Indole Acetic Acid Producing And Phosphate Solubilizing Bacteria Native To Kenyan Soils Promote Growth of Common Bean (Phaseolus Vulgaris L.)


 Use of phosphate solubilizing bacteria (PSB) and rhizobia can have a positive effect on the growth of common bean. This study aimed at determining the mechanisms of action of native bacterial strains; and to determine their effect in enhancing growth of common bean. The strains were screened for their ability to solubilize insoluble inorganic phosphates and production of indole acetic acid in vitro. A greenhouse experiment was set up to evaluate the response of common bean to inoculation with selected bacterial strains. Six of the bacterial isolates tested showed a positive result for IAA production. Rhizobium pusense showed the greatest solubilization efficiency of 648 followed by Bacillus megaterium (322.3) and Rhizobium phaseoli (308.7). Inoculation of common bean with Rhizobia and PSB had a significant effect on the number of nodules per plant. The highest shoot biomass was observed when Rhizobium phaseoli was co-inoculated with P. polymyxa (4.3g plant-1) compared to the single Rhizobium phaseoli inoculation (1.14 g plant-1). The shoot tissue nitrogen and phosphorous concentration was increased as a results of co-inoculation up to 32.5% and 75.4% respectively. Therefore, tested bacterial strains have great potential in being formulated and used as biofertilizers that can be evaluated under varying field conditions.


Introduction
Over 200 million people in Sub-Saharan Africa (SSA) depend on common bean as a primary staple food 1 .
The production of common bean (Phaseolus vulgaris L.) is, however, constrained by low soil fertility in many soils leading to a threat to food security. Nitrogen (N) and phosphorus (P) are the most limiting nutrients for plant growth. Phosphorus is generally de cient in some soils due to its ready xation by iron and aluminum oxides 2,3 . The low level of production in SSA has been attributed partly to low levels of soil plant-available P and drought stress, caused by climate change variability 4,5 . Phosphorous is one of the most de cient nutrients for cultivation of common bean 4 because of the high P xing soils 6 . Because common bean requires P to enhance energy for its metabolic activities, the crop possesses high requirements for P and is, hence sensitive to low plant-available P in soils 7 .
Thus, one area of increasing interest is the use of microorganisms which act through a number of mechanisms such as nitrogen xation, solubilization of phosphorous, production of indole acetic acid, cytokinins among others that facilitate nutrient acquisition 8, 9 . Various soil microorganisms produce naturally occurring auxins, indole-3-acetic acid (IAA) as a direct mechanism to promote plant growth 10 . In particular, plant growth promotion and root nodulation are both affected by IAA. Bacterial IAA increases root surface area and length, and thereby provides the plant with greater access to soil nutrients. In addition, bacterial IAA loosens plant cell walls facilitating increased amount of root exudation that provides additional nutrients to support the growth of bene cial rhizosphere bacteria 11 . Most Rhizobium strains that have been examined have been found to produce IAA and several studies have suggested that increases in auxin levels in the host plant are necessary for nodule formation 11 . Nodule bacteria including Rhizobium leguminosarum, R. undicolam, R. etlii, Sinorhizobium meliloti, R. phaseoli, R. pusense among others synthesizes IAA, thereby playing an important role in legume-rhizobia interaction [12][13][14][15][16][17] . It was reported that IAA acts as a signal molecule which is involved in plant signal processing, motility, or attachment of bacteria in root which help in legume-Rhizobium symbiosis 18 .
Soil holds large amounts of phosphate, yet it is found in insoluble form. Phosphate solubilizing bacteria (PSB) are reported to solubilize the phosphate in the soil through acidi cation, chelation, or enzymatically 19 . These microorganisms mineralize organic phosphorus in soil by solubilizing complexstructured phosphates such as tricalcium phosphate to inorganic forms available to plants. Many of the PSB lower the pH of the medium by secretion of organic acids such as acetic, lactic, malic, succinic, oxalic and citric acids 2 . The ability of some microorganisms to convert insoluble phosphates to plantavailable forms is an important characteristic for increasing plant yields 21  Although the role of phosphorus in nodulation, nitrogen fixation and growth of common bean has been reported, the role of phosphate-solubilizing bacteria in phosphorus availability, growth promotion and also their interaction with N 2 -fixing bacteria under tropical conditions requires thorough investigation.
The mechanisms of action of the native Rhizobia and PSB need to be understood in order to select highly effective strains that can be exploited for inoculant production. This study aimed to assess IAA production and phosphate solubilization e ciency of locally isolated endophytic bacteria and to test their e cacy in growth promotion of common beans under greenhouse conditions.

Results
Production of Indole-3 acetic acid (IAA) by Rhizobium species. There was signi cance difference on the concentration of IAA produced by the different Rhizobia isolates. Four out of the ten bacterial isolates tested were considered high IAA producers (>0.85) while the rest were low producers. From the four high IAA producers, R. pusense (Busia) and R. phaseoli (Bungoma) produced higher levels of IAA with absorbance values of 1.33 and 1.14 respectively compared to the other isolates ( Figure 1). Some of the isolates produced negligible amount of IAA ( Figure 1). The colour development showed that the Rhizobia isolates had different ability to produce IAA (Photo 1).
In vitro phosphate solubilization by the bacterial strain. Signi cant (p<0.05) differences in the inorganic phosphate solubilization was observed among the different bacterial isolates (Table 1). Rhizobium pusense (Busia) showed the greatest solubilization e ciency of 648 and consequently the highest solubilization index of 7.3 (Table 1). It was followed by B. megaterium and R. phaseoli with an SE of 322.3 and 308.7 respectively. Based on the classi cation scale of Silva-Filho and Vider (2000), R. pusense, B. megaterium and R. phaseoli (S5) were classi ed as high solubilizers (SI>3.0). Paenibacillus polymyxa, Pseodomonas sp, R. phaseoli (B3), B. aryabattai and B. megaterium were considered medium solubilizers (SI= 2.0-3.0); while the rest of the isolates were low solubilizers (Table 1). Photo 2 shows the solubilization diameter of some of the isolates.

Greenhouse Experiment
Effect of co-inoculation with Rhizobia and PSB on the nodulation of common bean. Inoculation of common bean with Rhizobia and PSB had a signi cant (p≤0.05) effect on the number of nodules per plant. The control treatment did not contain any nodules ( Figure 2). In terms of the effect of coinoculation, the results were varied depending on the speci c Rhizobia-PSB interaction. Some of the interactions led to a synergistic effect on the nodulation while in other treatments, single inoculation with Rhizobia elicited a higher nodulation ( Figure 2). For instance, the co-inoculation of R. pusense with B. aryabhattai (TRT-9) and B. megaterium (TRT-10) led to signi cantly higher number of nodules compared to the single R. pusense (TRT-1); whereas when co-inoculated with P. polymyxa (TRT-8), the nodule numbers are signi cantly lower ( Figure 2). The co-inoculation of R. phaseoli with P. polymyxa (TRT-11) and B. aryabhattai (TRT-12) led to signi cantly high number of nodules than the single R. phaseoli (TRT-2) inoculation. However, there was depressed nodulation when it was co-inoculated with B. megaterium ( Figure 2). Single inoculation with R. pusense (TRT-3) and R. pusense (TRT-4) had signi cantly higher number of nodules compared to the co-inoculation with the PSB. In terms of the nodule dry weight, a similar trend was observed in the inoculation effect as in the number of nodules. Correlation analysis showed a strong positive relationship (R 2 = 0.995; p<0.001) between the number of nodules and the nodule dry weight ( Table 2).
Effect of co-inoculation with Rhizobia and PSB on the shoot and root biomass of common bean. Inoculation of common bean with Rhizobia and PSB generally increased the shoot biomass compared to the control (Figure 3). Speci c Rhizobia-PSB co-inoculation had varied in uence on the shoot biomass of the common bean. Some of the interactions were synergistic while others led to a lower biomass compared to when singly applied ( Figure 3). A positive interaction was observed in most of the interactions except for the R. pusense co-inoculated with either B. aryabhattai or B. megaterium; R. phaseoli (B3) with B. megaterium; and co-inoculation of R. pusense with either P. polymyxa or B. megaterium ( Figure 3). A noticeable positive co-inoculation effect was observed when R. phaseoli (B3) was co-inoculated with P. polymyxa (4.3g plant −1 ) and with B. aryabhattai (3.4g plant −1 ) compared to the single R. phaseoli inoculation with a biomass of 1.14 g plant −1 ( Figure 3). Photo 3 shows a comparison on the above ground biomass of the bean crop between an inoculated and a control pot.
Inoculation of the common bean with the Rhizobia and PSB resulted in a signi cant (p≤0.05) increase in the root dry weight. The non-inoculated (control) treatment had signi cantly lowest root mass (0.9 g plant −1 ) compared to the inoculated treatments except for the single R. pusense (B2) inoculation (0.8 g plant −1 ) ( Figure 4). In terms of the speci c Rhizobia-PSB interactions, co-inoculation led to an increase in root biomass except for the R. pusense (S5) strain ( Figure 4). Co-inoculation of R. pusense (B4) with B.
megaterium resulted in signi cantly highest root biomass of 5.5g plant −1 ( Figure 4). Pearson correlation analysis showed a positive signi cant (p< 0.001) relationship (R 2 = 0.630) between shoot and root biomass ( Table 2). Comparison of the speci c rhizobia-PSB co-inoculation effect on shoot and root biomass showed a similar trend except for the R. pusense (B2) and R. phaseoli + P. polymixa (compare Figure 3 and 4). Effect of co-inoculation on the tissue nitrogen (N) and phosphorous (P) concentration. Tissue N concentration was signi cantly affected by the inoculation of common bean with Rhizobia and PSB. Stimulatory effect was observed in speci c Rhizobia-PSB interactions. For example, R. phaseoli + B. aryabhattai co-inoculation had signi cantly higher N concentration (2.38%) compared to the single R. phaseoli inoculation (2%) ( Table 3). For the R. pusense co-inoculation, the highest N concentration was achieved when co-inoculated with P. polymyxa (2.38%). Similarly, co-inoculation of R. pusense (S5) with P. polymyxa led to signi cantly higher N concentration (2.73%) compared to the single R. pusense (S5) with 2.06% ( Table 2). The highest tissue N concentration was observed in the R. pusense (B4) + B. aryabhattai co-inoculation (2.74%) that was signi cantly higher than the application of DAP with 2.48% N ( Table 3).
For tissue P concentration, application of DAP resulted in the highest P accumulation of 0.98%. This was followed by co-inoculation of R. pusense (S5) + P. polymyxa (0.93%) despite the single R. pusense (S5) having the least %P concentration of 0.47% (Table 3). Co-inoculation of R. phaseoli (B3) with the PSB, B. megaterium led to signi cantly higher %P compared to the other two PSB and the single inoculation (Table 3). Similarly, R. pusense (B2) + B. megaterium had the highest %P concentration compared to the single inoculation and with the co-inoculation with B. aryabhattai and P. polymyxa (Table 3). Single inoculation with the PSB P. polymyxa led to the highest %P concentration (0.73%) compared to the other single inoculations (Table 3). Selected orthogonal contrast. In terms of the number of nodules, there was signi cant contrast among all the selected orthogonal contrasts. There was a highly signi cant (p<0.001) contrast between the control versus the inoculations (single and co-inoculation) and between the co-inoculation and application of DAP (Table 4). In terms of the shoot dry weight, there were no signi cant contrasts between the single inoculation versus the co-inoculation; and between the co-inoculation versus DAP application. Compared to the control, there were signi cant contrasts with the inoculation (both the single and co-inoculation) ( Table 4). For the root dry weight, all the selected contrasts were signi cant except for the single versus the co-inoculation ( Table 4). The orthogonal contrast between all the selected contrasts were not signi cant except for the co-inoculation versus control on the concentration of tissue N (Table 4). There was a signi cant contrast between the co-inoculation versus the control, inoculation versus control, and the control versus the rest on the tissue P concentration. There was no signi cant contrast between the single inoculation versus the control; and the co-inoculation versus DAP application (Table 4).

Discussions
Among the plant growth promoting hormones produced by microorganisms, indole acetic acid is the most common and physiologically active 27 . The results from the present study showed that the tested bacterial strains were able to synthesize IAA in vitro. Six of the bacterial isolates under study exhibited their positive reaction by developing pink colour when reacted with Salkowski's reagent which indicates positive result for IAA production. Further, the results indicated that the microbes differed in their ability to produce IAA. Earlier studies have shown that IAA production by microbes differed between different species or even within strains of the same species 28, 29,30 . From the ten bacterial species and strains tested in this study, the Rhizobium produced higher levels of IAA compared to the Pseudomonas and Paenibaccillus species. This concurs with Tsavkelova et al. 31 , who reported that the strains of Rhizobium are among the most active IAA producers. Similarly Mandal et al. 32 reported that Rhizobia were the rst group of bacteria, which are attributed to the ability of PGPR to release IAA.
The present study tested the in vitro phosphate solubilizing capacity of 17 bacterial isolates comprising of genus Rhizobium, Bacillus, Paenibacillus and Pseudomonas. Sixteen of the 17 isolates were able to solubilize the insoluble tricalcium phosphate by the formation of the halo zones. The size of the solubilization varied among the bacteria isolates tested. Similarly, Andrade et al. 33 reported a wide variations in solubilization indexes and zones among the tested isolates. The solubilization zone occurs due to the presence of some substances, such as organic acids, that are released by microorganisms into the medium that can form metal complexes with calcium, and thereby solubilize the P 34 . All the Bacillus strains tested in this study were able to solubilize the phosphate con rming what has been reported by previous studies indicating that they are e cient P solubilizers. For instance Andrade et al. 33 , indicated that isolates of the genus Bacillus sp. were the most frequent P solubilizers and classi ed them as high e ciency solubilizers. Bacillus megaterium M510 was found to solubilize both aluminium phosphate and iron phosphate in addition to moderate the solubilization of tri-calcium phosphate 35 . Das et al. 36 indicated that Bacillus species isolated from rice rhizosphere solubilized phosphates, which was consistent with the results of our research for the isolate B. megaterium.
In addition to their bene cial nitrogen xing activity with legumes, rhizobia can improve plant P nutrition by mobilizing inorganic and organic P. The present in vitro study showed that all the Rhizobium species were able to form solubilization zones with the tricalcium phosphate. Notably, the present results showed that Rhizobium pusense recorded signi cantly highest solubilization e ciency and solubilization index compared to the Bacillus strains. Earlier studies have also shown Rhizobium as e cient P solubilizers 37,38,39 .
Results from the present study showed that inoculation with Rhizobium strains signi cantly affected the nodulation of common bean. This is in agreement with what has been previously reported by other authors. For instance Bastos 40 showed that inoculation with e cient rhizobial isolates promoted nodulation in common bean. The rhizobia strains used in this study were shown to have high IAA producing e ciency, thus resulting in enhanced nodule number and weights compared to the uninoculated control. Gosh et al. 12 reported that the number of effective nodules were increased when inoculated with IAA-producing rhizobia in Cajanus cajan. Similarly, Pii et al. 41 showed that IAA producing Rhizobium strains led to increased nodulation.
Inoculation of the four Rhizobium strains and their co-inoculation with the three PSB generally increased nodules compared with the uninoculated control. Co-inoculation of common bean with rhizobia and PSB led to an increase in nodule number and nodule dry weight compared to the single rhizobia inoculation. This could be attributed to the multi-strain's ability to effectively nodulate and enhance solubilization of other essential soil minerals such as phosphorus 42 . Similar stimulatory effects on nodulation by coinoculation of rhizobia and PSB has been reported by other authors 43,44,45 .
On the other hand, the present study suggested that the coinoculation of Rhizobium and the PSB might not always increase nodules compared with the individual inoculation with one of the four Rhizobium strains. This results have been reported previously in white clover by 46 , who showed that the coinoculation of CHB1120 and G31 signi cantly increased nodules of white clover compared with the individual inoculation of CHB1120, but the co-inoculation of CHB1121 and two PGPR signi cantly decreased nodules in comparison with the individual inoculation of CHB1121. This suggests that the compatibility between these two kinds of microorganisms should be evaluated before application.
Results from this study showed that common bean in pots inoculated with either single P. polymyxa or B. aryabhattai developed nodules. Other than rhizobia, it was expected that the other bacterial strains will not elicit nodule formation. However, over the years, a vast number of bacteria other than rhizobia have been found in nodules 47,48,49 . A review by 50 highlighted that some of these non-rhizobial nodule endophytes have nif and nod genes and elicits nitrogen xing nodules on nodules just like the rhizobia. A study by 51 reported that the diazotrophic bacteria used in their study were found to have the nitrogen xing genes and nodulated and enhanced nodulation in chickpea plants under greenhouse conditions. There is therefore a need to test more of the non-rhizobial nodule endophytes for their nitrogen xing ability and presence of nod genes to further understand their mechanisms of plant growth promotion.
Results from the present study showed that inoculation of common bean with rhizobia generally increased the shoot biomass and root dry weight compared to the control. The improved growth of plants subjected to Rhizobium inoculation is effectively attributed to its positive effect due to the symbiotic relationship between the rhizobia and the common bean 52 . Inoculation of seeds by phosphate solubilizing microorganisms is known to improve solubilization of insoluble phosphorus, which can therefore increase plant growth by enhancing the symbiotic e ciency of the common bean 53,54 .
In the present study, combined inoculation of the common bean with the rhizobia and PSB resulted in higher shoot and root dry weights compared to the single rhizobia inoculation. This can be attributed to better establishment of Rhizobium-legume symbiosis due to more secretion of plant growth promoting hormones, and improved nutrient availability especially P 53 , 55 . Similarly, Kumar et al. 56 reported a growth enhancement of common bean by application of Bacillus and their combination with Rhizobium. Khalifa and Almalki 57 showed that co-inoculation of phosphate-solubilizing B. megaterium and Sinorhizobium meliloti had a positive effect on the growth of common bean. Co-inoculation of Rhizobium MAP7 along with Brevibacillus MAP4 signi cantly increase the shoot dry weight compared to the treatment with Rhizobium MAP7 alone 58 .
Similar to the results on nodulation, some of the co-inoculation did not lead to improved growth as compared to the single inoculation. For instance, the shoot biomass of common bean co-inoculated with R. pusense and B. megaterium was signi cantly lower compared to their individual inoculations. This suggest that the two strains were not compatible. Other studies have shown the lack of positive effects of co-inoculation in respect to single rhizobium inoculation 59,60 . Therefore, compatibility studies should be done before coming up with the right microbial consortia for formulations of biofertilizers to ensure enhanced crop growth and maximum bene ts from the plant growth promotion of the introduced microorganisms.
Results from this study revealed that inoculation of the common bean with Rhizobia strains increased the shoot N content compared to the uninoculated control. This could be attributed to the formation of nodules by the Rhizobium that stimulated biological nitrogen xation by the crop. Similarly, de Souza et al. 61 reported increased shoot N concentrations by Rhizobium when common beans were inoculated with R. leguminosarum strains. Similarly, shoot P concentration was increased as a results of inoculation with the PSB and Rhizobia. Earlier study by Chen et al. 62 showed that the use of phosphate solubilising bacteria as inoculants increases the P uptake by plants. These ndings are similar to the study by Neila et al. 63 who observed that native rhizobia increase shoot phosphorus in bean. The present study showed a stimulatory effect in speci c Rhizobia-PSB interactions in the total N and P concentration in the plant tissues. For example, R. phaseoli + B. aryabhattai co-inoculation had signi cantly higher N concentration compared to the single R. phaseoli inoculation. The enhancement in total N and P content of shoot in present study might be due to increase of nitrogen and phosphorous acquisition due to altering root structure and nodule formation in the crop 64 . Co-inoculation of common bean with Rhizobium and Bacillus strains was shown to improve nitrogen and phosphorus content compared to single Rhizobium inoculation 65 . A study by Nimnoi et al. 66 showed that the total N and P content of shoot was enhanced by co-inoculation of Nocardia alba strain S4301 with Bradyrhizobium japonicum USDA110 as compared to single inoculation of Nocardia alba strain S4301 and un-inoculated control treatments in soybean. The increased N content from the co-inoculation of Bacillus and Rhizobia strains could also be attributed to the nitrogen xing ability of the Bacillus. A study on nitrogen xing potential of diverse species of Bacillus has reported the presence of nifH gene and hence the capability to x atmospheric nitrogen 67,68 .
Native soil bacteria possesses ability to produce indole acetic acid growth hormones and to solubilize insoluble phosphate. Combined inoculation of rhizobia and PSB promoted the growth of common bean. Therefore, the phosphate solubilizing strains and the nitrogen xing bacterial strains have great potential in being formulated and used as biofertilizers that can be tested under varying eld conditions. This study highlights the importance of the use of phosphate solubilizing and IAA producer microorganisms as biofertilizers to enhance common bean growth.

Materials And Methods
Bacterial isolates description. Seven native plant growth promoting rhizobacteria (PGPR) strains were obtained from Soil Microbial Ecology Laboratory, Egerton University from the isolation work done earlier by Korir et al. 49 . The PGPR were isolated from root nodules of common beans and belongs to the genus Bacillus and Paenibacillus. The native rhizobia strains isolated from bean growing areas of Busia and Bungoma Counties. The isolated strains were characterized and molecularly identi ed at the University of Jena, Germany. The strains were coded as S1-S5 for the different isolates from Busia and B1-B5 for the isolates from Bungoma Counties respectively.
In vitro screening of bacterial isolates for their plant growth promoting properties.
Indole-3 acetic acid (IAA) Production. Indole-3 acetic acid (IAA) production was analysed using modi ed colorimetric method as described by 69 . Ten bacterial isolates were grown in yeast extract mannitol broth supplemented with tryptophan (0.1%) and incubated at 28°C for 2 to 3 days in a shaking incubator. Then, 3 ml of the log phase broth culture (10 9 cfu ml −1 ) was centrifuged at 7826 x g for 15 minutes and 2 ml of the cell-free supernatant was transferred to a dry clean tube (15 ml capacity) to which 1 ml of 10 mM orthophosphoric acid and 4 ml of Salkowsky's reagent (1 ml of 0.5 M FeCl 3 in 49 ml of 35% Perchloric acid) was added and incubated in the dark at ambient temperature (25°C) for 25 minutes. The pink colour development was compared to the blank (sterile LB broth with 0.1% of tryptophan and reagents) at wavelength of 530 ηm and the absorbance was used as index of IAA production. Absorbance values of greater than 0.85 was considered high while those lesser than 0.85 considered low 69 .
In vitro phosphate solubilization. The solubilization capacity of previously isolated Rhizobia and Bacillus strains was checked on Pikovskaya's medium ( 70 Nautiyal, 1999). The growth and solubilization diameter was determined after incubation at 28 ± 2°C for seven days. The size of the halo of solubilization was obtained by subtracting the value of the colony diameter from the total halo solubilization diameter. On the basis of diameter of clearing halo zones, solubilization e ciency (SE) and solubilization index (SI) were calculated using the following formulae 71  Greenhouse experiment. The plant growth promotion of the legume and stimulation of nodulation was tested for strains that showed capacity for phosphate solubilization and IAA production. Four Rhizobia and three Bacillus strains that exhibited high IAA production absorbance (>0.85) and phosphate solubilization e ciency (> 2.0) respectively were used in the greenhouse study.
Inoculum preparation. Rhizobium inoculum was prepared in yeast extract mannitol (YEM) medium and PSB in Pikovskaya's medium 70 . The bacterial cultures were inoculated in 500 mL conical asks containing 150 ml of either the YEM or Pikovskaya's medium and incubated at 28 ±2 o C under shaking for three days to give an optical density of 0.5.
Treatment structure in the greenhouse. The experiment was conducted following guidelines by Figueiredo et al. 73 . Microbe-free vermiculite was obtained by autoclaving for 30 min at 121°C and 101 KPa, once a day for three consecutive days. Pots of 5.3 L capacity (15.0 cm inner diameter and 30 cm length), were lled with 3 kg of vermiculite. Two common bean seeds were planted per pot and thinned to one seedling per pot one week later, after which it was inoculated as per treatment i.e. un-inoculated, un-inoculated + inorganic NP source, PSB, Rhizobium and combination of PSB and Rhizobium (1:1). For single inoculation, 1ml of broth culture containing a Rhizobium or PSB (10 9 cfu ml -1 ) was inoculated per plant.
For the co-inoculation, 0.5 ml of YEM broth containing a Rhizobium (10 9 cfu ml -1 ) plus 0.5 ml of nutrient broth containing the phosophobacteria (10 8 cfu ml -1 ) was applied per plant. Ten milliliters of nitrogen-free nutrient solution (Broughton and Dilworth, 1970) was applied to the pots once a week until owering started. The treatment structure is shown in Table 4. The experiment was laid out in completely randomized design (CRD) with three replicates. The pots were watered regularly to maintain the substrate at eld capacity. Data analysis. Data on in vitro IAA production and phosphate solubilization (SE and SI) was subjected to analysis of variance (ANOVA) and the means separated using Tukey least signi cant difference (α = 0.05). For the greenhouse experiment, data was rst tested for normal distribution and the count data on nodule number was log transformed (Log 10 x+1) before analysis so as not to violate the assumptions of ANOVA (Payton et al., 2006). To determine the effects due to inoculation, analysis of variance at p < 0.05 was done and means separated using the least signi cance difference (Tukey's test at α = 0.05). Data was analysed using SAS Statistical Package Version 9.3 74 .

Figure 1
Absorbance (at 530 nm) values for different Rhizobium species using colorimetric method. Letters in brackets represent the location of isolation; S-Busia and B-Bungoma; and the numbers are codes given to the isolates. Means followed by different letters are signi cantly different from each other at α≤0.05.

Figure 2
Effect of co-inoculation of common bean on number of nodules. Error bars represent the standard error of the means. Effect of co-inoculation of common bean on the root biomass. Letters in brackets represent the location of isolation; S-Busia and B-Bungoma; and the numbers are codes given to the isolates. Error bars represent the standard error of the means

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.