. 3b). M. polymorpha in a mildly acidic soil, nodulation was
At the site with a high rhizobial background, nodule increased above that achieved by no inoculation or
occupancies of inoculant rhizobia in year 1 of the trial inoculation with an acid-sensitive strain (NA39). This
were typically very low than for the site with a low indicates the value of using strains that are well matched
background rhizobial population (Fig. 3c and d). By to edaphic conditions. Uninoculated controls produced
year 3 of the trial, M. truncatula inoculated with very few nodulated plants, reinforcing the estimate of a
WSM688 was the only treatment in which nodule small rhizobial population at the time of sowing the
occupancy was confirmed (Fig. 3d). experiment. M. polymorpha had variable nodulation
with the three rhizobial strains (10–85% of plants) while
4. Discussion M. truncatula nodulated to a similar extent with each of
the three strains (40–50% of plants) which indicates the
This study demonstrates the importance of under- likelihood that host–strain interactions influenced
standing both nodulation and nodule occupancy to nodulation. In a previous study M. polymorpha also
understand the outcome of competition for nodulation had a variable nodulation response when inoculated
in a field environment. Inoculation success was with strain CC169 (15%) or strain WSM688 (50%)
demonstrated to be highly dependent upon host–strain (Howieson et al., 1991). Plants at the site with a low
interactions in field sites with contrasting soil rhizobial rhizobial background produced very large crown
populations. nodules.
Plants at the site with high background rhizobia
4.1. Plant nodulation contained many small nodules. At this site each medic
was well nodulated regardless of inoculation, indicating
The medics at the two locations differed substan- that this background population of rhizobia was
tially in their nodulation response at the two sites sufficiently large and diverse to nodulate both hosts.
(Fig. 1). The differences in rhizobial populations at Surprisingly, 100% nodulation was not observed for all
these sites is likely to be due to edaphic and cultural plants, even with the large background of rhizobia.
attributes influencing the presence or absence of annual While much care was taken in excavating plant roots,
446 M.D. Denton et al. / Applied Soil Ecology 35 (2007) 441–448

some nodules may have been lost in the process. M. strains. However, in the presence of such a large
truncatula showed a small but significantly greater background population, it is difficult to understand why
preference to nodulate than did M. polymorpha, strain WSM688 was so successful in colonising nodules
indicating slightly greater compatibility with the of M. truncatula. Strain WSM688 had not been
background rhizobial population. previously inoculated at this site and the density of
plants and nodules was not higher in this treatment in
4.2. Nodule occupancy year 1 of the experiment (data not presented), indicating
that there was no initial competitive advantage for strain
Nodule occupancies were particularly informative: WSM688.
occupancy of inoculant rhizobia differed markedly at The success of WSM688 in year 3 of the experiment
the two field sites and for each host legume. At the site is notable given that only 20% of nodules were
with a small background of rhizobia, assessment of colonised with this strain in the first year of the trial.
nodule occupancy indicated that both medics nodulated Brockwell et al. (1995) have outlined the difficulty in
with the acid-tolerant strains WSM540 and WSM688 in assuring successful nodulation by inoculant strains
the year of inoculation far more successfully than the under the pressure of competing with large numbers of
acid-sensitive strain NA39. The acid tolerant strains background rhizobia. The successful nodulation
persisted through a year with a non-host (wheat) and between M. truncatula with WSM688 indicates that
two periods of hot, dry summers to nodulate both the specific interactions between the legume and
medics in year 3. This result indicates the importance of rhizobia and are likely to have determined the outcome
using an acid-tolerant strain of rhizobia to survive of nodulation. This finding indicates that under certain
saprophytically in this mildly acidic soil. The particular conditions, particular Medicago–Sinorhizobium com-
attributes that allowed these acid-tolerant strains to binations favour the nodulation of particular strains of
persist in the soil and nodulate medics in the third year rhizobia, despite the presence of large competing
remain to be elucidated. Acid tolerance in rhizobia has background strains of rhizobia. Host species have
been shown to involve maintenance of intracellular pH recently been observed to express a degree of selectivity
(O’Hara et al., 1989) which may be maintained by in nodulation with inoculants in a field environment
increased concentrations of potassium or phosphorus (Yates et al., 2005). The present results differ from other
(Watkins et al., 2003). inoculation experiments that show that the rhizobial
At the site with a large rhizobial background, population is the primary factor influencing the
establishment of inoculant strains was generally poor in outcome of nodulation (Brockwell et al., 1982; Thies
the first year, with a maximum of 20% colonisation of et al., 1991).
M. truncatula nodules by WSM688. This is not Given that M. polymorpha is considered more
surprising, given the large background of rhizobia selective for rhizobia genotypes than M. truncatula
(7.1  104 rhizobia/g soil) into which inoculant rhizo- (Brockwell, 2001), M. polymorpha was considered to be
bia were placed. Large populations of rhizobia can have a more likely candidate to nodulate with WSM688.
a significant impact on inoculation and are likely to be However, some naturalised M. polymorpha were
of concern when the background strains of rhizobia observed at the Black Rock field site and are likely
have a poor capacity to fix N2, when inoculation to have maintained a population of soil rhizobia well-
becomes futile (Denton et al., 2002, 2003). At the site matched with M. polymorpha. In this case, efficient
with a large background rhizobial population, inocula- nodulation of M. polymorpha with the soil rhizobial
tion with different strains provided 0–20% nodulation in population may have restricted nodulation of with
year 1 and inoculant strains occupied very few nodules inoculant strains.
in year 3. This compared poorly with the site that had a While the reason for this particular result is
small rhizobial background in which nodule occupancy unknown, it is instructive to note the importance that
typically varied between 50 and 100%. legume–rhizobia relationships may play in environ-
An interesting result was that strain WSM688 ments with large populations of soil rhizobia. By
occupied 45% of M. truncatula nodules assessed in selecting (or genetically enhancing) legumes and
year 3 of the experiment at the site with a large rhizobia to achieve highly specific plant-rhizobial
background rhizobial population. Strain WSM688 signaling, it may be possible to manipulate the
occupied nodules of M. truncatula successfully at the outcomes of inoculation, despite the size of the
site with a small rhizobial population where there was background rhizobial population into which the legume
very little competition for nodulation from indigenous is sown. It is likely, however, that the genetic structure
 M.D. Denton et al. / Applied Soil Ecology 35 (2007) 441–448 447

and physiology of the rhizobial population present at a Brockwell, J., Pilka, A., Holliday, R.A., 1991. Soil pH is a major
site may be as important to the successful outcomes of determinant of the numbers of naturally occurring Rhizobium
meliloti in non-cultivated soils in central New South Wales. Aust.
inoculation, as is the size of the population. Further J. Exp. Agric. 31, 211–219.
studies would be required to understand the mechan- Corbin, E.J., Brockwell, J., Gault, R.R., 1977. Nodulation studies on
isms that allow the successful nodulation outcome for chickpea (Cicer arietinum). Aust. J. Exp. Agric. Anim. Husbandry
M. truncatula with WSM688 on these soil types. 17, 126–134.
Coventry, D.R., Hirth, J.R., 1992. Effects of tillage and lime on
These results highlight the importance of (i) using
Rhizobium trifolii populations and survival in wheat-subterranean
acid-tolerant strains and (ii) selecting rhizobia that are clover rotation in southeastern Australia. Soil Till. Res. 25, 67–74.
well matched to the legume genotypes of interest for Denton, M.D., Coventry, D.R., Bellotti, W.D., Howieson, J.G., 2000.
successful nodulation in a field environment. These data Distribution, abundance and symbiotic effectiveness of Rhizobium
indicate that, if legumes are suitably matched with leguminosarum bv. trifolii from alkaline pasture soils in South
rhizobia, the possibility exists that effective inoculation Australia. Aust. J. Exp. Agric. 40, 25–35.
Denton, M.D., Coventry, D.R., Murphy, P.J., Howieson, J.G., Bellotti,
can proceed, whether the nodulation constraint is a large W.D., 2002. Competition between inoculated and naturalized
soil population of rhizobia or an acid soil. Rhizobium leguminosarum bv. trifolii for nodulation of annual
clovers in alkaline soil. Aust. J. Agric. Res. 53, 1019–1026.
Denton, M.D., Reeve, W.G., Howieson, J.G., Coventry, D.R., 2003.
Acknowledgements Competitive abilities of a commercial strain and common field
isolates of Rhizobium leguminosarum bv. trifolii for clover nodule
The authors wish to thank Mr. John Cozens, Black occupancy. Soil Biol. Biochem. 35, 1039–1048.
Rock and Mr. Kym Fromm, Pekina for the use of land Dilworth, M.J., Howieson, J.G., Reeve, W.G., Tiwari, R.P., Glenn,
A.R., 2001. Acid tolerance in legume root nodule bacteria and
for these experiments. Dr. Annette Anderson and Mrs.
selecting for it. Aust. J. Exp. Agric. 41, 435–446.
Irnayuli Sitepu are thanked for assisting with running Dowling, D.N., Broughton, W.J., 1986. Competition for nodulation of
PCR and isolating rhizobia from nodules. Dr. Alan legumes. Ann. Rev. Microb. 40, 131–157.
Richardson, CSIRO, Plant Industry, Canberra is Ewing, M.A., 1999. Annual pasture legumes: a vital component
thanked for providing the RP05 primer. The input of stabilizing and rehabilitating low-rainfall Mediterranean Ecosys-
tems. Arid Soil Res. Rehabil. 13, 327–342.
Dr. Alan H. Gibson early in this project was much
Hill, M.J., Donald, G.E., 1998. Australian Temperate Pastures Data-
appreciated. This work was supported by the Grains base. [computer file]. National Pastures Improvement Coordinat-
Research and Development Corporation. ing Committee/CSIRO Animal Production.
Howieson, J.G., 1995. Rhizobial persistence and its role in the
development of sustainable agricultural systems in Mediterranean
References environments. Soil Biol. Biochem. 27, 603–610.
Howieson, J.G., Ballard, R.A., 2004. Optimising the legume symbio-
Ballard, R.A., Charman, N., 2000. Nodulation and growth of pasture sis in stressful and competitive environments within southern
legumes with naturalised soil rhizobia. 1. Annual Medicago spp. Australia—some contemporary thoughts. Soil Biol. Biochem.
Aust. J. Exp. Agric. 40, 939–948. 36, 1261–1273.
Bena, G., Lyet, A., Huguet, T., Olivieri, I., 2005. Medicago–Sinorhi- Howieson, J.G., Ewing, M.A., Thorn, C.W., Revell, C.K., 1991.
zobium symbiotic specificity evolution and the geographic expan- Increased yield in annual species of Medicago grown in acidic
sion of Medicago. J. Evol. Biol. 18, 1547–1558. soil in response to inoculation with acid tolerant Rhizobium
Brockwell, J., 1963. Accuracy of a plant-infection technique for meliloti. In: Wright, R.J. (Ed.), Plant–Soil Interactions at Low
counting populations of Rhizobium trifolii. Appl. Microb. 11, pH. Kluwer Academic Press, Dordrecht, pp. 589–595.
377–383. O’Hara, G.W., Goss, T.J., Dilworth, M.J., Glenn, A.R., 1989. Main-
Brockwell, J., 2001. Sinorhizobium meliloti in Australian soils: popu- tenance of intracellular pH and acid tolerance in Rhizobium
lation studies of the root-nodule bacteria for species of Medicago meliloti. Appl. Environ. Microb. 55, 1870–1876.
in soils of the Eyre Peninsula, South Australia. Aust. J. Exp. Agric. Peoples, M.B., Baldock, J.A., 2001. Nitrogen dynamics of pastures:
41, 753–762. nitrogen fixation inputs, the impact of legumes on soil nitrogen
Brockwell, J., Bottomley, P.J., Thies, J.E., 1995. Manipulation of fertility, and the contributions of fixed nitrogen to Australian
rhizobia microflora for improving legume productivity and soil farming systems. Aust. J. Exp. Agric. 41, 327–346.
fertility: a critical assessment. Plant Soil 174, 143–180. Rayment, G.E., Higginson, F.R., 1992. Australian Laboratory Hand-
Brockwell, J., Gault, R.R., Zorin, M., Roberts, M.J., 1982. Effects of book of Soil and Water Chemical Methods. Inkata Press, Mel-
environmental variables on the competition between inoculum bourne, p. 330.
strains and naturalized populations of Rhizobium trifolii for nodu- Reeves, T.G., Ewing, M.A., 1993. Is ley farming in Mediterranean zones
lation of Trifolium subterraneum L. and on rhizobia persistence in just a passing phase? In: Proceedings of the XVII International
the soil. Aust. J. Agric. Res. 33, 803–815. Grassland Congress. SIR Publishing, Wellington, pp. 2169–2177.
Brockwell, J., Hely, F.W., 1966. Symbiotic characteristics of Rhizo- Richardson, A.E., Viccars, L.A., Watson, J.M., Gibson, A.H., 1995.
bium meliloti: an appraisal of the systematic treatment of nodula- Differentiation of Rhizobium strains using the polymerase chain
tion and nitrogen fixation interactions between hosts and rhizobia reaction with random and directed primers. Soil Biol. Biochem.
of diverse origins. Aust. J. Agric. Res. 17, 885–899. 27, 515–524.
448 M.D. Denton et al. / Applied Soil Ecology 35 (2007) 441–448

Thies, J.E., Singleton, P.W., Bohlool, B.B., 1991. Influence of the size Watkins, E.L.J., O’Hara, G.W., Glenn, A.R., 2003. Physiological
of indigenous rhizobial population on establishment and symbiotic responses to acid stress of an acid-soil tolerant and an acid-soil
performance of introduced rhizobia on field-grown legumes. Appl. sensitive strain of Rhizobium leguminosarum biovar trifolii. Soil
Environ. Microbiol. 57, 19–28. Biol. Biochem. 35, 621–624.
Vincent, J.M., 1970. A Manual for the Practical Study of Root-Nodule Yates, R.J., Howieson, J.G., Real, D., Reeve, W.G., Vivas-Marfisi, A.,
Bacteria. IBP Handbook No. 15. Blackwell, Oxford, 164 pp. O’Hara, G.W., 2005. Evidence of selection for effective nodula-
Vlassak, K.M, Vanderleyden, J., 1997. Factors influencing nodule tion in the Trifolium spp. symbiosis with rhizobium legumino-
occupancy by inoculant rhizobia. Crit. Rev. Plant. Sci. 16, 163–229. sarum biovar trifolii. Aust. J. Exp. Agric. 45, 189–198.