Tolerances to Soil Acidity Factors Among Rhizobia

INTRODUCTION AND LITERATURE REVIEW

This thesis reports on the effects of certain soil acidity factors upon the growth and symbiotic performance of slow growing strains of the symbiotic, nitrogen‑fixing bacteria Rhizobium.

The natural, nonsymbiotic habitat of rhizobia is the soil, and the immediate environment is that of the soil solution. In an acid soil, besides the high hydrogen ion activity in solution, there are several mineral elements whose pH‑dependent solubility and reaction with the soil matrix can produce relative toxic or deficient growth effects on higher plants (Pearson and Adams 1967; Pearson 1975; Kamprath 1972). Presumably these same factors could be stressful in rhizobia also, though much less information is available on this (Munns 1977b; 1978; Rerkasem 1977; Carvalho 1978).

The most commonly reported mineral elements producing these effects in acid soils are calcium, magnesium, phosphorus, and molybdenum in the deficiency category, and aluminum, manganese, and hydrogen‑ion in the toxicity category (Pearson and Adams 1967; Pearson 1975; Munns 1978). Though any one of these soil acidity factors alone can produce a biological stress, in many acid soils several of them occur together, and would then interact in their effects on the plant or microbe (Munns 1977b; Pearson 1975).

The factors chosen for study of their effects on rhizobia reported in this thesis are calcium, phosphorus, manganese, and aluminum, at low pH in pure culture studies, as well as the effects of two acid‑infertile soils.

The Bacterium

The Eubacteriales genus Rhizobium is characterized as gram‑negative, nonspore‑forming, short rods, about 0.5 to 0.9 by 1.2 to 3.0 µM, though it is commonly pleomorphic under adverse stress condtions (Jordon and Allen 1975; Vincent 1962). Members of the genus characteristically invade roots of legumes and induce root nodules to form, and the different strains exhibit host range affinities. The nodule bacteroids are involved in fixing molecular nitrogen into combined forms utilizable by the host plant (Vincent 1970).

The bacteria are chemoorganotrophs with a respiratory metabolism. They are aerobic, and often able to produce excellent growth under 0₂ tensions less than 0.01 atmosphere. The optima for temperature and pH is 25‑25°C and 5.0‑8.5 respectively. Most satisfactory growth is provided by media containing yeast or other plant extracts, and yeast extract‑mineral salts media containing mannitol or glucose are among the most conventional (Jordan and Allen 1975).

The taxonomic division of Rhizobium (Jordon and Allen 1975) is based upon host affinity for nodulation, and two major groups are recognized:

Group l ‑ 2 to 6 peritrichous flagella, rapid growth on yeast extract media.

R. leguminosarum R. trifolii

R. phaseoli R. meliloti

Group 2 ‑ polar or subpolar flagellum, slow growth on yeast extract media.

R. japonicum R. lupins

Another large group of mostly slow growing strains is recognized by many, referred to as the "cowpea rhizobia" or the "cowpea miscellany", as well as a smaller, odd group termed the lotus rhizobia (Vincent 1970, 1974).

The representative preferred hosts for all of the species are compiled elsewhere (Jordon and Allen 1975; Vincent 1970, 1974).

As Vincent (1974) points out, the distinction between the fast and slow growing rhizobia is well established on the basis of growth characteristics, acid production in synthetic media, utilizable carbon source, complement of glycolytic enzymes and internal antigens, as well as evidence from numerical taxonomy.

A Review of the Pertinent Literature

Of all the soil acidity factors, the effects of pH per se on rhizobia has been the most thoroughly researched. Early work on the response of several strains to low pH liquid media revealed that the different species of rhizobia had different critical pH for growth, R. meliloti being the most acid sensitive with a critical pH of 4.9, and the slow growing R. japonicum and R. lupini were the most acid tolerant, having critical pH of 3.3 and 3.15 respectively (Fred and Davenport 1918).

From a study of 8 strains of R. japonicum, Wright (1925) found the critical pH range in liquid media to be 4.1 to 4.5, though 5 of the 8 strains showed net viability decrease at pH less than 4.4 within 2 weeks.

Fairly good agreement with these values was found by Bryan (19.23) who introduced strains of different species into acid soils and found the critical pH for their recovery to be 5.0 for R. meliloti, 4.5‑4.7 for R. trifolii, and 3.5‑3.9 for R. japonicum. Comparing 2 fast growing species for acid tolerance in liquid media, Jensen (1942) confirmed that R. trifolii is more tolerant than R. meliloti, but both were inhibited below pH 5.

The most extensive investigation of the sensitivity to pH among rhizobia was conducted by Graham and Parker (1964) who tested 79 strains in liquid medium and found the general order of tolerance to low pH to be R. meliloti other fast growers slow growers. Of great significance was the demonstration of differential strain tolerances within each species. Their findings, as well as more recent reports (Graham and Hubbell 1975; Broughton et al. 1975) puts the critical pH range for the more tolerant slow growers at 4.3‑4.9, a range more conservative than the earlier studies.

The results from several studies on the effects of acidity as tested in soil give good agreement with artificial media tests concerning the more important observations; the critical pH ranges, the relative tolerances between rhizobium species, and within species the existence of strain to strain differences. Studies on isolates of the slow grower R. japonicum from Iowa soils show that some strains (sero‑groups) are indeed acid tolerant with a lower pH limit for growth of 4.0, while there exist large sero-group differences in adaptation to soils in the 7.5‑8.0 range (Damirgi et al. 1967; Ham et al. 1971). Among the fast growers, R. trifolii is found to persist better in soils at a lower pH than R. meliloti, though they both exhibit increased population levels in more neutral soils (Jones 1966; Jensen 1969; Munns 1965a; Nutman and Ross 1969; Peterson and Gooding 1941). Even the acid sensitive R. meliloti is found to have significant strain differences to survival in acid soil (Munns 1965a; Robson and Loneragan 1970a) so that these authors noted the possible benefits accruing from selection of R. meliloti strains for their performance in acid soils. Also, Lie (1971) found the same strain differentiations occurring at low pH among R. leguminsoraum, and these differences were not predictable from the strain's symbiotic performance at a neutral pH. Related studies of the cowpea type rhizobia show that in this acid tolerant group there also exists strain variation (Norris 1973; Rerkasem 1977).

The general association of the most acid tolerant rhizobia with the slow growth habit was a fundamental tenet in Norris' hypothesis (Norris 1965) that the prototype Rhizobium was a cowpea type organism that evolved in the wet, acid soils of the tropics, along with the tropical legumes, this group also supposedly being the more primitive line of the host plant family. Norris associated a broad host‑range compatibility with this primitive condition of the symbiosis. From tests on over 700 strains, he emphasized that the slow growing strains produced a net alkaline reaction on agar medium and that they were associated with, or isolated from, the most acid tolerant hosts, as well as themselves being the most acid tolerant group of rhizobia, and that there would be an ecological advantage for an alkaline producing strain over an acid producing strain (the fast grower) in an acid soil. Though there are reports giving support for some aspects of his hypothesis (Norris 1973; Brockwell et al. 1966; Rerkasem 1977), Parker (1968) has challenged many theoretical aspects of the hypothesis, and specifically Parker (1971) was able to demonstrate that the relative pH alteration of media is very dependent on its composition and that it would be unreliable to assume that the rhizobia produce similar alterations of pH in the buffered soil itself. Neither slow nor fast growing rhizobia were able to alter the pH of an unmodified soil extract solution, despite good growth, thereby showing no relationship with their respective abilities to change the pH in conventional media. Also, Munns et al. (1978) in a screening of 40 strains of the cowpea miscellany found only a weak association between alkali production in media and symbiotic acid tolerance, alkali production being a very imprecise indicator of acid tolerance. There appears to be little disagreement over the relative acid tolerances of the species of Rhizobium, but there is not enough information available concerning the evolution of the symbiosis or the bacterial mechanisms of acid tolerance to put these together into a unifying concept that relates them all.

Several soil studies have shown that applied rhizobia fail to persist or colonize in acid soils (Vincent and Waters 1954a & b; Vincent 1958; Robson and Loneragan 1970a; Mulder et al. 1966). Related research has shown that large innoculum levels can somewhat replace the beneficial effect of timing (Spencer 1950; Vincent and Waters 1954b; Mulder and Van Veen 1960; Robson and Loneragan 1970a; Rerkasem 1977), this being demonstrated for R. trifolii, R. meliloti, and the cowpea miscellany.

There are reports in the literature that poorly effective, indigenous, strains of rhizobia are more common in acid soils, with the effectiveness increasing in neutral soils (Holding and King 1963; Jones 1966; Jones and Burrows 1969). Also, Van Schreven (1972) demonstrated differences in strain response to the effect of subculturing at low pH on agar media for 300 days on this subsequent effectiveness. Holding and King (1963) suggested that the more ineffective strains found at low pH may be more tolerant of elements such as Mn and Al as compared with more effective strains. While all the above work on the relationship between ineffectiveness and soil acidity has been done with R. trifolii, Munns et al. (1978) working with the cowpea miscellany group found no trend for the symbiotic acid tolerant strains to be reduced in comparative effectiveness.

Abundant literature exists on the effects of acidity on different aspects of the symbiosis as separate from effects on the Rhizobium per se (Vincent 1965; Andrew 1976a; Jensen 1943, 1947; Lowther and Longeragan 1970; Munns 1968), and the reader is directed to excellent, current review on this subject (Andrew 1978; Munns 1977a & b, 1978; Lie 1971, 1974; Rerkasem 1977; Carvalho 1978).

Calcium

The requirement for Ca as an essential nutrient for rhizobia is quite small, as determined in liquid media (Loneragan and Dowling 1958; Norris 1959; Bergersen 1961; Vincent 1962). Vincent (1962) shows that the Ca requirement is about 25 µM for normal growth in liquid culture. In that study he also found no effect of pH down to 5.5 on the response to Ca levels. Humphrey and Vincent (1962) have shown that the Ca is an essential component of the cell wall fraction of the Rhizobium.

While Bergersen (1961) did find differences between strains in growth characteristics in low Ca broth, Norris (1959) reported differences in Ca requirement between fast and slow growing rhizobia when studied in electrodialyzed clay systems. Rerkasem (1977) points out that the differential Ca requirements noted by Norris may only reflect a difference in ability to obtain or retain calcium in those conditions. Nonetheless, she reports that many fast growers responded to increased Ca at pH 3.9, while many of the slow growers studied showed no such response above pH 2.8. Also, Lowther and Longeragan (1970) obtained a positive response to Ca for the fast growing R. trifolii in the rhizosphere of subclover in solution culture at pH 5.0.

Reports are scant on the effects of neutral calcium salts on rhizobia in soil. Albrecht and Davis (1929) reported better survival of rhizobia in soil, as assessed by nodulation on a second crop of soybeans, when CaCl₂ was supplied. Anderson and Maye (1952) suggested that the gradual improvement of modulation on an acid soil might be attributable to the Ca or P applied in the basal superphosphate fertilizer used. However, recent work of Rerkasem (1977) showed no effect of CaS0₄ addition on growth or survival of either a fast or slow growing cowpea rhizobium in soil at pH 4.5.

The low Ca requirement for rhizobia appears below that needed for adequate nodule initiation and function, and general symbiotic growth, those aspects of the slymbiosis as affected by Ca being reviewed elsewhere (Munns 1977a & b, 1978; Andrew 1978; Robson 1978). It appears then that a deficiency of Ca in acid soil would probably have deleterious effects on modulation and legume growth before it would limit rhizobial growth and survival, but there is insufficient data on important rhizosphere effects to make this a definite conclusion (Munns 1977b).

Phosphorus

Phosphorus as a soil acidity factor is important because it is often limiting to plat growth, being available at low levels in acid soils (Hsu 1965; Fox 1978; Kamprath 1973; Sanchez 1976). Though it is accepted as an essential nutrient for rhizobia, the quantitative requirement for P in the free living state has not been reported. Luria (1960) reports the range of dry weight P content in bacteria to be 2 to 6 percent.

Truesdell (1917) added 3 levels of phosphate, as the Na, K or Ca salt to sterilized soil, and between the 2 highest levels obtained large growth responses with introduced R. meliloti. Kamata (1962) reported that some strains of R. japonicum did not nodulate P‑deficient soybean roots, and this was correlated to the relative P response of the two strains in culture media. Clearly the quantitative response of rhizobia to P in synthetic media and soil needs further elucidation.

In the legume‑Rhizobium symbiosis, phosphorus is required in the ATP functioning in the nitrogen fixation process (Bergersen 1971). There are reports of added P improving nodulation, but no clear indication that these are not simply due to improved host growth (Munns 1977b; Carvalho 1978; Fox 1978; Andrew and Jones 1978). Manganese

In acid soils, manganese toxicity can be a stress factor to plant growth (Pearson 1975; Pearson and Adams 1967; Sanchez 1976). While it is difficult to find data on soil solution concentrations associated with manganese toxicity, Asher and Edwards (1978) say that a value of about 100 µM is sufficient to cause toxicity to leguminous plants, and this is in fair agreement with solution levels found in acid, Mn‑toxic soils (Morris 1948). Studies on the effects of Mn on nonsymbiotic legumes in solution culture show that most species are reduced in growth at manganese concentrations between 100 and 200 µM (Morris and Pierre 1949; Andrew and Regarty 1969).

Wilson and Reisenauer (1970) demonstrated a lower optimum growth range for Mn by rhizobia to be between 0.1 and 1 µM, and found no depressive effect of the highest level, 10 µM.

Two strains of R. phaseoli have been reported to differ in tolerance to Mn at very high levels, up to 400 ppm in culture media (Dobereiner 1966). However, when these same strains were used as inocula in acid soil with manganese added to inhibitory level, they displayed different effects on reduction in nodule numbers or in nitrogen fixation per unit nodule weight, indicating Mn toxicity effects are different for bacterial growth than for the symbiotic relationships. The author also stressed the importance of selecting rhizobia for their tolerance to soil acidity when selecting inocula for such problem soils.

Masterson (1968) found that R. trifolii isolates from mineral acid soils were less effective on their host than isolates from acid peat soils, implying that the more tolerant strains were less effective. He suggested the peat may provide some protection from high levels of Mn, Al or Fe found at low pH. He also reported data of Sherwood where culturing strains of R. trifolii on media with 800 to 1500 ppm Mn showed little effect, and in some cases raised the effectiveness. Related to these observations are those of Holding and Lower (1971) showing that high levels of Mn (16 mm) in continuous culture can reduce the symbiotic effectiveness of R. trifolii. The effect was transient though, in that good effectiveness was recoverable upon transfer to media low in Mn.

In studying the effects of Mn on nitrogen fixation in Beijerinckia and Azotobacter, Becking (1961) found differential strain tolerance to 20 ppm Mn for Azotobacter, and the Beijerinckia strain was little affected by 40 ppm, except at a low pH of 3.0‑3.2.

The effects of Mn on the symbiotic and nonsymbiotic legume are reported elsewhere (Andrew and Regarty 1969; Robson and Loneragan 1970b; Vose and Jones 1963; Morris and Pierre 1949; Souto and Dobereiner 1969; Dobereiner and Aronovich 1965; Dobereiner et al. 1965), as well as in recent reviews (Andrew 1978; 1976; Munns 1978). A survey of this literature reveals that Mn as a soil acidity factor can be quite limiting to symbiotic legumes, and that the effects could be more clearly understood with more knowledge of the effects of Mn at low pH on the rhizobia.

Aluminum

The high levels of aluminum that occur in soil solutions at low pH can be quite toxic to plant growth, and its abundance on the exchange sites of acid mineral soils make it the most important soil acidity factor (Kamprath 1972; Foy 1976; Pearson 1975; Andrew 1978). The few reports dealing with the effects of Al on rhizobia and other, microorganisms are reviewed here.

The important effects of Al on the functioning of the symbiotic and nonsymbiotic legume are detailed elsewhere (Munns 1965b; Carvalho 1978; Andrew et al. 1973; Foy and Brown 1964; Foy et al. 1969), and discussed in recent reviews (Asher and Edwards 1978; Andrew 1978). The work of Carvalho (1978) alone shows clearly that Al is more detrimental to symbiotic versus nonsymbiotic growth for Stylosanthes, though there was a range of tolerance among the 6 species tested. And since only 1 rhizobial strain was used, it merits further research to see if strains with superior Al-tolerance could be identified and compared with less tolerant strains in their symbiotic performance with a tolerant host.

Much of the work on Al suffers from lack of control of Al concentration and pH (Munns 1965b; Andrew 1978). In studies concerning the levels of P and Al available in nutrient solutions at low pH, Munns (1965b) reported that if P was kept below 50 µM, (1.5 ppm) at pH 4, or below 10 PM at pH 4.5, then Al concentrations on the order of 100 µM (2.7 ppm) could be maintained without reaction with phosphate. Andrew (1978) pointed out that many reports on the effects of Al on plants have utilized treatments that greatly exceeded the solubility limitations in the pH‑P‑Al system, and notes that even at low pH, Al concentrations in soil solutions rarely exceed 4 ppm.

Early research by Whiting (1923) demonstrated that rhizobia were not inhibited in their growth by addition of insoluble Al salts, but were inhibited by Al salts which lowered the pH of the media, though no data was given.

The existence of differential species tolerance to Al among Azotobacter has been shown by Katznelson (1940). In liquid media, with a pH range from 4 to 7, and with Al up to 100 ppm, A. indicum was much more tolerant than A. chroccoccum, the former not being seriously inhibited except at pH 4.0 with 100 ppm Al. The effects of Al at low pH were clear, if not the actual aluminum levels; many of the treatments probably exceeded the solubility of gibbsite.

In a similar study, Becking (1961) investigated effects of Al on the nitrogen fixation (growth) of 2 strains of Beijerinckia and one of Azotobacter in liquid media. The results showed the superior tolerance of Beijerinckia over Azotobacter, although the Al standard solution was neutralized before addition, and the organisms lowered the pH of the medium.

Chlorella pyrenoidosa, a green alga, has been reported by Foy and Gerloss (1972) to be quite tolerant of aluminum, and a strain having yet greater tolerance was obtained by adaptation to increasing Al stress. Again, while the relative effects of A1 were clear, the quantitative levels of Al reported are open to question because some of the treatments would have exceeded the solubility of variscite.

The only direct studies of effectively controlled Al level on rhizobia relate to survival, not growth rate. Rerkasem (1977) has shown that a slow growing strain of cowpea rhizobia that was little affected by introduction to soil with pH 4.5 displayed a decrease in viability when in the same soil acidified to pH 3.9‑4.1 with aluminum sulphate, while a fast cowpea strain was inhibited by pH 4.5 alone. The presence of the rhizosphere environment of the host root enabled both strains to better survive in acid soil. Also, in short term tests on retention of viability (20 minutes) over a wide pH range, she also found that many slow growers were unaffected by low pH or high Al, while the fast growers were inhibited by the acidity itself, with no extra detrimental effect of the Al. The addition of 1 MM Al₂(SO₄)₃caused all 12 strains of the fast growers to flocculate, whereas none of the 12 slow growers were flocculated, the difference between the two groups probably reflecting some basic characteristic as cell surface properties.

A recent study by Carvalho (1978) investigated the effects of Al on the symbioses of 6 Stylosanthes species with the slow growing, cowpea strain C8756. In looking at the effect of Al on the microbial component of the system, she compared the viability of the strain in nutrient solutions at pH 4.5 and with none, 25 or 100 µM Al and found no differences, though all treatments displayed a slow decrease in population density through 8 days.

While the studies of Rerkasem and Carvalho show that some rhizobia can survive at high levels of Al, there is insufficient evidence to establish Al‑tolerance in rhizobia as distinct from acid‑tolerance, and no data concerning effects on their growth rate.

Figure 4 shows the results of adding Al as two different sources on growth of TAL174N. The equal effect of both sources verifies that the inhibitory effect of the Al compounds was due to Al.

Table 3 lists tolerances to the various factor combinations among strains of the cowpea miscellany. The data were combined from detailed sampling trials and the screening trials with growth assessed by visual turbidity. The data are based on 25 days' growth from an average inoculum level of 10^{3.14 +
0.67} cells/ml. With 5 µM P, the average time to reach turbidity was 8.3 days at pH 6.3, 10.2 days at pH 4.5, and 13.4 days at pH 4.5 + 50 µM Al. These values are derived only from the strains that grew in the given medium within 25 days.

Table 4 shows a similar summary for R. japonicum. The time period was again 25 days, and the inoculum level was 10^{3.14 + 0.42} cells/ml. At 5 µM P, the average time to achieve turbidity was 7.0 days at pH 6.3, 11.3 days at pH 4.8, 17.3 days at pH 4.8 + 25 µM Al, and 15 days at pH 4.5 + 50 µM Al. In yeast and control defined media, the tunes were respectively 5.2 and 7.3 days.

DISCUSSION

The data show that Al is a potent stress to the growth of free living rhizobia. Even for tolerant strains, Al reduced growth rate, and often lengthened lag phase. The data do verify recent evidence that some strains of slow‑growing rhizobia can survive high concentrations of Al (Rerkasem 1977; Carvalho 1978); but the large reduction in growth rate shown here could be critical for colonization of soil and rhizosphere, and for induction of modulation (Munns 1968a; Vincent 1974; Dart 1976). Aluminum has been clearly shown to inhibit modulation of Stylosanthes spp. (Carvalho 1978).

The Al concentrations imposed in these trials were chosen from data of displaced solutions from acid soils (Pearson 1975; Pearson and Adams 1967). Also, the 50 µM concentration corresponded to an Al activity of 21.8 µM calculated from the first‑approximation Debye‑Huckel equation (Adams 1974), and this activity is well within the range found in acid soil solutions (Pearson 1975). Thus the 50 µM Al level is a realistic one that rhizobia might encounter in acid soil.

Acidity itself was a severe stress, preventing growth of about one third of the rhizobia. Tolerance of acidity did not necessarily confer tolerance of Al; about half the strains tolerant of pH 4.5 could not tolerate the Al toxicity that would normally be associated with the acidity in soil. Tolerance to both acidity and Al was rated at low (5‑10 µM) P, because the high (1000 µM) P concentration precludes the existence of toxic Al concentration and in any case is only remotely likely to exist in soil.

The low P concentration itself inhibited growth of some strains, but with less severity than acid or Al. According to data from soil studies (Fox et al. 1978), the 5 µM P concentration would not be deficient for plant growth. In fact, most soil solutions contain P at concentrations <1 µM (Reisenauer 1966; Gilman and Bell 1978). Soils, unlike the test media, are buffered with respect to phosphate. Only in laboratory media would rhizobia normally encounter the extremely high concentration of 1000 µM. As Parker et al. (1977) pointed out, conventional bacteriological media, which are absurdly rich, may encourage rhizobial dependence on luxury levels of nutrition. The data here suggest that the large majority of rhizobia could grow adequately at P levels 100 to 1000‑fold more dilute than in common media (Vincent 1970; Bergersen 1961). Routine use of media at more dilute nutrient levels might keep in check any tendency of the bacteria to develop luxury dependence.

Comparison of tolerance categories for the two groups of rhizobia suggest that the cowpea rhizobia have more tolerance to Al. However, there were perhaps too few strains on which to judge R. japonicum. Nonetheless, both groups show similar features. First, within each group there is strain to strain variation in tolerance; second, acid-tolerance and Al‑tolerance are separate, as they are for higher plants (Munns 1965b; Andrew et al. 1973; Carvalho 1978); and third, Al at realistic concentrations appeared to be more commonly a severe stress than low pH or low P at the levels tested here.

This study was limited to slow‑growing rhizobia. Some fast‑growing strains might also have useful acid and Al‑tolerance. Studies from soils acid enough to support moderate concentrations of soluble Al suggest that R. trifolii might contain such strains (Munns 1965a; Jones 1966; Jensen 1969).

The screening procedure of experiment B may prove to be a useful and simple method for detecting tolerant strains. It obviates the need for pH control during growth by the simple expedient of using a small inoculum, so that population density remains too small to raise the pH and precipitate Al until detectable turbidity is approached. Since Parker (1971) has shown that pH change in media by rhizobia is a function of organic composition, an improvement in the screening procedure might be to alter the media to prevent a large pH change. A desirable modification also might be reduction of the EDTA concentration to 10 µM. The response of TAL174N to 50 µM Al was less in the presence of Fe^IIIEDTA than in the presence of FeCl₃ (compare figs 4 and 2a). This may be an effect of EDTA. Ferric EDTA was used because if would not interfere with availability of P, whereas FeCl₃ could exceed the solubility of Fe(OH)₂H₂PO₄ (Norvell 1972). The stability constant of Fe^IIIEDTA so greatly exceeds that of AlEDTA that essentially no Al complex should form at pH 4.5 (Norvell 1972). However, after making sufficient growth in the presence of Al, the rhizobia might separate and absorb enough Fe to allow EDTA to start complexing Al. If so, the use of EDTA would diminish the effects of Al.

The importance of saprophytic competence in rhizobia has been emphasized by Chatel et al. (1968) and Parker et al. (1977). Introduced bacteria often are unable to tolerate biotic or abiotic stresses in a new environment (e.g., Alexander 1971). Identification of strains of Rhizobium having superior tolerance to mineral stresses may be a step towards improving chances of selecting successful inoculants for acid infertile soils. It now needs to be shown if rhizobial tolerances to acidity factors in pure culture relate to their tolerance in soil.

SECTION IV

RELATIONSHIP BETWEEN RHIZOBIAL TOLERANCES IN PURE

CULTURE AND SYMBIOTIC PERFORMANCE IN ACID SOIL

INTRODUCTION

Differential strain tolerance to soil acidity has been demonstrated for different rhizobia species (Munns 1965a; Lie 1971; Rerkasem 1977; Munns et al. 1978). Lie (1971) reported that a R. leguminosarum strain that gave the best symbiotic performance at pH 4.6 was a comparatively mediocre strain at neutral pH. This observation was confirmed by Munns et al. (1978) on several strains of the cowpea miscellany, where effect of acidity on a strain's performance bore no relationship to its effectiveness at higher pH.

The detrimental effects of Al on symbiotic legumes are attributed to reduced nodulation and plant growth (Sartain and Kamprath 1975; Carvalho 1978). For Stylosanthes species, Carvalho (1978) demonstrated that the nitrogen fixation activity of nodules formed in the absence of Al was not reduced upon exposure to levels up to 100 µM, but nodule formation was. Increased nodulation of soybeans was partly attributed to reduced Al saturation in soil from liming, while it was also highly correlated with root Ca level (Sartain and Kamprath 1975). Differential symbiotic performance of rhizobia under conditions of Al stress have not been demonstrated.

This section describes the symbiotic performance in soil of 25 strains of rhizobia on 3 varieties of Vigna unguiculata (cowpea). Comparative effectiveness of a strain at low pH with its effectiveness at a more neutral pH (symbiotic acid tolerance) was then compared with the strains' unknown tolerances to acidity and Al in pure culture, as determined in Sections I and II.

MATERIALS AND METHODS

Two greenhouse pot trials were conducted to test the symbiotic acid tolerance of several rhizobial strains. Three cultivars of the acid tolerant legume V. unguiculata were grown on two soils; one variety on the first soil, and 2 other varieties on the second soil. The treatments for each soil were a low pH of 4.6 and a more neutral pH of 6.0‑6.2. Previous experience with the two soils showed them to be very nitrogen deficient and naturally acid.

Seed of Vigna unguiculata cv. Blackeye 5 (BE5) was obtained from the California Crop Improvement Association, Agronomy Department, University of California, Davis. Seed of varieties TVu 1190 and TVu 4557 were supplied by the International Institute of Tropical Agriculture, Ibadan, Nigeria, by way of the University of Hawaii NifTAL Project, Paia, Maui. The day of planting, seeds were surface sterilized by submergence in 30% H₂0₂ for 5 minutes, and thoroughly rinsed with distilled water.

Twenty‑five strains of rhizobia were tested, and are listed in the Appendix. Thirteen strains were used as inocula for a trial with BE5, and 23 strains with the TVu varieties, with only 2 strains tested on BE5 not also tested on the other 2 varieties. All of the strains were tested for tolerance to acidity and Al in pure culture in Sections I and II. For inoculation, cultures from agar slants were suspended in full nutrient broth (treatment b, Sec. B, Sec. I), diluent, and delivered to seed in pots immediately before it was covered with soil. Viable counts of inoculum showed that 10^{4.9
+ 0.25} cells per seed was applied in the BE5 trial, except for strains TAL163 and TAL98 which, not by design, received 10^6.4 and 10^6.2 cells per seed, respectively. In the trial with the TVu varieties 10^{5.9 ± 0.5} cells/seed was applied. In each trial, a given strain was applied at the level for all treatments with that strain.

The soil used in the trial with the BE5 variety was Goldridge loamy fine sand (Typical Hapludult, fine loamy, mixed, mesic), B‑horizon material collected from Sebastopol, California. It had a saturation paste pH of 4.6, and levels of available cations as listed in Table 1. Mineralogical analysis of the clay fraction showed it contained predominately kaolinite. Two pH treatments were imposed; the unadjusted pH of 4.6 and a pH of 6.0 from the addition of CaCO₃ Limed treatments received 0.9 grams lime per kg soil. Basal nutrients applied to all treatments were 10 mMoles KH₂P0₄, 2 mMoles K₂SO₄, 10 mg Zn and 0.1 mg Mo per kg soil. Nutrients and lime were added to 1.9 kg soil in plastic pots, mixed, watered, and incubated for a week before planting. The final pH of each treatment was checked the day before planting. Combined nitrogen treatments received 12 mMoles NH₄N0₃ per pot, in split applications. A zero control, with no inoculum or nitrogen, was included, and all treatments were replicated. Six seeds were planted per pot, and thinned upon emergence to 3 uniform seedlings.

The soil used in the trial with the TVu varieties was Josephine silt loam (Typic Haploxerult, fine loamy, mixed, mesic), B‑horizon material collected near Georgetown, California. Mineralogical analysis of the clay fraction showed it contained predominately kaolinite and oxides. The saturation paste pH of the soil was 5.4, and available cations are listed in Table 1. Two soil pH treatments were imposed; (i) pH 4.6 by addition of 23.1 me Al as Al₂(SO₄)₃ per kg soil, and (ii) pH 6.2, by addition of 10.6 me C0₃ as CaCO₃ per kg soil. Basal nutrients applied to all pots were the same as for the Goldridge soil, except that 12 mMoles KH₂PO₄ per kg soil was added, and each pot also received 1.9 kg soil. Nutrients and lime were mixed, watered and incubated for 10 days with a final pH check before planting. Combined nitrogen treatments received a total of 16 mMoles NH₄N0₃ in split applications. A zero control was also included. Nine strains, the +N and zero controls were tested in triplicate, and the remaining 14 strains in duplicate. Four seeds of each variety were planted in the same pot, and thinned upon emergence to two plants per variety.

Greenhouse soil temperatures varied between 29°C (day) and 21°C (night). Pots were watered with distilled water, and maintained between 75 to 100% of the determined field capacity for each soil. Observations were taken daily on plant color and relative size of foliage. At 45 days in the trial with BE5, and at 42 days in the trial with the TVu varieties, plants were washed out of the soil, bagged in plastic and refrigerated until examined for nodule number and nodule fresh weight. Shoots were separated and dried at 75°C for weighing.

The time sequence of modulation in both treatments was much the same as for the previous 2 varieties.

Figure 6 summarizes the yield response from the strains as related to nodule number, and in Fig. 7 as related to nodule mass. The trends are much the same as for the 1190 variety; a general yield response with increasing nodule mass and number, these both being reduced by acidity; the greater sensitivity to acidity of the symbioses as compared with the nonsymbiotic condition (+N); the unrelatedness of yield at pH 6.2 to that at 4.6 for a given strain; and the majority of the best yielding strains at low pH were previously determined to be most tolerant of soil acidity factors.

The indigenous rhizobia in the Josephine soil were also effective on this variety at pH 6.2. Analysis of the yield data using Duncan's multiple range test showed that at high pH no strain was significantly better (5% level) than this zero control, and 3 strains were significantly poorer, being then considered ineffective on this variety ‑n15, n12, and TAL174. At low pH, 10 of the strains were significantly better than the zero control, though only 2 gave absolute lower yields than the zero control. The remaining strains were not significantly different from the control.

Figure 8 shows the comparative yields at pH 6.2 and 4.6 for each strain. Ignoring the ineffective strains (n12, n15, and TAL174), those with the least relative yield reduction due to effects of soil acidity are those determined as tolerant of acidity and aluminum. Again, the agreement between tolerances in media and soil is by no means complete; strains M7 and n9, which were acid sensitive in media, yielded better at low pH than such acid‑Al tolerant strains as CB756, TAL169, 172 and 420, while all had about the same effectiveness at pH 6.2.

Table 4 lists the results of association tests between tolerance in media and soil. As was found with variety 1190, the t test shows a significant difference between the 2 populations for net yield of plants at low pH, while the X² test of association of tolerances is not significant.

DISCUSSION

The data presented here show that there is large strain differences among rhizobia of the cowpea miscellany for symbiotic acid tolerance on different varieties of V. unguiculata. Soil analytical data indicate that the main stress in the acid Goldridge soil is that of pH per se, while in the acidified Josephine it is the high level of exchangeable and soluble Al. Nonetheless, strains which gave the best yield on varieties on either soil at low pH had the common property of being the most tolerant of acidity factors as tested in pure culture, especially to Al.

On studies of symbioses of V. radiata (mung bean) in the Goldridge soil, Munns et al. (1978) confirmed that it was not necessary to sacrifice effectiveness to get acid tolerance. This was also found in the present study on this same soil with the BE5 variety of cowpea.

On all 3 varieties, the effects of the soil acidity complex were to generally reduce yield by reducing nodule abundance and mass on the N-deficient soils. The more severe effects of low pH on the symbiotic growth of the TVu varieties than with BE5 is attributed to the high Al level in the Josephine soil. No nitrogen analysis of plant material was done, as it was clear that the growth response of the initially yellow plants on these soils was due to the availability of N through rhizobial association. Also, Munns et al. (1978) have already shown that the yield of symbiotic V. radiata in the Goldridge soil correlated well with %N in the tops. Therefore, the inhibitory effects of soil acidity seen here are attributable directly to effects on the overall performance of the Rhizobium.

The drastic effects of pH and Al on nodulation seen here are in agreement with previous studies (Sartain and Kamprath 2975; Carvalho 1978; Munns et al. 1978). Also, from this present study there is good evidence that nodulation is reduced because growth of some strains of rhizobia is impeded by acidity and/or Al.

The best agreement between strain tolerance in media and in soil is seen with the highest and lowest yielding strains at low pH. Tolerance in media was for survival and growth alone. For symbiotic performance in a stressed soil, other events of modulation or nodule function may be limited by acidity or Al for a giver strain, a limitation not related to the ability to make growth in their presence. This may be why some effective (pH 6) and tolerant (in laboratory media) strains perform poorly in acid soil, despite adequate modulation. Also, some strains classed as acid and Al sensitive may simply survive better in stressed soil than pure culture studies would indicate. Tests of strain survival alone in pH‑Al stressed soil could clear up some of these points.

Spain (1975) has discussed the forage potential of acid, allic soils in South America. He notes that there are several tolerant legumes that have promise for good growth on such soils, provided they are effectively nodulated. If indigenous, effective rhizobia are not present in these soils, any inoculum for these plants would have to have the property of Al-tolerance for adequate saprophytic competence of the bacteria and subsequent host growth on such soils.

This is the first reported investigation of the relationship of rhizobial tolerances to acidity factors in media with that symbiotically in acid soil. Similar work is underway in Colombia (J. Halliday, personal communication) and it does not seem unlikely that with experience better relationships will be found. Even at this early stage of development of such screening of strains for defined tolerances, it appears to be a promising technique for use in selection of inocula for acid, unlimed soils. The value of screening strains for acidity related tolerances in defined laboratory media is seen in that if a decision was made to eliminate Al-sensitive strains for use in acid soil trials, then 60 to 70% of the poorest strains would have been retained for further testing in soil, none of the best yielding strains would have been eliminated, and this is the right kind of error to have for a pre‑screening procedure. Obviously, the most critical test of the validity of such pre‑screening would be its relation with strain performance in field trials.

BIBLIOGRAPHY

Adams, F. 1974. Soil solution. IN: E. W. Carson (ed.) The plant root and its environment. Univ. Press of Virginia, Charlottesville. p. 441‑481.

Albrecht, W. A. and F. L. Davis. 1929. Physiological importance of calcium in legume inoculation. Botanical Gazette 88:310‑321.

Alexander, M. 1971. Microbial ecology. John Wiley & Sons. New York.

Anderson A. J. and D. V. Moye. 1952. Lime and molybdenum in clover development on acid soils. Aust. J. Agric. Res. 3:95‑110.

Andrew, C. S. 1976a. Effect of Ca, pH and N on the growth and chemical composition of some tropical and temperate pasture legumes. I. Nodulation and Growth. Aust. J. Agric. Res. 27:611‑623.

Andrew, C. S. 1976b. Screening tropical legumes for manganese tolerance. IN: M. J. Wright (ed.) Plant adaptation to mineral stress in problem soils. USDA Beltsville, Maryland.

Andrew, C. S. 1978. Mineral characterization of tropical forage legumes. IN: C. S. Andrew and E. J. Kamprath (eds.) Mineral nutrition of legumes in tropical and subtropical soils. CSIRO, Melbourne, Australia. (In press).

Andrew, C. S. and M. P. Hegarty. 1969. Comparative response to Manganese excess of 8 tropical and 4 temperate pasture legume species. Aust. J. Agric. Res. 20:687‑696.

Andrew, C. S., A. D. Johnson and R. L. Sandland. 1973. Effects of aluminum on the growth and chemical composition of some tropical and temperate pasture legumes. Aust. J. Agric. Res. 24:325‑339.

Andrew, C. S. and R. K. Jones. 1978. The phosphorus nutrition of tropical forage legumes. IN: C. S. Andrew and E. J. Kamprath (eds.) Mineral nutrition of legumes in tropical and subtropical soils. CSIRO, Melbourne, Australia. (In press).

Armiger, W. H., C. D. Foy, A. L. Fleeting and B. E. Caldwell. 1968. Differential tolerance of soybean varieties to an acid soil high in exchangeable aluminum. Agron. J. 60:67‑70.

Asher, C. J. and D. G. Edwards. 1978. Relevance of dilute solution culture studies to problems of low fertility tropical soils. IN: C. S. Andrew and E. J. Kamprath (eds.) Mineral nutrition of legumes in tropical and subtropical soils. CSIRO, Melbourne, Australia. (In press).

Baskett, A. C. 1952. The resistance of Bact. lactis aerogenes to proflavine. II. The direct induction of adaptation. Proc. Roy. Soc. London Ser. B 895:251‑262.

Becking, J. H. 1961. Studies on the nitrogen‑fixing bacteria of t.. genus Beijerinckia II. Mineral nutrition and resistance to high levels of certain elements in relation to soil type. Plant and Soil 14:297‑322.

Bergersen, F. J. 1961. Growth of Rhizobium in synthetic media. Aust. J. Biol. Sci. 14:349‑360.

Bergersen, F. J. 1971. Biochemistry of symbiotic nitrogen fixation in legumes. Ann. Rev. Plant Phys. 22:121‑140.

Brock, T. D. 1974. Biology of microorganisms. Prentice‑Hall Inc., Englewood Cliffs, New Jersey.

Brockwell, J., S. K. Auso and G. A. Rea. 1966. Acid production by rhizobia from the genera Trifolium and Lotus. J. Aust. Inst. Ag. Sci. 32:295‑297.

Broughton, W. J., P. Y, Chan, S. Padmanabhan, and K. P. Tan. 1975. Rhizobia in tropical legumes. I. Some characteristics of Malaysian rhizobia. Malaysian Agric. Res. 4:141‑153.

Bryan, U. C. 1923. Effect of acid soils on nodule forming bacteria. Soil Sci. 15:37‑40.

Carvalho, M. M. de. 1978. A comparative study of the responses of six Stylosanthes species to acid soil factors with particular reference to aluminum. Ph.D. Thesis. University of Queensland, Australia.

Chatel, D. L., R. M. Greenwood, and C. A. Parker. 1968. Saprophytic competence as an important character in the selection of Rhizobium for inoculation. Int. Cong. Soil Sci. Trans. 9th II:65‑73.

Damirgi, S. M., L. R. Fredrick, and I. C. Anderson. 1967. Serogroups of Rhizobium japonicum in soybean nodules as affected by soil types. Agron. J. 59:10‑12.

Dart, P. 1977. Infection and development of leguminous nodules. IN: R. W. F. Hardy and W. S. Silver (eds.) A treatise on dinitrogen fixation. Section III: Biology. John Wiley and Sons, New York, p. 367‑472.

Dean, A. C. R, and Sir Cyril Hinshelwood. 1953. Observations on bacterial adaptation. IN: R. Davies and E. F. Gale (eds.) Adaptation in microorganisms. 3rd Symposium of the Soc. for Gen. Micro., Univ. Press, Cambridge, England, p. 21‑45.

Dobereiner, J. 1966. Manganese toxicity effects on modulation and nitrogen fixation of beans in acid soils. Plant and Soil 24:153‑166.

Dobereiner, J and S. Andonovich. 1965. The effects of soil liming and soil temperature on fixation of nitrogen of Centrosema pubescens, Benth., on soil with manganese toxicity. Proc. 9th Int. Grasslands Congress, 2:1121‑1124. (In Portugese, English summary).

Dobereiner, J., N. E. Arruda, and A. de F. Pentado. 1965. Problems of soybean inoculation in acid soil. Proc, of the 9th Int. Grasslands Congress, 2:1153‑1157.(In Portugese, English summary).

Doudoroff, M. 1940. Experiments on the adaptation of Escherichia coli to sodium chloride. J. Gen. Physio. 28:585.

Fox, R. L. 1978. Studies on phosphorus nutrition in the tropics. IN: C. S. Andrew and E. J. Kamprath (eds.) Mineral nutrition of legumes in tropical and subropical soils. CSIRO, Melbourne, Australia, (In press).

Fox, R. L., R. K. Nishimoto, J. R. Thompson and R. S. de la Pena. 1974. Comparative external phosphorus requirements of plants growing in tropical soils. Tran. 10th Int. Con&. Soil Sci., Moscow, IV:232‑239.

Foy, C. D. 1974. Effects of aluminum on plant growth. IN: E. W. Carson (ed.) The plant root and its environment. Univ. Press of Virginia, Charlottesville.

Foy, C. D, and J. C. Brown. 1964. Toxic factors in acid soils: II. Differential A1 tolerance of plant species. Soil Sci. Soc. Am. Proc. 2:27‑32.

Foy, C. D., A. L. Fleming, and W. H. Armiger. 1969. Aluminum tolerance of soybean varieties in relation to calcium nutrition. Apron. J. 61: 505‑511.

Foy, C. D, and G. C. Gerloff. 1972. Response of Chlorella pyrenoidosa to Aluminum at low phi. J. Phycol. 8:268‑271.

Fred, E. B. and A. Davenport. 1918. Influence of reaction on nitrogenassimilating bacteria. J. Ag. Res. 14:3I7‑336.

Frink, C. R. and M. Peech. 1962. Determination of aluminum in soil extracts. Soil Sci. 93:317‑324.

Gillman, G. P. and L. C. Bell. 1978. Soil solution studies on weathered soils from tropical north Queensland. Aust. J. Soil Res. 16:67‑77.

Graham, P. H. and C. A. Parker. 1964. Diagnostic features in the characterization of root nodule bacteria of legumes. Plant and Soil 20:383‑396.

Graham, P. H. and D. H, Hubbell. 1975. Soil‑plant‑Rhizobium interaction in tropical agriculture. IN: E. Bornemisza and A. Alvarado (eds.) Soil management in tropical America. Soil Sci. Dept. N. Carol. State Univ., Raleigh, NC.

Ham, G. E., L. R. Fredrick and I. C. Anderson. 1971. Serogroups of Rhizobium japonicum in soybean nodules sampled in Iowa. Apron. J. 63: 69‑72.

Helyar, K. R. 1978. Effects of aluminum and manganese toxicity on legume growth. IN: C. S. Andrew and E. J. Kamprath (eds.) Mineral nutrition of legumes in tropical and subtropical soils. CSIRO, Melbourne, Aust. (In press).

Holding, A.. J. and J. King. 1963. The effectiveness of indigenous populations of Rhizobium trifolii in relation to sail factors. Plant and Soil 18:191‑198.

Holding, A. J. and J. F. Lowe. 1971. Some effects of acidity and heavy metals on the Rhizobium‑leguminous plant association. IN: T. A. Lie and E. G. Mulder (eds.) Biological nitrogen fixation in natural and agricultural habitats. Plant and Soil Spec. Vol., p. 153‑166.

Hsu, P. H. 1965. Fixation of phosphate by aluminum and iron in acidic soils. Soil Sci. 99:398‑402.

Humphrey, B. A, and J. M. Vincent. 1952. Calcium in cell walls of Rhizobium trifolii. J. Gen. Micro. 29:557‑561.

Jackson, S. and Sir Cyril Hinshelwood. 1959. An investigation of the nature of certain adaptive changes in bacteria. Proc. Roy. Soc. London Ser. B 885:562‑576.

Jensen, H: L. 1942. Nitrogen fixation in leguminous plants. I. General characters of root‑nodule bacteria isolated from species of Medicago and Trifolium in Australia. Proc. Linn. Soc. N.S.W. 67:98‑108.

Jensen, H. L. 1943. Nitrogen fixation in leguminous plants. IV: Influence of reaction on formation of root nodules in Medicago and Trifolium. Proc. Linn. Soc. N.S.W. 68:207‑220.

Jensen, H. L. 1947. Nitrogen fixation in leguminous plants. VII: The N‑fixing activity of root nodule tissue in Medicago and Trifolium. Proc. Linn. Soc. N.S.W. 72:265‑291.

Jensen, H. L. 1969. The distribution of lucerne and clover rhizobia in agricultural soils in Denmark. Tiddskr. Pianteavl. 73:61‑72. (In Danish, English summary).

Jones, D. G. 1966. The contribution of white clover to a mixed upland sward. II. Factors affecting the density and effectiveness of Rhizobium trifolii. Plant and Soil 24:250‑260.

Jones, D. G. and A. C. Burrows. 1969. Acid production and symbiotic effectiveness in Rhizobium trifolii. Sail Biol. Biochem. 1:57‑61.

Jordan, D. C, and 0. N. Allen. 1975. Rhizobiaceae. IN: R. G. Buchanan and N. G. Gibbons (eds.) Bergeys manual of determinative bacteriology. 8th edition. Williams and Wilkins Co., Baltimore, Md., p. 2v1‑2o4.

Kamata, E. 1962. Morphological and physiological studies on nodule formation in leguminous crops. 7. Variation in nodule forming capacity in strains of Rhizobium japonicum Proc. Crop Sci. Soc. Japan 31:78‑82. (In Japanese, English summary).

Kamprath, E. J. 1972. Soil acidity and liming. IN: Soils of the humid tropics. Natl. Acad. Sci., Washington, D. C.

Kamprath, E. J. 1973. Phosphorus. IN: A review of soils research in tropical Latin America. P. A. Sanchez (ed.) N.C. Ag. Ex. Stet. Tech. Bull. No 219.

Katznelson, H. 1940. Survival of Azotobacter in soil. Soil Sci. 49:21‑35.

Lie, T. A: 1971. Symbiotic nitrogen fixation under stress conditions. IN: T. A. Lie and E. G. Mulder (eds.) Biological nitrogen fixation in natural and agricultural habitats. Plant and Soil Spec. Vol., p. 117‑ 127.

Lie, T. A. 1974. Environmental effects on nodulation and symbiotic nitrogen fixation. IN: A. Quispel (ed.) The biology of nitrogen fixation. North‑Holland Publ. Co., p. 555‑582.

Loneragan, J. F. and E. J. Dowsing. 1958. The interaction of Ca and H ions in the nodulation of subterranean clover. Aunt. J. Agric. Res. 9:464472.

Lowther, W. L. and J. F. Loneragan. 1970. Calcium in the nodulation of legumes. Proc, of the 11th Int. Grasslands Cong., p. 446‑450.

Lund, Z. F. 1970. The effect of calcium and its relation to several cations in soybean root growth. Soil Sci. Soc. Amer. Proc. 34:456‑459.

Luria, S. E. 1960. Bacterial protoplasm:composition and organization. IN: T. C. Gunsalus and R. Y. Stanier (eds.) The bacteria. Vol. T9 p. 15. Academic Press, New York.

Masterson, C. L. 1968. Effects of some soil factors on R. trifolii. Traps. 9th Int. Cong. Soil Sci. II:95‑102.

Morris, H. D. 1948. The soluble manganese content of acid soils and its relation to the growth and manganese content of sweet clover and lespedeza. Soil Sci. Soc. Amer. Proc. 13:362‑371.

Morris, H. D. and W. H. Pierre. 1949. Minimum concentrations o£ manganese necessary for injury to various legumes in culture solutions. Apron. J. 43:107‑112.

Mulder, E. G. and W. L. Van Veen. 1960. Effect of pH and organic compounds on nitrogen fixation by red clover. Plant and Soil 13:91‑113.

Mulder, E. G., T. A. Lie, K. Dilz and A. Hauwers. 1966. Effect of pH on symbiotic N fixation of some leguminous plants. 9th Int. Con&. for Microb. Symposia., p. 133‑149.

Munns, D. N. 1965x. Soil acidity and growth of a legume. I. Interaction of lime with N and P on growth of Medicago sativa L. and Trifolium subterraneum L. Aust. J. Agr. Res. 16:773‑741.

Munns, D. N. 1965b. Reactions of Al and P in solution, and effects of Al, P, Ca, and pH on M. sativa and T. subterraneum in solution culture. Aust. J. Agr. Res.16:74‑755.

Munns, D. N. 1968. Modulation of Medicago sativa in solution culture. I. Acid‑sensitive steps. Plant and Soil 2.8:129‑146.

Munns, D. N. 1977a. Mineral nutrition gild the legume symbiosis. IN: R. W. F. Hardy and A. H. Gibson (ells.) Treatise on dinitrogen fixation. Section IV: Agronomy and Ecology. John, Wiley and Sons, New York, p. 353‑392.

Munns, D. N. 1977b. Soil acidity and related matters. IN: J. M. Vincent, A. S. Whitney and J. Bose (eds.) Exploiting the legume‑Rhizobium symbiosis in tropical agriculture. Univ. Hawaii Coll. Trop. Agr. Misc. Publ. 145., p. 211‑236.

Munns, D. N. 1978. Soil acidity and nodulation. IN: C. S. Andrew and E. J. Kamprath (ells.) Mine‑wo.l,nutrition of legumes in tropical and subtropical soils. CSIRO, Melbourne, Aust. (In press).

Munns, D. N., H. H. Keyser, V. W. Fogle, J. S. Hohenberg, T. L. Righetti, D. L. Lauter, M. G. Zaroug, K. L. Clarkin and K. W. Whitacre. 1978. Tolerance of soil acidity in symbioses of Vigna radiata with rhizobia. (Submitted to Agronomy Journal).

Norris, D. 0. 1959. The role of calcium and magnesium in the nutrition of Rhizobium. Aust. J. Agric. Res. 10:651‑698.

Norris, D. 0. 1965. Acid production by Rhizobium: A unifying concept. Plant and Soil 22:143‑166.

Norris, D. 0. 1973. Seed pelleting to improve modulation of tropical and subtropical legumes. 5. The contrasting response to lime pelleting of two Rhizobium strains on Leucaena lecocephla. Aust. 3. Lip. Agr. An. Hus. 13:98‑101.

Norvell, W. A. 1972. Equilibria of metal chelates in soil solution. IN: J. J. Mortvedt, P. M. Giordano, and W. L. Lindsay (ells.) Micronutrients in agriculture. Soil Sci. Soc. AM., Madison, Wisc., p. 115‑138.

Nutman, P. S. and G. J. S. Ross. 1969. Rhizobium in the Rothamsted and Woburn Farms. Rotramsted Lip. Sta. Rep. 2:148‑167.

Parker, C. A. 1968. On the evolution of symbiosis in legumes. Festskrift til Hams Lauritz Jensen. Gadgaard Nielsens Bogtrykkeri, Lemvig, Denmark, p. 107‑116.

Parker, C. A. 1971. The significance of acid and alkali production by rhizobia on laboratory media. Proc. 4th Aust. Legume Modulation Conf. Paper No. 10. Canberra, Australia.

Parker, C. A., M. J. Trinik, and D. L. Chatel. 1977. Rhizobia as soil and rhizosphere inhabitants. IN: R. W. F. Hardy and A. H. Gibson (eds.) Treatise on dinitrogen fixation. Section IV: Agronomy and Ecology. John Wiley and Sons, New York, p. 311‑352.

Pearson, R. W. 1975. Soil acidity and liming in the humid tropics. Cornell International Agric. Bull. 30. Ithaca, New York.

Pearson, R. W. and F. Adams. 1967. Soil acidity and liming. Agronomy No. 12. Amen. Soc. Agron., Madison, Wisc.

Peterson, H. B, and T. H. Gooding. 1941. The geographic distribution of Azotobacter and Rhizobium meliloti in Nebraska soils. Univ. Neb. Coll. Agr. Res. Bull No. 121.

Probert, M. E. 1978. Availability of phosphorus and sulphur in tropical soils in relation to legume growth. IN: C. S. Andrew and E. J. Kamprath (eds.) Mineral nutrition of legumes in tropical and subtropical soils. CSIRO, Melbourne, Aust. (In press).

Reisenauer, H. M. 1966. Mineral nutrients in soil solution. IN: P. L. Altman and D. W. Dittman (ads.) Environmental Biology. Fed. Amer. Soc. Exp. Biol., Bethesda, Maryland, p. 5137‑508.

Rerkasem, B. 1977. Differential sensitivity to soil acidity of legume‑Rhizobium symbioses. Ph.D. Thesis. University of W. Australia.

Robson, A. D. 1978. Mineral nutrients limiting nitrogen fixation in legumes. IN: C. S. Andrew and E. J. Kamprath (eds.) Mineral nutrition of legumes in tropical arid subtropical soils. CSIRO, Melbourne, Aust. (In press).

Robson, A. D. and J. F. Loneragan. 1970a. Nodulation and growth of Medicago truncatula on acid soils. Aust. J. Agric. Res. 21:427‑445.

Robson, A. D. and J. F. Loneragan. 1970b. Sensitivity of annual Medicago species to manganese toxicity as affected by Ca and pH. Aust. J. Agric. Res. 21:223‑232.

Sanchex, P. A. 1976. Properties and management of soils in the tropics. John Wiley and Sons, New York.

Sartain, J. B. and E. J. Kamprath. 1975. Effect of liming a highly Al-saturated soil on the top and root growth and soybean nodulation. Agron. J. 67:507‑510.

Souto, S. M. and J. Dobereiner. 1969. Manganese toxicity in tropical forage legumes. Pesq. Agropec. Bras. 4:129‑138. (In Portugese, English summary).

Spencer, D. 1950. The effect of Ca and soil pH on modulation of T. subterraneum L. clover on a yellow padsol. Aust. J. Agric. Res. 1:374‑381.

Stranier, R. Y. 1953. Adaptation, evolutionary and physiological. IN: R. Davies and E. F. Gale (eds.) Adaptation :in micro‑organisms. arc: Symposium of the Soc. for Gen. Micro., Univ. Press, Cambridge, England, p. 1‑20.

Stanier, R. Y., E. A. Adelberg and J. L. Ingraham. 1976. The microbial world. Prentice‑Hall, Inc. Englewood Cliffs, New Jersey.

Steel, R. G. D. and 3. H. Torrie. 1960. Principles arid procedures of statistics. McGraw‑Hill Book Co., Inc., New York.

Steinborn, J, and R. J. Roughley. 1975. Toxicity of sodium and chloride ions in Rhizobium spp. in broth and peat culture. J. Appl. Bact. 39:133‑138.

Spain, J. M. 1975. Forage potential of allic soils of the humid lowland tropics of Latin America. IN: E. C. Doll and G. 0. Mott (eds.) Tropical forages in livestock production systems. Amer. Soc. Agron. Spec. Publ. 24. Amer. Soc. Agron., Madison, Wisconsin.

Truesdell, H. W. 1917. The effect of phosphorus on alfalfa and alfalfa bacteria. Soil Sci. 3:77‑98.

Van Schreven, D. A. 1972. On the resistance of effectiveness of Rhizobium trifolii to a low pH. Plant and Soil 37:49‑55.

Vincent, J. M. 1958. Survival of the root nodule bacteria. IN: E. G. Hallsworth (ed.) Nutrition of the legumes. Proc. of Univ. of Nottingham 5th Easter School in Ag. Sci. Butterworth Sci. Publ., London, p. 108‑123.

Vincent, J. M. 1962. Influence of calcium and magnesium on growth of Rhizobium. J. Gen. Micr. 28:653‑653.

Vincent, J. M. 1965. Environmental factors in fixation of nitrogen by the legume. IN: W. Bartholomew and F. E. Clark (eds.) Soil nitrogen. Agron. Monograph No. 10. Amen. Soc. Agron., Madison, Wisconsin, p. 384. 435.

Vincent, J. M. 1970. A manual for the practical study of the root‑nodule bacteria. I.B.P. Handbook No. 15. Blackwell Sci. Publ., Oxford, England.

Vincent, J. M. 1974. Root‑nodule symbioses with Rhizobium. IN: A. Quipsel (ed.) The biology of nitrogen fixation. Frontiers of Biology Vol. 33. North‑Holland Publ. Co., Amsterdam, p. 265‑34i.

Vincent, J. M. and L. W. Waters. 1954a. The root‑nodule bacteria as factors in clover establishment in the red basaltic soils of the Lismore District, N.S.W. II. Survival and success of inocula in laboratory trials. Aust. J. Agric. Res. 5:61‑76.

Vincent, J. M. and L. W. Waters. 1954b. The root‑nodule bacteria as factors in clover establishment. in the red basaltic soils of the Lismore District N.S.W. III. Field inoculation trials. Aust. J. Agric. Res. 5:77‑89.

Vose, P. B. and D. G. Jones. 1963. The interaction of manganese and calcium on nodulation and growth in varieties of Trifolium repens. Plant and Soil 18:372‑401.

Watanabe, F. S. and S. R. Olsen. 1965. Test of an ascorbic acid method for determining phosphorus in water and NaHCO₃ extracts from soil. Soil Sci. Soc. Amer. Proc. 29:677‑678.

Werner, D. and K. Berghauser. 1976. Discrimination of R. japonicum, R. lupini, R. trifolii, R. leguminosarum and of bacteroids by uptake of 2‑ketoglutaric acid, glutamic acid and phosphate. Arch. Microb. 107:257‑262.

Whiting, A. L. 1923. Inorganic substances, especially aluminum, in relation to the activities of soil microorganisms. J. Amer. Soc. Agron. 15:277‑289.

Wilson, D. 0, and H. M. Reisanauer. 1970. Effect of manganese and zinc ions on growth of Rhizobium. J. Bact. 102:729‑732.

Wright, W. H. 1925. The nodule bacteria of soybeans. I. Bacteriology of strains. Soil Sci. 20:95‑130.