APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Sept. 1987, p. 2090-2097 Vol. 53, No. 90099-2240/87/092090-08$02.00/0Copyright X) 1987, American Society for Microbiology
Influence of Lime and Phosphate on Nodulation of Soil-GrownTrifolium subterraneum L. by Indigenous Rhizobium trifoliit
ANGELA S. ALMENDRAS't AND PETER J. BOTTOMLEY' 2*
Departments of Soil Sciencel and Microbiology,2 Oregon State University, Corvallis, Oregon 97331-3804
Received 15 April 1987/Accepted 11 June 1987
Previous research had identified four serogroups of Rhizobium trifolii indigenous to the acidic Abiqua soil(fine, mixed, mesic Cumulic Ultic Haploxeroll). Nodulation of subterranean clover (Trifolium subterraneum L.)by two of the serogroups, 6 and 36, was differentially influenced by an application of CaCO3 which raised thepH of the soil from 5.0 to 6.5. These studies were designed to characterize this phenomenon morecomprehensively. Liming the soil with either CaCO3, Ca(OH)2, MgO, or K2CO3 significantly (P = 0.05)increased the percent nodule occupancy by serogroup 36, whereas the percent nodule occupancy by serogroup6 was decreased, but the decrease was significant (P = 0.05) only after application of either CaCO3 or Ca(OH)2.Application of KH2PO4 (25 mg of P kg of soil-'), which did not change soil pH, also significantly (P = 0.05)increased the percent nodule occupancy by serogroup 36. Application of KH2PO4 in combination with Ca(OH)2produced the same increase in nodule occupancy by serogroup 36 as did individual application of the twomaterials. Soil populations of serogroup 36 consistently, and in the majority of cases significantly (P = 0.05),outnumbered those of serogroup 6 before planting and after harvest regardless of soil treatment or the outcomeof nodulation. Soil chemical and plant analyses provided no evidence that liming was simulating phosphateaddition by increasing the availability and subsequent uptake of soil Pi by the subclover plants. Liming did,however, result in a significant transformation (30 to 50 mg of P kg of soil1) of Pi from the residual soil Pifraction into an NaOH-extractable organic P fraction during the preplant equilibration period.
For many years, factors associated with acid soils haveoften been diagnosed as impediments to the establishmentand subsequent productivity of subterranean clover (Trifo-lium subterraneum L.). Although liming such soils tends, ingeneral, to alleviate many establishment problems (2, 3), thecritical steps that are influenced by such a treatment can benumerous. For example, improvement in establishmentand/or growth as a result of liming has been attributed toincreased molybdenum and/or phosphate availability to theplant (2, 12, 14, 21, 37). Spencer (44), however, suggestedthat liming enhanced nodulation as a result of increasingboth the soil calcium level and soil pH. Subsequently, ahigher requirement for calcium in nodulation under acidconditions (26, 29, 36) than for subclover growth (28) wasconfirmed.Concerns have also been expressed for the influence of
acid soil conditions on survival and proliferation of themicrosymbiont Rhizobium trifolii. Lime improved the pro-liferation of the latter in both the absence (10, 11, 34, 50) andthe presence (32, 40) of the host plant. Norris (34), however,raised the issue of whether the increase in divalent cationconcentration or the increase in pH was really the criticalfactor. Numerous reports in the literature have shown vari-ation among strains of R. trifolii regarding their abilities togrow and/or nodulate clover singly, or in competitive situa-tions, in the presence of acid soil-related stresses (25, 38, 41,46-48, 52-55).
* Corresponding author.t Oregon State University Agricultural Experiment Station Tech-
nical Paper no. 8196.t Present address: Department of Agronomy and Soil Science,
Visayas College of Agriculture, Baybay, Leyte, Republic of thePhilippines.
In this connection, we previously reported upon theheterogeneous nature of an indigenous R. trifolii populationin root nodules of subclover growing in an acidic (pH 5.0)soil (15). The composition of the population recovered fromroot nodules of plants grown in unamended soil was differentfrom that found in nodules of plants grown in the same soilafter it had been limed with CaCO3 (16, 17). The objectivesof this study were (i) to characterize more comprehensivelythe effects of lime on the growth and nodulation of subter-ranean clover; (ii) to enumerate and monitor the populationdynamics of specific indigenous serogroups in response toliming of the soil; and (iii) to explore the possibility that otherinteractive roles of lime, independent of calcium or pH perse, were the cause of the nodule occupancy changes.
MATERIALS AND METHODS
Soil. Surface samples (0 to 0.3 m) of soil were collectedfrom the Al horizon of a silty-clay loam of the Abiqua series(a member of the fine, mixed, mesic, Cumulic UlticHaploxerolls), mixed thoroughly, and, while moist (14 +
0.2% [wt/wt] water), passed through a 2-mm mesh screen.Soil analyses were carried out by using the standard methodsof the soil testing laboratory, Department of Soil Science,Oregon State University (5), except that extractable alumi-num was determined by a modified Aluminon method (23)after extraction with 1 M KCl (4). Salient characteristics ofthe soil relevant to this study are as follows: pHH2O, 5.2;pHCaCl2, 4.7; dilute acid-ammonium fluoride-extractablephosphate, 1 to 5 mg of P kg of soil-'; cation exchangecapacity, 30.6 cmol (+) kg of soil-'; base saturation, 60%;extractable aluminum, 25 mg of Al kg of soil-'; organic
2090
on April 2, 2020 by guest
http://aem.asm
.org/D
ownloaded from
LIME AND PHOSPHATE EFFECTS ON R. TRIFOLII 2091
carbon, 35 g of C kg of soil-'. Fresh samples of soil were
collected for each of the four experiments described below.Experiment 1. We compared the effect of four liming
materials [CaCO3, Ca(OH)2, MgO, and K2CO3] on noduleoccupancy by indigenous serogroups of R. trifolii. Theamounts of liming material used were based on their respec-
tive equivalent weights needed to raise the soil pH to 6.5 as
determined by lime requirement measurements (42). Eachlime material (analytical grade) was mixed thoroughly withthe soil samples. Since the soil was low in extractablephosphate (1 to 5 mg of P kg-'), a supplemental treatment ofKH2PO4 added at 25 mg of P kg-' to unlimed soil was
included. All the soil samples were brought to, and main-tained at, a water potential of 33 kPa (37 g of water 100 g ofdry soil-') and allowed to equilibrate for 4 weeks at 21 + 2°Cuntil the pH values of the lime treatments had stabilized atpH 6.6 + 0.2. For every treatment, 0.4 kg of soil was placedinto each of four plastic-lined pots (8.3 by 22 cm) equippedwith a perforated polyvinyl chloride tube (diameter, 1.8 cm)extending the full length of the pot to facilitate watering. Atotal of 10 to 15 surface-sterilized seeds of T. subterraneumL. cv. Mt. Barker were sown into each pot. The plants were
thinned to three per pot when the seedlings were 3 to 5 cm inheight. The plants were grown and maintained under green-
house conditions previously described (15). At 10 weeksafter sowing (early flowering stage), the plants were har-vested, and as many nodules as possible were recoveredfrom the root systems of the three plants in each pot.
(i) Evaluation of nodule occupancy. Nodules were surfacesterilized by standard methods (49), and 25 were selected atrandom from each replicate of each treatment. For eachnodule triplicate smears were prepared and stained with thenon-cross-reacting fluorescein-labeled immunoglobulin con-
jugates of serogroups 6, 27, or 36, and occupancy was
determined by immunofluorescence (13).(ii) Enumeration of the soil populations of serogroups 6, 27,
and 36. After harvest, 10-g subsamples of root-associatedsoil were taken from each pot, and rhizobia were extractedand enumerated by immunofluorescence as described previ-ously (13) with a minor modification. To flocculate the soilcolloids in the gelatin-ammonium phosphate suspension,1.83 g of a dry, finely ground mixture of CaCl2 2H20 andMgCO3 was added (13.3:5 [wt/wt] was predetermined to bethe most efficient ratio for flocculating Abiqua soil). Toevaluate the viability of the cells recovered from the soil andenumerated by immunofluorescence, dilution series of theabove soil samples were made over the range of 10-1 to10-7.Portions of each dilution were inoculated onto four replicateseedlings of T. subterraneum L. cv. Mt. Barker. The plantswere scored for nodulation, and nodules were recoveredfrom the final positive dilutions of each treatment. Thenodule occupancy by serogroups 6 and 36 was evaluated byimmunofluorescence.Experiment 2. The effects of four phosphate levels (0, 25,
55, and 115 mg of P kg of soil-') upon growth and nodulationof T. subterraneum L. cv. Mt. Barker were evaluated.Monobasic potassium phosphate (KH2PO4) was mixed thor-oughly with soil immediately prior to potting and seeding. Tothe four samples treated with phosphate, molybdenum (1 mgof Mo kg-') as Na2MoO4 2H20 and sulfur (22 mg of S kg-')as K2SO4 were applied, and a control consisting ofunamended soil was included. For each treatment, 0.7 kg ofsoil was placed into each of four pots (8.3 by 22 cm), andsowing and seedling thinning were carried out as describedabove. At 11 weeks after sowing, plants were harvested,nodules were recovered, and nodule occupancy by sero-
groups 6 and 36 was determined by immunofluorescence asdescribed above.
Shoots and roots (devoid of nodules) were separated andoven dried at 65°C to a constant weight. Plant samples wereground to pass through a 1-mm screen and redried beforebeing analyzed for N, P, K, and S. Samples of root and shoottissue were digested by the Kjeldahl method (6) and ana-lyzed for total N and P. Potassium was extracted fromcomposite samples of shoot and root tissue by nitric aciddigestion followed by perchloric acid digestion (22) andquantified with an atomic absorption spectrophotometer (no.4000; The Perkin-Elmer Corp., Norwalk, Conn.). We deter-mined the sulfur content in composite shoot samples ofplants subjected to each treatment by digestion by themethod of Tabatabai and Bremner (45) and analysis by themethod of Johnson and Nishita (24).Experiment 3. The results of experiments 1 and 2 estab-
lished a link between the effects of Ca(OH)2 and phosphateupon nodule occupancy by serogroup 36. A factorial exper-iment was designed with two levels of Ca(OH)2 (0 and 4 gkg-') and of KH2PO4 (0 and 25 mg of P kg-') to evaluatemore comprehensively the impact of lime and phosphate,separately and in combination, on (i) the populations of R.trifolii during the preplant equilibration period, (ii) selectedsoil chemical properties, and (iii) plant growth, phosphateuptake, and composition of nodule occupancy. Soil used ineach of the following three experiments was prepared asfollows. A 10.5-kg portion of soil was limed with 4 g ofanalytical-grade Ca(OH)2 kg-' and another 10.5-kg portionwas unamended. Both soil samples were maintained at awater potential of 33 kPa and equilibrated for 20 days atroom temperature (20°C). Treatments specific to each exper-iment are described in the relevant subsections below.
(i) Experiment 3a. During the 20-day equilibration period,direct enumeration of serogroups 6 and 36 by immunofluo-rescence was carried out at 0, 6, and 20 days after liming bythe procedures described above.
(ii) Experiment 3b. After the 20-day equilibration period,KH2PO4 was mixed thoroughly with subsamples of eachsoil, which were potted (0.5 kg pot-') into four replicates pertreatment. Soil samples were taken at intervals (0, 5, 10, and15 days) after application of KH2PO4. One replicate wasused at each sampling time, and the soil was divided for bothsoil solution analyses and determination of extractable alu-minum and phosphate. Extractable Al was displaced with 1M KCI (4), and extractable phosphate was displaced with0.03 M NH4F in 0.025 M HCI (35). Soil solutions wererecovered from 400-g portions of soil by centrifugal displace-ment (39) with the following modifications. Soil solutionswere composited by treatment and passed through 0.4,um-pore-size membrane filters (Millipore Corp., Bedford,Mass.). Aluminum and phosphate were determined colori-metrically by the Aluminon (23) and molybdate blue (33)methods, respectively. Calcium, Mg, and K were deter-mined by atomic absorption spectrophotometry.
(iii) Experiment 3c. After the 20-day equilibration period,different portions of the unlimed and limed soil samples wereamended uniformly with Mo and S as described above andselectively with the respective KH2PO4 treatment. Soil waspotted (0.7 kg pot-') into four replicates per treatment, andgrowth, plant analysis, and nodule occupancy were evalu-ated as described above.Experiment 4. Results of experiment 3 revealed that liming
the soil with Ca(OH)2 to raise the soil pH to 6.5 decreasedsoil solution phosphate, dilute acid-fluoride-extractablephosphate, and total phosphate uptake by subclover plants.
VOL. 53, 1987
on April 2, 2020 by guest
http://aem.asm
.org/D
ownloaded from
2092 ALMENDRAS AND BOTTOMLEY
TABLE 1. Nodule occupancy by indigenous serogroups ofR. trifolii on T. subterraneum L. cv. Mt. Barker as affected
by source of lime and phosphate applicationsa
Soil % Occupancy by serogroupc:treatmentb 6 36 27 6M 36M Othersd
None 50.0 15.0 17.0 12.0 12.0 35.0CaCO3 13.0 52.0 18.0 13.0 17.0 34.0Ca(OH)2 16.5 32.8 20.0 12.0 7.0 52.8MgO 29.0 46.0 25.0 18.0 19.0 29.0K2CO3 27.0 47.3 26.0 19.0 18.0 31.0KH2PO4 49.3 45.3 28.3 32.3 30.0 23.5
a The least significant difference at P = 0.05 was 23.9 for serogroup 6 and13.9 for serogroup 36. For the others, there was no significant difference at P- 0.05.
b Soil pH values immediately prior to sowing the seed were 5.2 ± 0.2, 5.30.2, and 6.6 ± 0.2 for unamended, phosphate-amended, and lime-amended
soils, respectively." Mean of four replications per treatment; percentages reported for
serogroups 6, 27. and 36 include both singly occupied and coinhabitednodules; percentages for 6M represent nodules occupied by combinations ofserogrotip 6 with serogroups 27 and 36; percentages for 36M represent nodulesoccupied by combinations of serogroup 36 with serogroups 6 and 27.
d Nodules not reacting with fluorescent antibody conjugates specific forserogroups 6, 27, and 36.
An experiment was conducted to determine whether liminghad influenced the distribution of soil phosphate in itsvarious forms during the preplant period. Three levels ofCa(OH)2 (0, 1, and 4 g kg-') were used, which resulted in soilpH values of 5.0 + 0.1, 5.8 + 0.1, and 6.4 ± 0.1 respectively,after the 20-day equilibration period. Subsamples were takenfrom each treated sample, and Pi and organic phosphatewere fractionated sequentially by the procedure of Hedley etal. (20), with the minor modification that concentrated HCIwas used in place of concentrated H2SO4 for digesting theresidual phosphate fraction.
RESULTS
Experiment 1. Regardless of the composition of the mate-rial, liming significantly (P = 0.05) increased the percentnodule occupancy by serogroup 36 (Table 1). Percent occu-
pancy by serogroup 6 was lower after all lime treatmentsthan in unamended soil, yet the decrease was significant (P= 0.05) only in the calcium-containing treatments. Limingdid not influence significantly the nodule occupancy byindigenous serogroup 27. A phosphate application (25 mg ofP kg-'), which did not raise the soil pH, also significantly (P= 0.05) increased the percent nodule occupancy byserogroup 36, but had no effect on nodule occupancy byserogroup 6. The number of nodules cooccupied by mem-
bers of both serogroups 6 and 36 also increased substantially(12 to 32%) as a result of the phosphate application. Neitherthe numbers of nodules formed (ranging between 21 and 42per plant) nor the shoot dry weight yields of the plants were
significantly affected by the treatments under these growthconditions (data not shown).The population densities of the serogroups were evaluated
in postharvest samples of root-associated soil. Although thepopulations of serogroups 6 and 36 were similar (104 g-1), inall cases serogroup 36 outnumbered serogroup 6 by 2.5- to6-fold (Table 2). Significant differences (P = 0.05) betweenthe populations of serogroups 6 and 36 were observed inunamended soil and soil treated with Ca(OH)2, K2CO3, or
KH2PO4. Although there was a nonsignificant trend for thepopulation densities of both serogroups 6 and 36 to be two-to threefold greater in the amended soils regardless of
TABLE 2. Population densities of indigenous serogroups ofR. trifolii in soil after nodule harvest
Population density (101 cells g of dry soil-')Soil treatment
of serogroupa:
36 6 27
None 0.14a 0.05b 0.01bcCaCO3 0.29a 0.05ab 0.01bCa(OH)2 0.42a 0.18b 0.02bcMgO 0.33a 0.08a 0.02aK2CO3 0.43a 0.12b 0.01cKH2PO4 0.36a 0.09b 0.01c
a Mean population of four replicates for each treatment, determined byimmunofluorescenqe; numbers for the same soil treatment that are notfollowed by the same letters differ significantly at P = 0.05 according to t testsfor paired comparisons made between serogroups 6 and 36, 6 and 27, and 36and 27.
treatment, no evidence was obtained that the soil treatmentsresulted in selective proliferation of one serogroup over theother. The direct counts of the populations in soil werecomparable to those obtained by evaluating nodule occu-pancy of final positive dilutions of plant infection-soil dilu-tion tests (data not shown). These data suggest that cells ofboth serogroups 6 and 36 observed by immunofluorescencewere indeed viable, possessed nodulating capability, andmade up similar and significant portions of the total R. trifoliipopulation in soil.Experiment 2. Application of 25 mg of P kg-' immediately
prior to planting resulted in a significant increase (P = 0.05)in percent nodule occupancy by serogroup 36 but had noinfluence on the nodule occupancy by serogroup 6 (Table 3).Occupancy by the latter was overwhelmingly dominant inthis particular experiment. No significant differences wereobserved between the percentages of nodules occupied byserogroup 36 at any of the three concentrations of phosphateapplied. The percentage of nodules co-occupied by sero-groups 6 and 36 was significantly higher (P = 0.05) in the soilreceiving Mo and S than in unamended soil and was furtherincreased (P = 0.05) by phosphate application.
Shoot dry weights were increased significantly (P = 0.05)by the combined application of Mo and S, but no furtheryield increase was measured in response to phosphate ap-plication (Table 4). The improved yield is assumed to be dueto the Mo application. Although the latter was not measured
TABLE 3. Nodule occupancy of T. subterraneum cv. Mt. Barkeras affected by applications of molybdate, sulfate, and phosphate
% Occupancy by serogroupb,cSoil treatment"
6 36 Md Otherse
None 80.0 18.0 15.0 17.0Mo + S 91.0 29.0 27.0 6.0Mo + S + 25 mg of P kg-' 81.5 46.5 39.3 10.1Mo + S + 55 mg of P kg-' 92.0 54.0 48.0 4.0Mo + S + 115 mg of P kg-' 94.0 45.0 45.0 6.0
a Mo, S, and P applied immediately prior to sowing seeds; Mo and S wereused at 1 and 22 mg kg of soil-', respectively.
b Percentages reported for serogroups 6 and 36 include both the singlyoccupied and coinhabited nodules. Results are the mean of four replicationsfor each treatment.cThe least significant difference at P = 0.05 was 12.8, 9.6, and 8.5 for
groups 36, M, and Others, respectively. For group 6, there was no significantdifference at P = 0.05.
d Nodules co-occupied by serogroups 6 and 36.e Nodules not reacting with fluorescent antibody conjugates specific for
serogroups 6 and 36.
APPL. ENVIRON. MICROBIOL.
on April 2, 2020 by guest
http://aem.asm
.org/D
ownloaded from
LIME AND PHOSPHATE EFFECTS ON R. TRIFOLII 2093
TABLE 4. Dry matter yield and nutrient concentration and uptake of T. subterraneum cv. Mt. Barkeras affected by molybdate, sulfate, and phosphate
Patatoltamta Dry wt Nutrient concn (mg g-)b Nutrient uptake (mg potUl)bPlant part Soil treatmentgpoY _1(g pot') N P K S N P K S
Shoot None 2.0 31.8 2.9 19.9 2.0 62.1 5.7 38.9 3.8Mo + S 2.6 36.0 2.3 20.4 2.0 94.2 5.9 53.5 5.1Mo + S + 25 mg of P kg-' 2.6 32.9 2.4 21.5 2.1 86.4 6.3 57.6 5.3Mo + S + 55 mg of P kg-' 2.6 34.6 2.9 21.2 2.1 89.6 7.6 56.0 5.7Mo + S + 115 mg of P kg-' 3.1 36.2 3.4 23.2 2.6 108.7 10.2 70.2 7.9
LSDo.05c 0.6 NSd 0.7 NS NDe 16.4 2.3 15.1 ND
Root None 0.24 23.7 2.5 11.5 NMf 5.9 0.6 2.8 NMMo + S 0.43 27.0 1.5 13.2 NM 11.6 0.6 5.7 NMMo + S + 25 mg of P kg-' 0.31 23.0 1.4 11.2 NM 7.4 0.5 3.5 NMMo + S + 55 mg of P kg-' 0.37 24.7 1.7 14.4 NM 9.4 0.7 5.3 NMMo + S + 115 mg of P kg-' 0.42 28.1 3.9 19.2 NM 11.6 1.6 8.1 NM
LSDo.05 NS NS 1.0 ND NS 0.5 NDa Mo, S, and P were applied as described in Table 3.b Mean of four replications for each treatment, except that K and S were analyzed on composite samples.c LSDo.05, least significant difference at P = 0.05.d NS, Not significantly different at P = 0.05.e ND, Statistical analyses not determined.f NM, Values not measured.
in plant tissues, S and K concentrations in plant tissue werenot affected by the Mo and S treatment, and the total uptakeof K and S was directly proportional to the yield increasesobserved in response to this particular treatment. Significantincreases in shoot and root phosphate uptake were observedonly when 115 mg of P kg of soil-' was applied. Thephosphate concentrations in shoot and root tissues in all theother treated soils ranged between 2.3 to 2.9 mg g-1 and 1.4to 2.5 mg g-1. These values are above the levels consideredto be critical for subclover growth.
Experiment 3. Application of a combination of Ca(OH)2and 25 mg of P kg of soil-' as KH2PO4 resulted in the sameincrease in nodule occupancy by serogroup 36 as that whenboth materials were applied separately (Table 5). Lime andphosphate in combination, however, prevented the reducedoccupancy by serogroup 6 which was observed in thepresence of Ca(OH)2 alone. The incidence of co-occupancy
TABLE 5. Nodule occupancy of T. subterraneum cv. Mt. Barkeras affected by single and combined applications of lime
and phosphate
% Occupancy by serogroupb.cSoil treatmenta
6 36 Md Otherse
Unlimed, -P 44 19 8 45Unlimed, + P 34 38 15 41Limed, -P 25 36 8 47Limed, + P 35 38 15 43
a Soil pH values immediately prior to sowing the seeds were 5.2 0.1 and6.4 ± 0.1 for unlimed and limed soils respectively; phosphate at 25 mg kg ofsoil-' was applied immediately prior to sowing; Mo, and S were applied to alltreated samples, just prior to sowing, at 1 and 22 mg kg of soil-' respectively.
b Percentages for serogroups 6 and 36 include both singly occupied andcoinhabited nodules.
c The least significant difference at P = 0.05 was 10.9, 11.6, and 1.8 forgroups 6, 36, and M, respectively. For the others, there was no significantdifference at P = 0.05.
d Nodules co-occupied by serogroups 6 and 36.e Nodules not reacting with fluorescent antibody conjugates specific for
serogroups 6 and 36.
by both serogroups 6 and 36 increased significantly (P =0.05) as a result of phosphate application and was indepen-dent of the presence or absence of lime. The direct counts ofboth serogroups 6 and 36 in the soil during the 20-dayequilibration period prior to planting revealed that a three-fold proliferation of both serogroups 6 and 36 had occurredduring the incubation period regardless of soil treatment orthe outcome of nodulation (Table 6). As reported forpostharvest samples, serogroup 36 populations consistentlyoutnumbered those of serogroup 6.No evidence was obtained from analysis of either plant
growth or nutrient content to suggest that the commonstimulation by lime and phosphate of occupancy by sero-group 36 was linked with lime stimulation of soil phosphateuptake by the plant. On the contrary, the data showed thatlime decreased the availability of both soil and fertilizerphosphate to the plant. A nonsignificant increase in shootdry weight of 26% in response to phosphate was accompa-
TABLE 6. Population sizes of R. trifolii serogroups 6 and 36during soil incubation after liming and prior to planting
Population density(105 cells g of dry
Time (days) Soil treatment soil-') of serogroupa:
6 36
0 Controlb 0.24 0.46
7 Unamendedc 0.34 0.55Limedd 0.36 0.58
20 Unamended 0.85 1.25Limed 0.82 1.21
a Mean of three replications for control at day zero; mean of two replica-tions per treatment at 7 and 20 days.
b Unamended soil prior to incubation.c Unamended soil maintained at 33 kPa water potential and at 21 ± 2°C.d Soil limed with Ca(OH)2 and maintained identically to the unamended
treatment. Soil pH values were 6.6 ± 0.1 and 6.4 ± 0.1 at 7 and 20 days afterliming, respectively.
VOL. 53, 1987
on April 2, 2020 by guest
http://aem.asm
.org/D
ownloaded from
2094 ALMENDRAS AND BOTTOMLEY
TABLE 7. Dry matter yield, nutrient concentration, and uptakeof T. subterraneum cv. Mt. Barker as affected by single and
combined applications of lime and phosphate
Nutrient NutrientPlant Dry wt concn (mg uptake (mgpart Soil treatmenta~ (g pot-') gI)b pot I)b
N P N P
Shoot Unlimed, -P 1.6 37.2 2.4 59.9 3.9Unlimed, +P 2.0 35.9 2.9 72.9 5.8Limed, -P 1.0 43.1 2.8 44.0 2.9Limed, + P 1.3 37.5 2.6 48.2 3.4
LSDo.05' 0.5 NS NS NS 1.9
Root Unlimed, -P 0.32 22.7 1.9 7.3 0.6Unlimed, +P 0.36 25.3 2.2 9.1 0.7Limed, -P 0.31 31.1 2.5 9.7 0.8Limed, +P 0.38 28.6 2.6 11.0 1.0
LSDo.05 NS NS NS NS NSa Soil pH values immediately before seeds were sown were 5.2 ± 0.1 and
6.4 ± 0.1 for unlimed and limed soils, respectively; 25 mg of P kg of soil'- wasapplied immediately prior to sowing; Mo and S were applied to all treatedsamples, just prior to sowing, at 1 and 22 mg kg of soil-', respectively.
b Mean of four replications for each treatment.c LSDo.05, Least significant difference at P = 0.05.d NS, Not significantly different at P = 0.05.
nied by a significant (P = 0.05) suppression of shoot dryweight in response to liming (Table 7). Although phosphateuptake in the shoots of plants in the unlimed soil treated withphosphate was significantly (P = 0.05) greater than in theunlimed control, liming substantially suppressed the uptakeof soil phosphate and significantly (P = 0.05) suppressed theuptake of fertilizer phosphate by the plants. These observa-tions correlated with the effects of the soil treatments uponthe values of Bray solution (0.03 M NH4F in 0.025 MHCl)-extractable Pi and soil solution Pi obtained from soilsamples taken after 20 days of equilibration with Ca(OH)2and throughout the first 15-day period after phosphate addi-
tion. Liming reduced the level of extractable and soil solu-tion Pi from 1 to 0.6 mg of P kg-' and from 0.7 to 0.3 ,uM,respectively. In contrast, addition of 25 mg of P kg-' raisedextractable and soil solution Pi to 1.8 mg of P kg-' and 1 ,uM,respectively throughout the first 15-day period after phos-phate addition. Levels of extractable and soil solution Pi inlimed soil receiving 25 mg of P kg-' were unchanged fromthose in the unlimed soil receiving no phosphate. The levelsof both 1 M KCl-extractable and soil solution aluminum inunamended soil were low (100 ,umol 100 g-1 and 10 to 20 F.M,respectively), with phosphate addition having a negligibleeffect on both of these values. Liming reduced the extract-able and soil solution aluminum levels below detection limitsregardless of phosphate addition.Experiment 4. Although there was no evidence that lime
enhanced the availability of soil phosphate for plant growth,lime did affect the distribution of soil phosphate between theinorganic and organic fractions in the preplant soil samples.Regardless of concentration, liming increased the percentageof organic phosphate, with a concomitant decrease observedin the Pi fraction (Table 8). The major portion of the decreasein Pi was accounted for in the residual phosphate fractionextracted with concentrated acid, whereas the major in-crease in the organic phosphate fraction was observed in the0.1 M NaOH-extractable phosphate fraction.
DISCUSSIONThe data presented in this study expand upon previous
findings on the effect of CaCO3 on nodule occupancy ofsubclover by indigenous R. trifolii (16, 17). From thosestudies it was hypothesized that the improved nodulation byindigenous serogroup 36 as a result of neutralizing soilacidity could have been due to selective proliferation ofserogroup 36 organisms from a lower population base in theacid soil than serogroup 6. Circumstantial evidence to sup-port such a hypothesis has been documented (see Introduc-tion). The results of this study, however, provided noevidence for this possibility. Serogroup 36 outnumberedserogroup 6 in unlimed and limed soil, independent ofsampling time and nodulating success. Although the relative
TABLE 8. Distribution of phosphate in samples of unlimed and limed Abiqua soila
Phosphate fraction Amt of phosphate (mg kg-l)b at lime levels (g kg-') of: % Change" due to0 1 4 lime application
Inorganic0.5 M NaHCO3 17.4 (1.3) 16.2 (1.2) 18.9 (1.4) 00.1 M NaOH 304.1 (22.5) 318.3 (22.9) 310.6 (23.0) +0.51 M HCI 124.6 (9.2) 120.4 (8.7) 117.3 (8.7) -0.5Residual 419.6 (31.1) 397.2 (28.6) 387.3 (28.7) -2.5
Total 865.7 (64.2) 852.1 (61.3) d 834.1 (61.9)*d -2.6
Organic0.5 M NaHCO3 44.8 (3.3) 37.3 (2.7) 39.3 (2.9) -0.50.1 M NaOH 299.3 (22.2) 362.9 (26.1) 322.7 (23.9) +2.8Residual 138.9 (10.3) 136.8 (9.9) 151.5 (11.3) +0.3
Total organic 483.0 (35.8) 537.0 (38.7)** 513.5 (38.1)* +2.6
Total phosphate 1,348.7 1,389.1 1,347.6
a See Materials and Methods for details of soil treatments.b Values in parentheses represent the percentage of total soil phosphate.cDifference in percentage between the unlimed and the mean of the two limed samples.d ** and *, Significantly different from the unlimed treatment at P = 0.05 and P = 0.1, respectively.
APPL. ENVIRON. MICROBIOL.
on April 2, 2020 by guest
http://aem.asm
.org/D
ownloaded from
LIME AND PHOSPHATE EFFECTS ON R. TRIFOL1I 2095
numbers of serogroups 6, 27, and 36 bore no resemblance tothe outcome of nodulation, we can infer, since the popula-tion densities were significantly different, that the data revealpreliminary information about differences in saprophyticcompetence within the indigenous population. Although theterm saprophytic competence was introduced into Rhizo-bium research almost 20 years ago (9), little evidence hasbeen forthcoming from nonsterile-soil studies to suggest thatstrains of R. trifolii do indeed differ in this respect and thatsuch differences are important to nodulation success orfailure.The increase in the number of nodules occupied by
serogroup 36 as a result of either lime or phosphate additionis noteworthy. Two hypotheses were attractive, since ifproven, they would have linked to this phenomenon con-cepts established about the chemistry of phosphate in acidsoils and other soil-acidity-related Rhizobium research. Thefirst is that liming the acid soil simulated phosphate additionby increasing the availability of soil phosphate to either thesoil microorganisms or the subclover plant or both (18, 21).The second is that both phosphate and lime were acting in asimilar manner by reducing available soil aluminum to a levelnontoxic to either members of serogroup 36 or the host plant(46, 47, 52-55). The results of this study revealed that limingdid not enhance soil phosphate uptake by the subclover plantand that phosphate did not reduce the low levels of extract-able and soil solution Al or influence the nontoxic levels ofAl in shoot or root tissues (data not shown). Liming did,however, stimulate a significant transformation of Pi into anorganic phosphate form during the preplant equilibrationperiod. The magnitude of this transformation (ca. 30 to 50 mgof P kg of soil-') is within the limits of values reported in theliterature for phosphate in soil microbial biomass, whichrange between 10 and 100 mg of P kg of soil-' (7, 19, 20, 31,43, 51). Soil biomass has also been shown to increase inresponse to liming (1, 8).Our data support the possibility that liming results in soil
Pi becoming more chemically available for immobilization,with special benefit to members of the soil microbial com-munity previously handicapped by its level of availability.Alternatively, liming releases acidity-related inhibition ofphosphate immobilization by the soil microbial populationwithout necessarily influencing the chemical forms of phos-phate in the soil. In either case, the assumption is that thephosphate status of members of the R. trifolii soil populationis improved as a result of lime application. Benefits to thenodulating capabilities of serogroup 36 organisms subse-quently ensue. Studies carried out under nonsoil conditions,and described in the accompanying paper (26), have shownthat a representative of serogroup 6 has better P1-seques-tering and nodulating abilities than those of a member ofserogroup 36 at low pH and at the low Pi concentrationstypical of Abiqua soil solution (-0.7 ,uM). Furthermore, theacid phosphatase activities of both strains are significantlylower at pH 5.0 than pH 6.5 (26). However, we must temperthese speculations with the fact that the soil population ofserogroup 36 consistently outnumbered that of serogroup 6regardless of soil treatment.The response of serogroup 6 nodule occupancy to soil
amendments was more complex than that of serogroup 36.Although Ca(OH)2 and CaCO3 reduced the serogroup 6nodule occupancy, thereby confirming our previous findings(16, 17), non-calcium-containing liming materials did notreduce the serogroup 6 occupancy significantly. Moreover,although the addition of phosphate alone had no effect onserogroup 6 nodule occupancy, the addition of phosphate in
combination with Ca(OH)2 counteracted the negative effectof the latter material in reducing occupancy. Exactly howphosphate counteracts the effect of Ca(OH)2 is not known,nor can a special role for the cations associated with eitherthe liming materials or the phosphate be ignored. The resultshave provided preliminary evidence for interactions betweenpH, cations, and phosphate on nodule occupancy by mem-bers of indigenous serogroup 6 which were not observedwith indigenous serogroup 36. On the basis of these differ-ences in nodulation responses by different members of anindigenous population of rhizobia, we can only wonder howinoculant strains can ever be introduced into competitivesituations with predictable success guaranteed.
In this study the conditions which stimulated occupancyby serogroup 36, but which did not reduce occupancy byserogroup 6 (i.e., phosphate alone or lime plus phosphate),resulted in co-occupation of significant proportions of nod-ules by both serogroups. These findings support other recentfindings that co-occupancy can be substantial in nodules ofsoil-grown legumes (13, 30, 38). Obviously, conditions whichimprove the nodulating capability of one group of indigenousorganisms do not necessarily result in exclusion of anothergroup from nodules. Increased incidences of co-occupancyhave been reported with 1:1 mixtures of two strains of R.trifolii as a result of raising the pH on agar-grown plants (25)or of application of lime to field-grown plants (38) of whiteclover (Trifolium repens L. cv. Milkanova). Our findingscertainly illustrate the need for a combination of soil chem-ical and microbiological expertise to elucidate the factorsinfluencing nodulation and growth performance of soil-grown legumes.
ACKNOWLEDGMENTS
This research was supported by funds from the U.S. Departmentof Agriculture (Science and Education competitive grants 81-CRCR-1-0726 and 85-CRCR-1-1704) and the Oregon Agricultural Experi-ment Station. A.S.A. acknowledges partial financial support from aWorld Bank Scholarship.We thank K.-T. Leung, D. H. Demezas, and M. H. Dughri for
technical assistance, D. Hanson and R. Dick for assistance withplant analysis, and C. Pelroy for typing the manuscript.
LITERATURE CITED
1. Adams, T. M., and S. N. Adams. 1983. The effects of liming andsoil pH on carbon and nitrogen contained in the soil biomass. J.Agric. Sci. 101:553-558.
2. Anderson, A. J., and D. V. Moye. 1952. Lime and molybdenumin clover development on acid soils. Aust. J. Agric. Res. 3:95-110.
3. Anderson, A. J., and D. Spencer. 1948. Lime in relation to clovernodulation at sites on the southern tablelands of New SouthWales. J. Aust. Inst. Agric. Sci. 14:39-41.
4. Barnhisel, R., and P. M. Bertsch. 1982. Aluminum, p. 275-300.In A. L. Page, R. H. Miller, and D. R. Keeney (ed.), Methodsof soil analysis, part 2. Chemical and microbiological properties,2nd ed. American Society of Agronomy, Madison, Wis.
5. Berg, M. G., and E. H. Gardner. 1978. Methods of soil analysisused in the Soil Testing Laboratory at Oregon State University.Special report 321. Oregon State University Agricultural Exper-iment Station, Corvallis, Oreg.
6. Bremner, J. M. 1965. Total nitrogen, p. 1149-1178. In C. A.Black, D. D. Evans, J. L. White, L. E. Ensminger, and F. E.Clark (ed.), Methods of soil analysis, part 2. Chemical andmicrobiological properties, 1st ed. American Society of Agron-omy, Madison, Wis.
7. Brookes, P. C., D. S. Powlson, and D. S. Jenkinson. 1984.Phosphorus in the soil microbial biomass. Soil Biol. Biochem.
VOL. 53, 1987
on April 2, 2020 by guest
http://aem.asm
.org/D
ownloaded from
2096 ALMENDRAS AND BOTTOMLEY
14:319-329.8. Carter, M. R. 1986. Microbial biomass and mineralizable nitro-
gen in solonetzic soils: influence of gypsum and lime amend-ments. Soil Biol. Biochem. 18:531-537.
9. Chatel, D. L., R. M. Greenwood, and C. A. Parker. 1968.Saprophytic competence as an important character in the selec-tion of Rhizobium for inoculation, p. 65-73. In Trans. 9th Int.Congr. Soil Sci., vol. 2. International Society of Soil Scienceand Angus and Robertson, Sydney, Australia.
10. Coventry, D. R., J. R. Hirth, T. G. Reeves, and V. F. Burnett.1985. Growth and nitrogen fixation by subterranean clover inresponse to inoculation, molybdenum application and soilamendment with lime. Soil Biol. Biochem. 17:791-796.
11. Coventry, D. R., J. R. Hirth, T. G. Reeves, and H. R. Jones.1985. Development of populations of Rhizobium trifolii andnodulation of subterranean clover following the cropping phasein crop-pasture rotations in southeastern Australia. Soil Biol.Biochem. 17:17-22.
12. Dawson, M. D., and H. S. Bhella. 1972. Subterranean clover(Trifolium subterraneum L.) yield and nutrient content as influ-enced by soil molybdenum status. Agron. J. 64:308-311.
13. Demezas, D. H., and P. J. Bottomley. 1986. Autecology inrhizospheres and nodulating behavior of indigenous Rhizobiumtrifolii. Appl. Environ. Microbiol. 52:1014-1019.
14. Drlica, D. M., and T. L. Jackson. 1979. Effects of stage ofmaturity on P and S critical levels in subterranean clover.Agron. J. 71:824-828.
15. Dughri, M. H., and P. J. Bottomley. 1983. Complementarymethodologies to delineate the composition of Rhizobium trifoliipopulations in root nodules. Soil. Sci. Soc. Am. J. 47:939-945.
16. Dughri, M. H., and P. J. Bottomley. 1983. Effect of acidity on
the composition of an indigenous soil population of Rhizobiumtrifolii found in nodules of Trifolium subterraneum L. Appl.Environ. Microbiol. 46:1207-1213.
17. Dughri, M. H., and P. J. Bottomley. 1984. Effect of acidity on
the composition of an indigenous soil population of Rhizobiumtrifolii in nodules of different cultivars of Trifolium subter-raneum L. Soil. Biol. Biochem. 16:405-411.
18. Haynes, R. J. 1982. Effects of liming on phosphate availability inacid soils. A critical review. Plant Soil 68:289-308.
19. Hedley, M. J., and J. W. B. Stewart. 1982. Method to measuremicrobial phosphate in soils. Soil Biol. Biochem. 14:377-385.
20. Hedley, M. J., J. W. B. Stewart, and B. S. Chauhan. 1982.Changes in inorganic soil phosphorus fractions induced bycultivation practices and by laboratory incubations. Soil Sci.Soc. Am. J. 46:970-976.
21. Holford, I. C. R. 1985. Effects of lime on yields and phosphateuptake by clover in relation to changes in soil phosphate andrelated characteristics. Aust. J. Soil Res. 23:75-83.
22. Jackson, M. L. 1974. Soil chemical analysis-advanced course.
M. L. Jackson, Department of Soil Science, University ofWisconsin, Madison.
23. Jayman, T. C. Z., and S. Sivasubramaniam. 1974. The use ofascorbic acid to eliminate interference from iron in the aluminonmethod for determining aluminum in plant and soil extracts.Analyst 99:296-301.
24. Johnson, C. M., and H. Nishita. 1952. Microestimation of sulfurin plant materials, soils, and irrigation waters. Anal. Chem. 24:736-742.
25. Jones, D. G., and S. J. Morley. 1981. The effect of pH on hostplant 'preference' for strains of Rhizobium trifolii using fluores-cent ELISA for strain identification. Ann. Appl. Biol. 97:183-190.
26. Leung, K., and P. J. Bottomley. 1987. Influence of phosphate on
the growth and nodulation characteristics of Rhizobium trifolii.Appl. Environ. Microbiol. 53:2098-2105.
27. Loneragan, J. F., and E. J. Dowling. 1958. The interaction ofcalcium and hydrogen ions in the nodulation of subterraneanclover. Aust. J. Agric. Res. 9:464-472.
28. Loneragan, J. F., K. Snowball, and W. J. Simmons. 1968.Response of plants to calcium concentration in solution culture.Aust. J. Agric. Res. 19:845-857.
29. Lowther, W. L., and J. F. Loneragan. 1968. Calcium andnodulation by subterranean clover (Trifolium subterraneum L.).Plant Physiol. 43:1362-1366.
30. May, S. N., and B. B. Bohlool. 1983. Competition amongRhizobium leguminosarum strains for nodulation of lentils(Lens esculenta). Appi. Environ. Microbiol. 44:960-965.
31. McLaughlin, J., A. M. Alston, and J. K. Martin. 1986. Measure-ment of phosphorus in the soil microbial biomass: a modifiedprocedure for field soils. Soil Biol. Biochem. 18:437-443.
32. Mulder, E. G., and W. L. van Veen. 1960. Effect of pH andorganic compounds on nitrogen fixation by red clover. Plant Soil13:91-113.
33. Murphy, J., and J. P. Riley. 1962. A modified single solutionmethod for the determination of phosphate in natural waters.Anal. Chim. Acta 27:31-36.
34. Norris, D. 0. 1959. The role of calcium and magnesium in thenutrition of Rhizobium. Aust. J. Agric. Res. 10:651-697.
35. Olsen, S. R., and L. E. Sommers. 1982. Phosphorus, p. 403-430.In A. L. Page, R. H. Miller, and D. R. Keeney (ed.), Methodsof soil analysis, part 2. Chemical and microbiological properties,2nd ed. American Society of Agronomy, Madison, Wis.
36. O'Toole, M. A., and C. L. Masterson. 1968. Interaction ofcalcium and hydrogen ion concentration on nodulation of whiteclover in peat. Ir. J. Agric. Res. 7:129-131.
37. Petrie, S. E., and T. L. Jackson. 1982. Effects of lime, P and Moapplication on Mo concentration in subclover. Agron. J. 74:1077-1081.
38. Renwick, A., and D. G. Jones. 1986. The manipulation of whiteclover 'host preference' for strains of R. trifolii in an upland soil.Ann. Appl. Biol. 108:291-302.
39. Roseberg, R. J., N. W. Christensen, and T. L. Jackson. 1986.Chloride, soil solution osmotic potential, and soil pH effects onnitrification. Soil Sci. Soc. Am. J. 50:941-945.
40. Rovira, A. D. 1961. Rhizobium numbers in the rhizospheres ofred clover and paspalum in relation to soil treatment andnumbers of bacteria and fungi. Aust. J. Agric. Res. 12:77-83.
41. Russell, P. E., and D. G. Jones. 1975. Immunofluorescencestudies of selection of strains of R. trifolii by S184 white clover(T. repens L.). Plant Soil 42:119-129.
42. Shoemaker, H. E., E. 0. McLean, and P. F. Pratt. 1961. Buffermethods for determining lime requirements of soils with appre-ciable amounts of extractable aluminum. Soil Sci. Soc. Am.Proc. 25:274-277.
43. Sparling, G. P., K. N. Whale, and A. J. Ramsey. 1985. Quanti-fying the contribution from the soil microbial biomass to theextractable P levels of fresh and air-dried soils. Aust. J. SoilRes. 23:613-621.
44. Spencer, D. 1950. The effect of calcium and soil pH on nodula-tion of T. subterraneum L. clover on a yellow podsol. Aust. J.Agric. Res. 1:374-381.
45. Tabatabai, M. A., and J. M. Bremner. 1970. An alkalineoxidation method for determination of total sulfur in soils. SoilSci. Soc. Am. Proc. 34:62-65.
46. Thornton, F. C., and C. B. Davey. 1983. Acid tolerance ofRhizobium trifolii in culture media. Soil Sci. Soc. Am. J. 47:496-501.
47. Thornton, F. C., and C. B. Davey. 1983. Response of theclover-Rhizobium symbiosis to soil acidity and Rhizobiumstrain. Agron. J. 75:557-560.
48. Thurman, N. P., D. M. Lewis, and D. G. Jones. 1985. Therelationship of plasmid number to growth, acid tolerance andsymbiotic efficiency in isolates of Rhizobium trifolii. J. Appl.Bacteriol. 58:1-6.
49. Vincent, J. M. 1970. A manual for the practical study ofroot-nodule bacteria. International Biological Programme hand-book no. 15. Blackwell Scientific Publications, Oxford.
50. Vincent, J. M., and L. M. Waters. 1954. The root-nodulebacteria as factors in clover establishment in the red basalticsoils of the Lismore district. New South Wales. II. Survival andsuccess of inocula in laboratory trials. Aust. J. Agric. Res.5:61-76.
51. West, A. W., G. P. Sparling, and W. D. Grant. 1986. Correlationbetween four methods to estimate total microbial biomass in
APPL. ENVIRON. MICROBIOL.
on April 2, 2020 by guest
http://aem.asm
.org/D
ownloaded from
LIME AND PHOSPHATE EFFECTS ON R. TRIFOLII 2097
stored, air-dried and glucose-amended soils. Soil Biol. Bio-chem. 18:569-576.
52. Wood, M., and J. E. Cooper. 1984. Aluminum toxicity andmultiplication of Rhizobium trifolii in a defined growth medium.Soil Biol. Biochem. 16:571-576.
53. Wood, M., and J. E. Cooper. 1985. Screening clover and Lotusrhizobia for tolerance of acidity and aluminum. Soil Biol.
Biochem. 17:493-497.54. Wood, M., J. E. Cooper, and A. J. Holding. 1984. Soil acidity
factors and nodulation of Trifolium repens. Plant Soil 78:367-379.
55. Wood, M., J. E. Cooper, and A. J. Holding. 1984. Aluminumtoxicity and nodulation of Trifolium repens. Plant Soil 78:381-391.
VOL. 53, 1987
on April 2, 2020 by guest
http://aem.asm
.org/D
ownloaded from
Recommended
