Potential of Managing Iron and Zinc Deficiency in Dry Beans with Interplantings of Annual Ryegrass and Increased Bean Density

2008 Field Study Results and Discussion

There were no differences between treatments for yield for the 2008 study. Hand-weeding of the study was ineffective in eliminating persistent weeds. This may have affected the yields obtained for dry beans.

There were no differences for soil and tissue iron and zinc between treatments. Though iron chlorosis symptoms were observed in pinto bean plants during the first month after planting, the beans were generally able to overcome these symptoms as the season progressed.

Although pinto beans in 2008 study did not exhibit sustained iron deficiency chlorosis symptoms, soy beans and black beans planted on adjacent plots by another researcher were observed to be more adversely affected by iron deficiency chlorosis that could potentially reduce their yields. This suggested that different dry bean cultivars have different tolerances to micronutrient deficiency. A study planned for 2010 would attempt to determine the potential role of annual ryegrass to mitigate iron deficiency in more susceptible dry bean cultivars.

There was a significant increase in mean iron concentration (p<0.0001) with respect to sampling time. Iron concentration was significantly higher in all plots during the last two sampling dates in August compared to the earlier sampling dates in June and July (Table 1). Since there was no interaction between treatments and sampling dates, a seasonal variable may have been responsible for the increased iron concentration. Gradual increase in mean temperature between May and August (Table 2) may have been responsible for these results. The mechanism influenced by temperature that increased iron availability over time was not determined by this study. We hypothesize that increasing temperature as the season progressed may have increased microbial activity and root interactions which may have processed more soil iron through means such as siderophore exudation and chelation.

There was a significant decrease in mean nitrate nitrogen (p<0.0001) with time (Table 3). Nitrate loses through the season can be expected from utilization by plants, volatilization and leaching though irrigation and precipitation. Reduction in nitrate nitrogen in the soil may partly explain the observed recovery of some cultivars of dry beans from the initial iron deficiency symptoms exhibited early in the season (late spring to early summer). High levels of nitrate nitrogen in soils that already have high levels of salt and free calcium carbonate are thought to interfere with iron metabolism in plant leaves, depress chlorophyll synthesis and induce iron deficiency chlorosis (Christensen and Johnson, 2008).

2009 Field Study Results and Discussion

There were no differences for soil and tissue iron and zinc concentration between treatments. Just as in the 2008 study, iron chlorosis symptoms observed in the pinto bean plants during the first month after planting disappeared as the season progressed. Unlike 2008, however, increase of soil iron concentration with time in all treatments was only marginal in 2009 (p=0.07). Unexpected heavy and prolonged precipitation in 2009 may have contributed to a reduction in environmental and soil temperature (Table 2), resulting in lower soil iron concentration compared to the previous year. Different sampling dates, due to delay in planting and failure of pinto beans to germinate, may also have failed to capture the time period when iron concentration was expected to rise.

Pre-emergent application of ‘Treflan’ herbicide resulted in effective season-long control of the prevalent weeds observed in the previous study. High density pinto beans produced at least twice higher yields (p<0.0001) than all the other treatments (Table 4). The effect of high density beans on soil and tissue iron availability, however, could not be ascertained from this study. Pinto beans intercropped with wheat produced the lowest yields (p<0.0001), suggesting that wheat out-competed dry beans for nutrients and light and may therefore not be a suitable crop to interplant dry beans with. There were no differences in yield between the lower density beans, beans intercropped with annual ryegrass and beans intercropped with annual ryegrass sprayed later with select herbicide, suggesting that the slower-growing annual ryegrass exerted less competition for nutrients and light on dry beans compared to wheat. Annual ryegrass may therefore be a better companion for dry beans when the two crops are intercropped. The specific role played by annual ryegrass in increasing iron availability was not established by this study. Subsequent studies utilizing less tolerant dry bean market classes intercropped with annual ryegrass will attempt to address this need.

Growth Chamber Study Results and Discussion

In 2009 study, bean seedlings developed long internodes and bolted toward fluorescent light tubes in the chamber, suggesting that light intensity was insufficient for those plants. Both chambers were designed for smaller statured plants and organisms. Adjusting the day-light period from 13 to 15 hours in subsequent study shortened bean seedling internodes and reduced the tendency of the plants to bolt toward the light source. Bean plants in the warmer growth chamber were larger than plants in the cool growth chamber (Figure 1).

There were no differences between treatments within each growth chamber for above and below ground biomass, as well as root nodule counts. However, temperature had a significant effect on the shoot biomass (Table 5). Bean plants in the high temperature growth chamber had significantly higher shoot biomass (p = 0.0004) than bean plants in the low temperature growth chamber. The youngest fully expanded leaves of the plants in the high temperature chamber had higher chlorophyll content (p<0.0001) than plants in the low temperature chamber. These results suggest that temperature increases plant nutrient availability and uptake and are consistent with our field study results. Just as in the field study, however, this study did not determine the actual mechanism influenced by high temperature that increased iron availability and prevented development of chlorosis symptoms.

Intercropped bean plants had significantly higher chlorophyll content than monoculture bean plants in the high temperature chamber (p<0.0001). There was no significant difference between treatments for chlorophyll content in the low temperature chamber. These results are, however, not consistent with our 2008-2009 field study results, partly because we used a bean cultivar (‘T-39’ Black bean) thought to be more susceptible to iron-deficiency in our growth chamber study, compared to the more tolerant pinto beans used for the field study.

Results from our field and growth chamber studies suggest that seasonal temperature changes can influence soil iron availability for plant uptake. Further studies are needed, however, to determine the specific mechanisms in the soil that change iron availability when temperatures change. Results from these studies also suggest that annual ryegrass has the potential to mitigate iron deficiency chlorosis in more susceptible dry bean cultivars. Additional studies to investigate this potential for three dry bean cultivars with varying susceptibility to iron deficiency were carried out in 2009 and repeated in 2010. Our results did not ascertain the potential of high dry bean plant density to alleviate iron deficiency.

2009 Field Study: Soil and Tissue Nutrient analysis

The study found significant differences between treatments for chlorophyll content. Navy and black beans had significantly less chlorophyll content (and therefore more chlorotic) compared to pinto beans (p < 0.0001). Grass intercropped bean plants had significantly higher chlorophyll content (p < 0.0001) than bean monocultures (Figure 4; Table 6).

There were significant differences between treatments for bean tissue iron concentration (p<0.0001). While there were no differences for tissue iron between Navy bean monoculture and navy-ryegrass intercropped treatments, these two treatments had the highest mean iron concentration in their tissues compared to all the other treatments (Table 6). Similarly, although there were no differences for tissue iron concentration between black bean monocultures and black beans interplanted with corn and oats, these three treatments had significantly higher mean iron concentration in bean tissues than black beans interplanted with wheat. Paradoxically, bean plants in treatments with less chlorophyll content accumulated more iron and nitrate-nitrogen (p = 0.003) in their tissues than less chlorotic beans (Table 6). All intercropped treatments had less mean nitrate-nitrogen in their tissues (p<0.0001) than all bean monoculture treatments (Table 6).

The study found significant differences between treatments for bean tissue manganese concentration (p=0.0006). Whereas there were no differences for tissue manganese between navy bean and black beans monocultures, these two treatments had the highest mean manganese concentration in their tissues compared to all the other treatments (Table 1). Apart from bean-corn intercrop, which had higher manganese concentration than bean-grass intercropped treatments with oats and wheat, all the other intercropped treatments were not significantly different for tissue manganese concentration (Table 1). There were no differences between treatments for soil manganese concentration.

There was a significant decline in tissue iron concentration (p=0.0026) and, at the same time, a significant increase in tissue zinc concentration (p = 0.0031) between the first and second sampling date for all treatments, though there were no differences between treatments for bean tissue zinc concentration (Figure 5 and 6). These results were reversed in soil zinc and soil iron concentration, (Figure 7) whereby there was a significant increase in soil iron for all treatments between the first and second sampling dates (p=0.02) and a significant decline in soil zinc concentration for all treatments between the first and second sampling date (p=0.042), though there were no differences between treatments for soil iron and soil zinc concentration.

All three monocropped dry bean cultivars produced higher yields (p = 0.0009) than all the intercropped beans.

Phopholipid Fatty Acid Analysis

The results of the Gas Chromatographic PLFA assay of soil samples obtained from four representative treatment plots found no significant differences between treatments. However, there was a general trend showing higher total bacterial, fungal and protozoan abundance in intercropped treatments (Figure 8).

Greenhouse Study Results

The study carried out between July 29 and September 5 found no differences between treatments for chlorophyll content. Bean plants grown in unsterilized soil, however, exhibited a significant increase in chlorophyll content (p=0.024) between the first and second sampling date. The reverse was the case in bean plants grown in sterilized soil, whereby there was a marginal decrease in chlorophyll content (p=0.06) between the first and second sampling date (Figure 9 and 10; Table 7).

When the study was repeated between September 4 and October 18, treatment had highly significant effects (p=0.009) on chlorophyll content for bean plants in sterilized soil (Figure 11; Table 8). Whereas there was no difference in chlorophyll content in bean plants between bean-wheat and bean-annual ryegrass intercropped treatments, both treatments had significantly higher chlorophyll content than bean monoculture treatment. There were also statistical differences between sampling dates (p=0.002), whereby the highest chlorophyll content was recorded on the first sampling date, September 26, and the lowest chlorophyll content recorded on the last sampling date, October 18 (Table 7).

Whereas there were no statistical differences between sampling dates for beans planted in unsterilized soil during this study period, bean plants in intercropped treatments had significantly higher chlorophyll content (p=0.026) than bean plants in monoculture treatment in the September 4-October 18 study (Table 8).

The study found significant differences in root nodule counts between the two soil media during the first study period (July 29-September 5). Bean root nodule counts were significantly higher (p=0.004) in unsterilized soil compared with bean plants in sterilized soil, with the highest counts in bean-annual ryegrass and bean-wheat intercropped treatments (Table 9). There were also significant differences in total above- and below-ground bean dry matter biomass weights between treatments across the two soil media (p=0.04). Bean monoculture and bean intercropped with wheat in the sterile soil media had the highest dry matter biomass compared to the other treatments (Table 9).

Field Study

Although black and navy bean monocropped treatments in the field study exhibited significantly more chlorosis than the intercropped treatments, the bean monoculture treatments had more tissue iron and zinc concentration, and ultimately higher yields than the intercropped plots. Bean plants with higher iron concentration in their tissues exhibiting less chlorophyll content may be explained by a phenomenon referred to as the "chlorosis paradox" described by Abadía (1992); Marschner (1995); and Morales et al. (1998). These inconsistencies may be attributed to the localization and binding state of iron in leaves (Marschner, 1995), whereby some of the iron may precipitate in the apoplasm of leaves and become less available physiologically (Mengel and Geurtzen, 1988; Marschner, 1995). These results are consistent with similar findings by Omondi et al. (2010), who observed that bean plants exhibiting iron deficiency chlorosis symptoms had more iron concentration in their tissues than beans that were less chlorotic.

Minimum soil Zn concentration before Zn chlorosis symptoms can occur is 0.5 ppm (Jacobsen et al., 2003). On the other hand, the deficiency range of soil iron concentration is 2.5 - 5 ppm (Jacobsen et al., 2003). Soil zinc concentration from the field study (2ppm) was far above the threshold found by Jacobsen et al. (2003) to cause zinc chlorosis. Average soil iron concentration on the other hand was 5 ppm which was within the range of 2.5 - 5ppm found by Jacobsen at al. (2003) as capable of causing iron chlorosis. These results therefore suggest that the chlorosis symptoms observed were caused by soil iron deficiency rather than zinc deficiency.

While soil iron concentrations from this study were low according to Jacobsen et al., (2003), they were generally above the critical concentration level of 5 ppm before the performance of dry beans can be seriously impacted by iron deficiency. This suggests that the performance of dry beans was influenced by competition from grass intercrops more than micronutrient deficiency. ‘Select’ herbicide is applied to remove grass intercrops from dry beans one month after planting. This allows for adequate root interaction between the two species to enable any benefits expected from the grass to accrue. However, given that it takes about two weeks for ‘Select’ herbicide to kill its target plant, competitive effects from the grass intercrops can offset any benefits that may have been obtained from intercropping. There is therefore a need to carry out further studies to determine the optimum time of grass removal that can mitigate yield losses due to grass competition.

An interesting result of this study was whereby tissue iron concentration declined at the same time as tissue zinc concentration increased between sampling dates. Curiously, the situation was reversed with regard to the soil micronutrients, whereby soil zinc concentration declined in apparent opposition to soil iron concentration, which increased as the season progressed. These results may be explained by studies by Ambler et al. (1970) which found that soil zinc interfered with the translocation of iron in soy beans by inhibiting the capacity of the root conversion of ferric to ferrous iron or by accentuating other reactions detrimental to iron transport. Both nutrient supply and nutrient balance are important considerations in plant nutrition, as the concentration of one nutrient in the soil will often affect the uptake or transport of another nutrient within the plant (Biertman and Rosen, 2005). Nutrient interactions not only involve the relationship between nutrient supply in the soil and plant growth, but also nutrient concentrations in plant tissue and plant growth (Biertman and Rosen, 2005). According to Biertman and Rosen (2005), “the precise nature of nutrient interactions depends on the nutrients involved and can vary for different plant species. Furthermore, the actual mechanism for the interaction may not be completely understood.” Concentration of soil zinc in our study was relatively low according to Jacobsen et al., (2003) and may, therefore, have not substantially affected performance of beans in accordance with Ambler et al. (1970). Higher zinc concentration may benefit from a polycultural cropping design such as this experiment, which would enable more utilization of the micronutrient and alleviate its potential interference with iron uptake.

Our study found significantly higher bean tissue manganese concentration in black and navy bean monocultures compared to the intercropped treatments. Mean manganese concentration was 117 ppm and 123 ppm in black beans and navy beans respectively. Studies by Fageria (2001) to determine adequate and toxic levels of copper and manganese in upland rice, common beans, corn and soybeans found that manganese concentration of 128 ppm caused toxicity symptoms in common beans. Mean tissue manganese concentration for black and navy beans in our study were close to the threshold established by Fageria (2001) as capable of influencing manganese toxicity and/or inducing iron deficiency chlorosis. Sideris and Young (1949) suggested that heavy metals such as manganese with similar ionic radii, valency and electronic configuration as iron can compete with iron for reaction sites that lead to chlorophyll formation. Similar results were reported by Epstein (1972), Marschner (1986) and Mengel and Kirkby (1987). Manganese, rather than iron, can react with porphyrin compounds, thereby inactivating them for subsequent conversion to chlorophyll (Sideris and Young, 1949). Studies by Twyman (1951) to determine the relationship between iron and manganese in the tissue and growth media on iron deficiency chlorosis found that high manganese concentration under adequate iron bioavailability caused chlorosis in various plants grown in nutrient medium, the intensity of which was correlated with manganese concentration in the medium. This chlorosis was attributed to competition between iron and manganese for active iron acceptors, antagonism between iron and manganese, whereby manganese slowed the entry of iron and encouraged formation on inactive iron receptors and a direct toxic action of manganese on processes involved in iron metabolism (Twyman, 1951). Although there were no differences between treatments for soil manganese concentration in our study, mean soil manganese concentration was 6.39 ppm for navy beans and 7.19 ppm for black beans. Fageria (2001) determined toxic concentration of soil manganese to be 6 ppm for dry beans. Results from our study, therefore, suggest that high levels of manganese were present in the soil, and that grass intercrops prevented excessive uptake and accumulation of the micronutrient in the bean tissues, thereby alleviating induction of iron deficiency chlorosis.

Findings by this study that all intercropped treatments had less mean nitrate-nitrogen in their tissues than all the bean monoculture treatments suggest that high nitrate-nitrogen in the monoculture treatments may have played a role in causing iron deficiency chlorosis in those treatments. These results are consistent with findings by Christensen and Johnson (2008) that high levels of nitrate-nitrogen can interfere with iron metabolism in the plant leaves, depress chlorophyll synthesis and induce iron deficiency chlorosis. Kosegarten et al. (1999) hypothesized that nitrate nutrition might induce iron deficiency chlorosis by inactivation of iron in the leaf apoplast. Intercropping dry beans with annual ryegrass as a means to reduce nitrate-nitrogen in the soil and bean tissues may therefore provide a potential solution to iron deficiency chlorosis, induced by high nitrate-nitrogen, in dry beans.

Annual ryegrass performance in the field study was comparable to those of oats and corn with regard to mitigating iron deficiency chlorosis. Navy and black beans intercropped with these three grasses were less chlorotic and had higher tissue iron concentration compared to treatments with spring wheat. Oats and corn have been found to have the ability to mitigate Fe deficiency chlorosis through root exudates of phytosiderophores which can chelate iron and make it more available (Romheld and Marschner, 1990; Singh et al., 2005). These results suggest that annual ryegrass is a potentially suitable grass to include in a dry bean intercropping management program designed to combat iron deficiency.

Greenhouse Study

Results from the greenhouse study found a decrease in chlorophyll content for bean plants growing in sterilized soil and a corresponding increase in chlorophyll content for beans grown in unsterilized soil. Increased chlorophyll content in bean plants grown in unsterilized soil suggest that, with time, greenhouse conditions (adequate temperature, light, moisture and air) may have helped to increase microbial activity in the soil, some of which may have produced siderophores that chelated iron and made it more available for plant uptake. Increased root nodule counts in this medium confirms that microbial action was at play, including biological dinitrogen fixation by root-nodule-forming bacteria of the genus rhizobia that exist in symbiotic association with leguminous plants. Considered as one of the most limiting nutrients to plants in terrestrial ecosystems (Sylvia et al., 2005), nitrogen is used by plants to synthesize amino acids, the primary components of proteins, and is also required by plants for other vital components such as chlorophyll and enzymes (Ludwick et al., 2002). Biological dinitrogen fixation is catalyzed by the nitrogenase enzyme system which consists of two metalloproteins, the iron (Fe-) protein and the molybdenum-iron (MoFe-) protein (Kim and Rees, 1994). The legume–rhizobia symbiosis is particularly sensitive to iron deficiency, given its critical role as a vital structural and functional component of the nitrogenase enzyme (Tang et al., 1991).

Significantly lower root nodule counts on bean plants grown in autoclaved soil is attributed to the high temperatures involved in autoclaving soil, which may have destroyed microorganisms in the soil, including rhizobia. High initial chlorophyll content in these bean plants may therefore be a result of resident mineral nitrogen in the soil and an increase in available micronutrients such as iron due to the high autoclaving temperatures. In a study by Abou-Shanab et al. (2003), autoclaving increased extractable Fe, Mn and Co but had no significant effect on other elements examined. Reduction of chlorophyll content in bean plants in autoclaved soil in our study, as the season progressed, may have been due to depletion of nitrogen, iron and other nutrients in the soil as the plant developed. However, interveinal chlorosis that developed in these plants could also have been caused by manganese toxicity. Studies by Boyd (1971) found that interveinal leaf chlorosis occurring in Argentine peanuts growing in autoclaved soil was the result of manganese toxicity. The toxic level of manganese was attributed to direct release of manganese complexed with organic fraction of the soil and the killing of microorganisms that normally transform available manganese into higher, less available and therefore less toxic oxides (Boyd, 1971). As has already been described elsewhere in this paper, high manganese concentration can induce iron deficiency chlorosis by competing with iron for reaction sites that lead to chlorophyll formation (Hewitt, 1948; Sideris and Young, 1949; Twyman, 1951; Epstein, 1972; Marschner, 1986; and Mengel and Kirkby, 1987). Our greenhouse study finding that bean plants intercropped with annual ryegrass and wheat growing in autoclaved soil had higher chlorophyll content than monoculture bean plants in the same medium are therefore consistent with our field studies that grass intercrops prevented excessive uptake and accumulation of manganese in bean tissues, thereby alleviating induction of iron deficiency chlorosis.

PLFA Analysis

Phospholipid fatty acid analysis (PLFA) provides a means to study in-situ microbial communities without the usual problems associated with their isolation or removal of cells from the environment (Petersen and Klug, 1994). This method examines the entire microbial community, thus providing a quantitative description of the structure and function of the micro-flora within a particular environment (Sinsabaough et al., 1999). PLFA characterizes the composition of microbial biomass by identifying extracted cellular phospholipid ester-linked fatty acids (Zelles et al., 1992). Microbial communities under different agricultural management profiles can also be distinguished using a simplified extraction of cellular fatty acid methyl esters (Cavigelli et al., 1995). The assay is based on the fact that all organisms except archaea have membranes made of phospholipids. Many different fatty acids exist in microbial lipids and can be used to identify microorganisms (Sylvia et al., 2005). Phospholipid fatty acids are an integral part of cell membrane and are rapidly metabolized when the cell dies in the soil and can therefore provide an accurate measurement of living organisms (Sylvia et al., 2005). PLFA assay involves extraction of lipids from the soil into a mixture of water, chloroform and methanol, followed by separation of different classes of lipids (Sylvia et al 2005). The PLFA fraction is then analyzed after hydrolysis and methylation under alkaline conditions. The PFLAs are extracted with organic solvents and analyzed by Gas Chromatography. Typically 20 to 50 fatty acids can be detected and differentiated (Sylvia et al 2005). Statistical analysis of fatty acid composition can enable identification of specific fatty acids that distinguish between microbial communities in different soils or in specific soils under different management conditions (Sylvia et al 2005). PFLA assays can therefore tell us which kinds of microorganisms are present and their relative amounts and diversity in the soil.

Although the Gas Chromatographic PLFA assay analysis of soil samples found no significant differences between treatments, there was a general trend showing higher total bacterial, fungal and protozoan abundance in intercropped treatments, which suggests that iron-chelating microorganisms may have interacted with grass root exudates or organic matter in these treatments to increase iron availability resulting in higher chlorophyll content. The exact nature that such interactions may occur was not established by this study and should be a subject for further investigation.

While results from our study showed that annual ryegrass can mitigate iron deficiency chlorosis in dry beans, grass intercropping did not result in higher dry bean grain or biomass yields in both field and greenhouse studies. Additional studies are required to determine the appropriate annual ryegrass and dry bean densities, as well as the optimum time of removal of annual ryegrass from the intercropped treatment that can not only alleviate iron deficiency chlorosis symptoms but also increase dry bean yields.

The study found significant differences between treatments for chlorophyll content. Bean-annual ryegrass intercropped treatments had significantly higher chlorophyll content (p=0.004) than bean monoculture, regardless of the nitrate-nitrogen treatment or growth chamber temperature (Table 10). Treatments with low soil nitrate-nitrogen had significantly higher (p=036) chlorophyll content compared with treatments with high soil nitrate-nitrogen (Table 2). Bean monoculture in high nitrate-nitrogen treatment in the low temperature chamber had the lowest chlorophyll content.

Treatments with low soil nitrate-nitrogen had significantly higher (p=024) root nodule counts compared with treatments with high soil nitrate-nitrogen, regardless of growth chamber temperature (Table 10). Treatments in the high temperature growth chamber in low nitrate-nitrogen treated soil media had significantly higher (p=0.018) root nodule counts than all the other treatments (Table 11). In this case, however, bean monoculture in high nitrate-nitrogen treatment in the high temperature chamber had the lowest root nodule count.

There were no differences between treatments for root fresh weight or dry weight biomass. However, temperature had a marginal effect (p=0.085) on above ground fresh weight biomass, with mean fresh weight of 10.6g per bean plant in the high temperature growth chamber and 6.0g per bean plant in the low temperature growth chamber. Temperature also had a significant interaction with nitrate-nitrogen treatment for above ground fresh biomass (p=0.018). Mean fresh weight of bean plants in low nitrate-nitrogen treatment was significantly higher (p=0.007) than mean fresh weight of bean plants in high nitrate-nitrogen treatment (Table 10) even though there were no differences in fresh weight between monoculture and intercropped bean plants in either low or high nitrate-nitrogen treatment. The study did not find any differences between above ground bean dry biomass as well as total above and below ground bean dry biomass. However, total above and below ground bean dry biomass showed a similar trend to the above ground fresh biomass weights.

Findings by this study that bean plants growing in annual ryegrass-intercropped treatments had higher chlorophyll content than monoculture bean plants are in agreement with field and greenhouse studies by Omondi et.al. (Unpublished) as well as field studies by the same author (Omondi et al., 2010), which showed that interplantings of annual ryegrass alleviated chlorosis symptoms in dry edible beans. These studies suggest that annual ryegrass may be an iron efficient plant that can utilize Strategy II mechanism, whereby root exudates containing phytosiderophores can be secreted into the rhizosphere root zone, thereby chelating iron and making it more bio-available for uptake by both the grass and dry beans growing in close proximity (Marschner and Romheld, 1994; Marschner, 1995; Vansuyt et al., 2000). However, there is still a need to conduct HPLC analyses to confirm the ability of annual ryegrass to excrete phytosiderophores in nutrient medium.

Although high temperature improved the performance of bean plants in all treatments, the study found that high nitrate-nitrogen reduced chlorophyll content, root nodule counts and above ground fresh biomass of bean plants regardless of the temperature treatment. These results not only suggest that high nitrogen induces iron deficiency, but also that black beans are iron inefficient plants which can secrete OH- and HCO- ions into the rhizosphere root zone in the presence of high nitrate-nitrogen (Aktas and Egmond, 1979; Romheld et al., 1984). These results are in agreement with Smolders et al. (1997), who found that increased nitrate assimilation by sharpflower rush (Juncus acuti?orus) led to increased apoplastic pH and to simultaneous immobilisation of iron and/or lower Fe3+ reduction. These results are also in conformity with studies by Aktas and Egmond (1979), which showed that high nitrate-nitrogen decreased dry matter production, caused ionic imbalances and reduced chlorophyll content of iron inefficient soybean cultivar ‘T-203’.

The study found significant differences in soil zinc concentration between treatments. There was significantly higher (p = 0.042) zinc concentration in pinto beans-annual ryegrass intercropped plots compared to pinto beans monoculture (Table 12). There was also significantly higher (p<0.001) organic matter in beans-ryegrass intercropped plots compared to the other treatments (Table 12). A linear regression analysis of the means of soil zinc concentration against soil organic matter content found a significant correlation (p = 0.49; r2 = 0.33) between soil zinc concentration and soil organic matter (Figure 17).

Sampling date had a significant effect on zinc availability on all plots (p = 0.05). There was a significant increase in zinc concentration between the first and second sampling date, with the highest increase in the bean-annual ryegrass plots (Table 13, Figure 16). Peak soil zinc concentration was observed on June 26 (Figure 16).

There were no differences between treatments for soil iron, soil pH and soil nitrate-nitrogen. There were also no differences between treatments for tissue iron, tissue zinc and tissue nitrate-nitrogen.

Findings by this study that there was significantly higher soil zinc concentration in the bean-annual ryegrass intercropped treatment compared with the monoculture treatment suggest that intercropping annual ryegrass with beans has the potential to mitigate zinc deficiency in the highly weathered acidic soils in western Kenya. Corresponding high organic matter content in the bean-annual ryegrass intercropped treatment compared to the other treatments (Table 12) suggests that organic matter contributed to the increase in soil zinc concentration. A linear regression analysis of the means of soil zinc concentration against soil organic matter content found that organic matter content explained 33% of the variability observed in soil zinc concentration between pinto-beans/annual ryegrass intercropped plots and pinto beans monoculture (Figure 17). However, the relatively low r2 value of this analysis discounts the possibility that soil organic matter by itself contributed to the increased soil zinc content in the intercropped plots.

Higher soil organic matter in the bean-annual ryegrass intercropped plots in this study suggests that the extensive rooting system of annual ryegrass may have interacted with the rooting system of pinto beans resulting in increased below ground biomass. Soil organic matter increases the buffering capacity of soil and plays an important role in nutrient availability and soil aggregate stability (Hussain et al., 1999). Soil organic matter can undergo partial decomposition and give rise to organic acids and other breakdown products such as carbon dioxide and water, which combine to form soluble micronutrient-organic complexes (Chen and Hadar, 1991). Soil organic matter provides many beneficial biological, chemical and physical properties to the soil. With regard to biological properties, soil organic matter provides slow-release carbon and energy source which can support a large, diverse and metabolically active microbial community (Sylvia et al., 2005). Soil organic matter is also a source of certain compounds that can promote plant growth (Sylvia et al. 2005). Chemically, soil organic matter increases the cation exchange capacity, buffering capacity, provides a slow-release supply of organically bound nutrients such as nitrogen, phosphorus, and sulfur, and enhances chelation thus increases bioavailability of micronutrients to plants (Hussain et al., 1999, Sylvia et al. 2005%2

• Growth chambers used in the study
• Means of chlorophyll content of black beans and black beans intercropped with two cereal species; wheat and annual ryegrass at two study periods
• Bean yield sampled from 3m rows in grams
• Black and navy bean monocultures were more chlorotic than corresponding bean intercropped with oats
• Average percentage microbial abundance against treatments showing a trend toward low abundance in bean monoculture treatment. These results were not significant.
• Bean plants growing in unsterilized (blue tag) and sterilized (red tags) on on September 26, 2010.
• Bean plants growing in unsterilized soil (left three pots) and sterilized soils (right three pots) on October 18, 2010. Chlorophyll content of beans in unsterilized soil increased as the opposite happened for beans in sterilized soil.
• Mean chlorophyll, root nodule counts, and total below and above ground biomass of bean-ryerass intercropped and bean monoculture plants under low and high soil nitrogen treatments.
• Mean soil zinc concentration and organic matter content between treatments
• Mean soil zinc concentration for all treatments between sampling dates
• Average tissue zinc concentration in parts per million against sampling time
• : Bean plants in bean-annual ryegrass treatments (two left pots) compared to bean monoculture treatments in the two pots on the right, both in sterilized soil media. Beans intercropped with annual ryegrass remained green during the duration of the study while beans in monoculture treatment progressively became chlorotic.
• Means of chlorophyll content of black beans and black beans intercropped with two cereal species; wheat and annual ryegrass at two study periods
• : Mean chlorophyll content and tissue iron between treatments in 2009. Pinto beans failed to germinate and had to be reseeded. Trifoliate leaves had therefore not sufficiently developed to be sampled for tissue analysis during the first sampling date and were therefore not included in the analysis.
• Mean soil iron concentration for all treatments between sampling dates
• Mean soil nitrate nitrogen concentration for all treatments between sampling dates
• Warm growth chamber bean plants (left) compared to cool growth chamber plants
• Mean shoot tissue dry weight biomass in grams between treatments in two growth chambers under two temperature regimes
• Mean chlorophyll and root nodule counts of bean-ryegrass intercropped and bean monoculture plants under low and high soil nitrogen treatments between two growth chambers set at high and low temperatures.
• Changes in zinc concentration with sampling time between pinto bean monoculture and pinto bean-annual ryegrass intercrop treatments.
• Mean monthly temperatures in 2008 and 2009 in degrees Fahrenheit (Weather Warehouse)
• Average tissue iron concentration in parts per million against sampling time
• Average soil iron and soil zinc concentration in parts per million against sampling time
• Regression of soil zinc concentration against soil organic matter.
• Means of chlorophyll content of bean plants in sterilized and unsterilized soil at two sampling dates during two study periods.