Papers

2. LITERATURE REVIEW

2.1 The Nitrogen and Nitrogen Fixation

2.1.1 Role of Nitrogen in the Biosphere

Nitrogen (N) is one of the most abundant elements on earth, and after carbon (C), hydrogen (H), and oxygen (O), the element living creatures need most. The atmosphere over each square foot of the earth’s surface—which is 78% dinitrogen (N 2) gas—contains approximately 6,000 pounds of nitrogen. However, the majority of the earth’s nitrogen (98%) is in rock, sediment, and soils. The amount of nitrogen in rocks is about 50 times more than that in the atmosphere, and the amount in the atmosphere is approximately 5,000 times more than that found in soils (Stevenson, 1982).

Nitrogen is an essential nutrient for life on Earth. It is an important element in proteins and nucleic acids, which are vital to all living and some non-living (viruses, prions, etc.) organisms on Earth (Kormondy, 1996). Although nitrogen makes up about 78 percent of the atmosphere, it is often a limiting factor for plants and animals (Kormondy, 1996). These organisms experience a shortage because they cannot use nitrogen in its dinitrogen gaseous form. Plants can only use nitrogen in the form of nitrate or ammonium ions, and animals can only use it in organic forms, which they obtain by consuming plants or other animals (Kormondy, 1996). The weathering of rocks releases these ions so slowly that it has a negligible effect on the availability of fixed nitrogen. So, nitrogen is often the limiting factor for growth and biomass production in all environments where there is suitable climate and availability of water to support life.

Nitrogen moves from the atmosphere to usable forms and then back to the atmosphere through a process known as the Nitrogen Cycle. Organisms, both living and dead, and soil act as the reservoir for usable nitrogen in organic and inorganic forms (Kormondy, 1996).

Nitrogen is an essential plant nutrient. It is the nutrient that is most commonly deficient, contributing to reduced agricultural and forest yields throughout the world. The Green Revolution was accompanied by a huge increase in the application of fertilizers, particularly nitrogen. Grain crop yields have constantly increased during the 20^th century to reach remarkable levels in many years. This increase in yield has been primarily facilitated by the introduction of genetically improved cultivars and use of chemical pesticides and fertilizers. Improvement in grain crop production technology from ‘slash and burn’ practice to the success of ‘Green Revolution’ was associated with increasing nitrogen (N) availability to crops. The expanding use of industrially manufactured N fertilizers has been one of the main factors behind the fast growth in agricultural productivity over the last few decades. Over the period 1950 to 1990, the per capita use of N fertilizer increased about ten fold from about 1.3 to 15 kg N person^-1, and has leveled off in the last decade of the 20^th century (Galloway, 1998).

Figure 2.1: N – A Limiting Factor in Agriculture and Forest Productivity (Source: Internet)

2.1.2 Nitrogen Fixation

A relatively small amount of ammonia is produced by lightning. Some ammonia also is produced industrially by the Haber-Bosch process, using an iron-based catalyst, very high pressures and fairly high temperature. But the major conversion of N₂ into ammonia, and thence into proteins, is achieved by microorganisms in the process called nitrogen fixation (Walker et al., 1983; Sprent, 1987; Purves et al., 1992).

Table 2.1: Some estimates of the amount of nitrogen fixed on a global scale

Type of fixation	N2 fixed (1012 g per year, or 106 metric tons per year)
Non-biological
Industrial	about 50
Combustion	about 20
Lightning	about 10
Total	about 80
Biological
Agricultural land	about 90
Forest and non-agricultural land	about 50
Sea	about 35
Total	about 175
Source: Bezdicek, DF & Kennedy, AC, 1998

2.1.3 Biological Nitrogen Fixation

Approximately 80% of the atmosphere is nitrogen gas (N₂). Unfortunately N₂ is unusable by most living organisms. Plants, animals, and micro-organisms can die of nitrogen deficiency, surrounded by N₂ they cannot use. All organisms use the ammonia (NH₃) form of nitrogen to manufacture amino acids, proteins, nucleic acids, and other nitrogen-containing components necessary for life.

Biological nitrogen fixation is the process that changes inert N₂ to biologically useful NH₃ (Lindermann, and Glover, 1998). This process is mediated in nature only by bacteria. Other plants benefit from nitrogen fixing bacteria when the bacteria die and release nitrogen to the environment, or when the bacteria live in close association with the plant. In legumes and a few other plants, the bacteria live in small growths on the roots called nodules. Within these nodules, nitrogen fixation is done by the bacteria, and the NH₃ produced is absorbed by the plant. Nitrogen fixation by legumes is a partnership between a bacterium and a plant.

Biological nitrogen fixation can take many forms in nature including bluegreen algae (a bacterium), lichens, and free-living soil bacteria. These types of nitrogen fixation contribute significant quantities of NH₃ to natural ecosystems, but not to most cropping systems, with the exception of paddy rice. Their contributions are less than 5 lbs of nitrogen per acre per year. However, nitrogen fixation by legumes can be in the range of 25-75 pounds of nitrogen per acre per year in a natural ecosystem, and several hundred pounds in a cropping system (Lindermann, and Glover, 1998)

2.1.4 Mechanism of biological nitrogen fixation

Biological nitrogen fixation can be represented by the following equation, in which two moles of ammonia are produced from one mole of nitrogen gas, at the expense of 16 moles of ATP and a supply of electrons and protons (hydrogen ions):

N₂ + 8H+ + 8e^- + 16 ATP = 2NH₃ + H₂ + 16ADP + 16 Pi

This reaction is performed exclusively by prokaryotes (the bacteria and related organisms), using an enzyme complex termed nitrogenase. This enzyme consists of two proteins - an iron protein and a molybdenum-iron protein, as shown below.

The reactions occur while N₂ is bound to the nitrogenase enzyme complex. The Fe protein is first reduced by electrons donated by ferredoxin. Then the reduced Fe protein binds ATP and reduces the molybdenum-iron protein, which donates electrons to N₂, producing HN=NH. In two further cycles of this process (each requiring electrons donated by ferredoxin) HN=NH is reduced to H₂N-NH₂, and this in turn is reduced to 2NH₃.

Depending on the type of microorganism, the reduced ferredoxin, which supplies electrons for this process, is generated by photosynthesis, respiration or fermentation.

Figure 2.2: Mechanism of biological nitrogen fixation (Bezdicek, DF & Kennedy, AC, 1998)

2.1.5 The nitrogen-fixing organisms

All the nitrogen-fixing organisms are prokaryotes (bacteria). Some of them live independently of other organisms - the so-called free-living nitrogen-fixing bacteria. Others live in intimate symbiotic associations with plants or with other organisms (e.g. protozoa

Table 2.2: Examples of of nitrogen-fixing bacteria

Examples of nitrogen-fixing bacteria (* denotes a photosynthetic bacterium)
Free living		Symbiotic with plants
Aerobic	Anaerobic	Legumes	Other plants
Azotobacter Beijerinckia Klebsiella (some) Cyanobacteria (some)*	Clostridium (some) Desulfovibrio Purple sulphur bacteria* Purple non-sulphur bacteria* Green sulphur bacteria*	Rhizobium	Frankia Azospirillum

Source: Bezdicek, DF & Kennedy, AC, 1998

Rhizobia are well known for their capacity to establish a symbiosis with legumes. They inhabit root nodules, where they reduce atmospheric nitrogen and make it available to the plant.

Rhizobia encompass a range of bacterial genera, including Rhizobium, Bradyrhizobium, Sinorhizobium, Mesorhizobium, Allorhizobium and Azorhizobium, which are able to establish a symbiosis with leguminous plants (Dixon and Wheeler, 1986; Sprent and Sprent, 1990; Zahran et al., 1995; Haukka et al., 1996; Zahran, 1997b; Nick et al., 1999; Tan et al., 1999; Wang et al., 1999; Tighe et al., 2000; Zahran et al., 2000). They elicit the formation of specialized organs, called nodules, on roots or stems of their hosts, in which they reduce atmospheric nitrogen and make it available to the plant. Symbiotic nitrogen fixation is an important source of nitrogen, and the various legume crops and pasture species often fix as much as 200 to 300 kg nitrogen per hectare (Peoples et al., 1995).

Rhizobia generally achieve their maximum population density in, the rhizosphere of legumes (Rovira, 1961). When rhizobia achieve high numbers in the rhizosphere, their metabolic byproducts or antibacterial secretions may be in sufficient concentration to influence competition for nodulation.

Rhizobium species are found worldwide; they interact with a large diversity of plants even outside the Leguminosae and have been found as endophytes. For a long time Rhizobium and Agrobacterium have been recognized as being closely related genera (Graham, 1964). Several proposals for the inclusion of Agrobacterium in Rhizobium have emerged (Graham, 1964; Sawada, 1993; Pulawska et al., 2000 and Gaunt et al., 2001).

2.1.6 Symbiotic Nitrogen Fixation

Biological N2 fixation (BNF) is the major way for N input into degraded ecosystems. Rhizobium–legume symbioses represent the major mechanism of BNF in degraded lands, com-pared with the N2 -fixing heterotrophs and asso-ciative bacteria (Abdel-Ghaffar, 1989; Wullstein, 1989) and actinorhizal plants, e.g. Frankia-Ca-suarina (Sayed et al., 1997) and Frankia-Atriplex (Caucas and Abril, 1996) symbioses.

Globally, symbiotic nitrogen fixation has been estimated to amount to at least 70 million metric tons of nitrogen per year (Brockwell et al., 1995). In 1999, world consumption of fertilizer nitrogen was 88 million tons and apart from the consumption of nonrenewable energy sources, environmental pollution from fertilizer nitrogen escaping the root zone is high because in many cases nitrogen fertilizers are not used efficiently by crops (Peoples et al., 1994). Concomitant with N₂-fixation, the use of legumes in rotations offers control of crop diseases and pests (Robsons, 1990; Graham and Vance, 2000).

For a long time it was assumed that Rhizobium symbiosis was confined to leguminous plants, but a few other plants have now been found to bear nodules with a rhizobium as symbiont (Akkermans and Houwers, 1983). Non-legume Rhizobium symbioses have been proven in several species of Parasponia of Ulmaceae family (Trinick, 1973, 1976, 1979; Akkermans et al., 1978). Rhizobium symbioses have also been reported in a number of Zygophyllaceae (Sabet, 1946; Athar and Mahmood, 1972).

2.1.7 Rhizobia Inoculants

The inoculation of legumes with rhizobia has been practiced for the past century, primarily in industrialized countries. Inoculant production and use, both small- and large-scale, are well documented in the literature (Smith, 1987). Putting legume inoculation into practice requires substantial infrastructure for production, storage, and distribution, sometimes under stressful local conditions. The technology for producing effective inoculant can be implemented readily in developing countries, but extension of information and training is needed to teach growers for proper handling and use of inoculants (Sims et al., 1984).

Continued research is needed to identify the best rhizobial strains available for improved varieties of legumes. Despite our extensive experience, a better understanding is needed of the questions of when, where, and how much inoculant to apply to seed or soil to ensure effective colonization of roots in competition with soil rhizobia (Thies et al., 1991).

2.1.8 Inoculation and Rhizosphere Ecology

The establishment of genetically improved strains of rhizobia for nitrogen fixation will be unsuccessful if introduced organisms do not compete well with less-efficient symbiotic organisms already in the soil. Studies on the ecology of nitrogen-fixing microbes in the rhizosphere is an essential component of biological N fixation research. A better understanding of the microbial genes, plant genes, and other soil and plant factors influencing microbial ability to develop and function in the soil adjacent to (rhizosphere) and away from plant roots. The interaction of roots in mixed-cropping systems is poorly understood. The genetic bases of the beneficial interaction of nitrogen-fixing rhizobia with mycorrhizae and other microbes involved in biocontrol or plant growth promotion are not understood.

2.1.9 Factors Influencing Rates of BNF

The process of symbiotic N fixation generally depends on the overall vigour and productivity of the host plant. The environmental conditions that favour tree growth also favour higher rates of N fixation, including adequate water and temperature, high supplies of soil nutrients such as P and Ca and favourable soil pH (Binkley and Giardina, 1997). Giller and Wilson (1991) also reports that the ability of free living or symbiotic N₂-fixers to actually fix N₂ in the field is strongly influenced by the prevailing environmental conditions. Three main environmental constraints which frequently occur simultaneously are high temperature, water deficit and soil acidity (Hungria and Vargas, 2000). Soil degradation due to fragile soil structure, erosion, low soil organic matter and inappropriate land use technology found in the tropics are also influence the biological nitrogen fixation (Giller and Wilson, 1991). Some extensive reviews are reported the factors that affecting nitrogen fixation (Vincent, 1962, Gibson, 1976; Sprent, 1976; Vincent, 1980). The factors limiting N fixation by legumes can be divided broadly into 4 main categories: i) biological, ii) environmental, iii) nutritional, and iv) management. Hungria and Vargas (2000) reported that tropical environmental; conditions can constrain N fixation.

2.1.9.1 Biological Factors

2.1.9.1.1 The Bacterium

The important pre-requisite for effective nodulation and N fixation is the presence of an effective strain in the site. The survival and persistence of an adequate number of effective Rhizobium in soils independent of a hosts stimulant is essential to ensure nodulation of young seedlings. In spite of rhizobia being a natural inhabitant of the soil, many soils lack effective strains, especially those strains effective for legume species not previously grown on a particular soil.

2.1.9.1.2 Host Bacterial Factors

The host controls rhizobial rhizosphere stimulation, initial infection including specific site and potential number of infections and nodulation, including potential number, size the patterns of distribution of the nodules on the root system.

2.1.9.1.3 Efficiency of N Fixation

In theory, in a fully functioning system, energy utilization in fixing N in developing new nodule and in maintaining existing nodule tissue appears to balance the energy used in taking up nitrates and reducing them to ammonia. Gibson (1971) shows that under certain conditions and with certain legumes this is true. However, considerable variation in the efficiency with which different host/bacteria associations utilize carbon to fix N has been shown.

2.1.9.1.4 Longevity of Nodules

The amount of N fixed by a leguminous crop depends very much on the longevity of the nodules on its roots. A number of factors affect longevity including the strain of bacteria forming the nodules, environmental stress effects, the physiological condition of the plant and parasites in the nodules.

2.1.9.2 Environmental Factors

2.1.9.2.1 Temparature

Maximum soil temperatures in the tropics exceed 40^oC at 5 cm and 50^oC at 1 cm depth (Eaglesham and Ayanaba, 1984; Hafeez et al., 1991; Lal, 1993) and can limit nodulation (Day et al., 1978; Graham, 1981; Eaglesham and Ayanaba, 1984; Dudeja and Khurana, 1989). Excessive soil temperature can kill the majority of the bacteria in the surface layers of soil, though some Rhizobia can survive periods at 70^oC in dry soil (Marshall, 1964). Upper temperature limits for N-fixation in tropical legumes vary between 27^oC and 40^oC (Gibson, 1971; 1975; Dart, 1974). In relation to rhizobial growth upper limits range between 32 and 47^oC, though tolerance varies among species and strains (Pankhurst and Gibson, 1973; Gibson, 1975; Dart et al., 1976; Day et al., 1978; Munevar and Wollum, 1981a; La Favre and Eaglesham, 1986; Karanja and Wood, 1988b). Day et al. (1978) reported that the population of Rhizobia in Nigeria was only 4-40 cells g^-1 in the surface 5 cm of soil, whereas it was up to 10⁴ cells g^-1 at a depth of 20-25 cm below the surface.

The root infection process is probably the component most affected by high temperatures, with sensitivity located at the nodulation sites (Barrios et al., 1963; Pankhurst and Gibson, 1973; Frings, 1976; Lie 1974; 1981). High temperature also inhibits root hair formation, reducing the number of sites for nodulation (Jones and Tisdale, 1921; Frings, 1976); root hair penetration and infection-thread formation (Joffe et al., 1961; Barrios et al., 1963; Pankhurst and Gibson, 1973; Dart, 1974; Frings, 1976). Nodule initiation, rhizobial release from the infection thread, and bacteroid developments are also affected by temperature (Roughley, 1970; Pankhurst and Gibson, 1973; Vincent, 1980). High temperatures may affect nodule senescence (Pankhurst and Gibson, 1973; Sutton, 1983; Hungria and Franco, 1993) and nodule function (Jones and Tisdale, 1921; Dart and Mercer, 1965; Lindemann and Ham, 1979; Munevar and Wollum, 1981b; Hungria et al., 1989; Piha and Munns, 1987; Hernendez-Armenta et al., 1989 and, Hungria and Franco, 1993).

Indirect effects of high temperature on the metabolism of the host plant and direct effects on N2 fixation have been recognized for a long time (Jones and Tisdale, 1921). However, the overall balance between photosysnthesis and respiration determines levels of N₂ fixation (Hungria and Vargas, 2000).

2.1.9.2.2 Light

The effects of light on N fixation can be ascribed to its effects on photosynthesis and to photoperiodic effects. A certain light intensity is necessary for maximum nodulation of seedlings and further nodule development of established plants. Decreased day length, effects of shading (Sprent, 1973) or excessive cloud cover all have adverse effects on nodulation. Gibson (1976) has shown that by transferring legumes from a low to high light intensity results in a rapid increase of both nodule size and number.

2.1.9.2.3 Moisture or Water Stress

In tropical areas, usually at matrix water potentials of –0.5 to–1.5 MPa, growth and nitrogen fixation are affected. Water stress affects rhizobial survival and growth and population structure in soil, the formation and longevity of nodules, synthesis of leghaemoglobin and nodule function (Hungria and Vargas, 2000). Severe stress may lead to irreversible cessation of n fixation (Sprent, 1971; Vincent, 1980; Walker and Miller, 1986; Venkateswarlu et al., 1989; Stamford et al., 1990; Guerin et al., 1991). However, rhizobial strains (Boonkerd and Weaver, 1982) and plant species (Walsh, 1995) vary in relation to drought tolerance.

Studies has shown that decrease in nitrogenase activity under drought conditions occur in a two stage process- i) there is a decrease in nodule cortical permeability, enforcing a primary limitation in oxygen supply to the bacteroid zone, restricting respiration and resulting in a simultaneous decrease in nitrogenase activity (Sprent, 1976; Durand et al., 1987; Walsh, 1995); ii) the second step occurs when nodule activity is already drastically reduced, and nitrogenase activity can be constrained by other factors, e.g. leghaemoglobin content (Serraj et al., 1999). Dixon and Wheeler (1983) reported that low water supply might affect N fixation by lowering plant carbohydrate supply, carbohydrate transportation to the nodules, or direct impairment of nodule development and activity. Srivastava and Ambasht (1994) found that Casuarina nodule mass increased by fourfold from the end of the dry season (June) into the wet season (July and August) in Egypt. Binkley and Giardina (1997) reported that soil moisture variations accounted for about 80% of the variations in N fixation rates.

2.1.9.2.4 Soil Atmosphere

The average composition of the principal active gases in the soil atmosphere are: oxygen- 20.3%, nitrogen-79.2%, carbon-di-oxide- 0.5% but there is considerable local variability affected by microbial and or plant root respiration. High moisture levels, poor soil structure and finely textured soils lowers the air content of the soil and slow gaseous diffusion maintains low concentrations of these gases in the soil atmosphere surrounding roots and nodules.

2.1.9.2.5 Soil Acidity

The tropical hill soils may be highly leached, acidic and infertile, and contain toxic concentrations of available aluminium (Al) and manganese (Mn). Acid soils pose problems for the plant, the bacteria and the symbiosis (Giller and Wilson, 1993). Low soil pH is often associated with increased Al and Mn toxicity and reduced Ca supply. In addition to Al toxicity, acidity may also be related to phosphorus (P), and molybdenum (Mo) deficiencies (Hungria and Vargas, 2000). These stresses affect the growth of rhizobia (Cooper et al., 1983; Coventry and Evans, 1989; Campo, 1995), of the host legume (Andrews et al., 1973; Kim et al., 1985), and symbiosis (Murphy et al., 1984; Brady et al., 1990; Campo, 1995).

The optimal pH for rhizobial growth is considered to be between 6.0 and 7.0 (Jordan, 1984) and relatively few rhizobia grow well at pH less than 5.0 (Graham et al., 1994). However, some rhizobia can tolerate acidity better than others, and tolerance may vary among strains within a species (Lowendorff, 1981; Vargas and Graham, 1988: Brockwell et al., 1995).

Acidity affects early steps in the infection process, including the exchange of molecular signals between symbiotic partners and attachment to the roots (Hungria and Vargas, 2000). Other stages of nodule establishment and function are also impacted by acidity, as is the growth of the host plant (Graham, 1981; Munns, 1986).

Liming has been considered the most efficient practice in overcoming soil acidity, with some of the benefits to legume crops not only due to increased soil pH, but also to increased availability of Ca to plant, bacteria and the symbiosis. Ca may affect rhizobial cell wall integrity (Bergersen, 1961; Vincent, 1962; de Maagd et al., 1989; Ballen et al., 1998), some enzymes (Norris et al., 1991; Ballen et al., 1998), or membrane transport systems (O’Hara et al., 1989).

2.1.9.2.6 Salinity

Plant development is hindered by the soil’s high salt content. Trees vary widely in their salt tolerance. In reforestation programmes of salt affected areas is to select tree species that are salt tolerant. Than the approach will inoculate the host plant with a micro-organism that would improve the salt tolerance of the plant (Reddell et al., 1986). Some legumes are relatively salt tolerant but high concentrations can damage the nodules indirectly by withdrawing water osmotically from nodules or in situations of high evaporation rates, can damage nodules directly by depositing salts on the nodules and roots (Sprent, 1971). Apart from the effects of NaCl salinity, carbonates and bicarbonates principally of Na have been shown to affect the symbiosis (Subba Rao et al., 1980).

2.1.9.3. Mineral Nutrition

2.1.9.3.1 Combined Nitrogen

Combined N added to the soil through fertilization can be a limiting factor that affects the symbiotic relationship between rhizobia and their partners (Gibson, 1981). It is widwly accepted that the capacity for nitrogen fixation by a nodulating or nodulated legume is influenced, at least in two ways, by mineral nitrogen in the soil in which it is grown. First, the process of nodulation may be promoted by relatively low level of available nitrate or ammonia, higher concentrations of which almost always depress nodulation (Davidson and Robson, 1986; Dixon and Wheeler, 1986). Second, the rate of N₂ fixation by an active, growing, and well-nodulated legume is always suppressed by NO₃^- ions (Hardina and Silsbury, 1989). It was suggested that nitrogen fertilizers added to soil may delay the symbiotic process through decreased multiplication of free-living rhizobia (Dart, 1974). Early as well recent reports show that N fertilizers inhibit at least three phases of legume nodulation, viz. root-hair infection (Abdel-Wahab et al., 1996) nodule initiation, and growth and development (Atkins et al., 1984; Imsande, 1986), as well as the level of nitrogenase activity (Sanginga et al., 1989; Purcell and Sinclair, 1990; Arreseigor, 1997) and promote premature nodule senescence (Munns, 1977; Gibson and Harper, 1985; Abaido et al., 1990).

N fixation should reduced or eliminated when the soil have high supplies of ammonium and nitrate (MacDicken, 1994). Sanginga et al. (1989) found N fertilization at the rate of 40 – 80 kg N ha^-1 on L. leucocephala reduced the nodule mass and N fixation. Goi et al. (1992) found that both nodulation and growth of Acacia auriculiformis were stimulated by ammonium, but impaired by nitrate additions. Peoples et al. (1994) found that the ¹⁵N: ¹⁴N ratio in L. leucocephala leaves declined with plantation age and Dommergues (1995) inferred that this patterns indicated a decline in N fixation as soil N supply increased. Dommergues (1995) reported that N fixation in tropical trees declines as soil N supplies increase.

But, Ingestad (1980) found that frequent and low concentration nitrate supply increased N fixation by grey alder (Alnus incana).

2.1.9.3.2 Phosphorus

Phosphorus availability can be a major limiting factor for plant growth, particularly in tropical soils.In addition, P deficiency has a stronger effect on N-fixing legumes than on other plants because of the high energy costs of N₂ fixation, which requires a greater quantity of inorganic P (Israel, 1987; Vadez et al., 1999). In bean, a legume of tropical origin, low P availability can seriously depress yield. Strategies to improve crop growth under P deficiency (Clarksonm, 1985; Vadez and Drevon, 2001;) include promoting the efficient uptake of P, or favouring effective use of endogenous P. The latter strategy involves both the appropriate distribution as well as the optimal utilization of P by the plant, resulting in greater biomass production and more N₂ fixed per unit of P taken up.

Probably the most common limiting nutrient in many areas is P. Legumes require larger amounts of P than any other plant and its effects are complex acting on nodulation, nitrogen fixation and plant growth. Nitrogen fixing plants have a high demand for phosphorus (Israel, 1987) and this is also applicable for tropical N fixing trees. P stimulates root growth, the number and density of nodules are greatly increased, N fixation begins earlier and the amount of N accretion per unit of nodule tissue is higher with legumes well supplied with P (Munns, 1977). Sun et al. (1992) also found that Acacia mangium growth and N fixation increased with P application. However, the supply of P to trees may depend heavily on soil pH and exchangeable aluminium (Binkley and Giardina, 1997).

2.1.9.3.3 Other Nutrients

Under adequate conditions of P. K stimulates infection and N fixation but high levels of K are inhibitory if P levels are low. According to Vincent (1962) a higher K requirement is found in older established pastures but again no direct role has been demonstrated for the symbiotic state though it is though that K may stimulate nodule activity by improving carbohydrate supplies.

Similarly, S deficiency effects nodulation by reducing nodule number, size and N fixation (Munns, 1977). It has been shown that S deficiency leads to a failure of protein synthesis whether nitrogen is available to the plant symbiotically or in a combined form (Vincent, 1962). A number of disorders due to deficiency of trace elements have been shown to affect the growth of legumes (Hewitt, 1958) including Ca, Mg, Mn, Al, etc (Cooper et al., 1983; Brady et al., 1990; Campo, 1995).

2.1.9.4 Agriculture and Forestry Management Systems

Management practices are those designed to improve plant growth in general through improved soil structure, irrigation, soil fertility, weed control and plant disease control. These factors increase nitrogen fixation through the improved growth of the host legume. Another important factor involved management practices improving N fixation directly either by improving nodulation or N fixing efficiency.

2.1.10 Approaches for the Estimation of Nitrogen Fixation

The process of symbiotic N fixation generally depends on the overall vigour and productivity of the host plant. Rates of biological N-fixation are highly variable and depend on the main effects and interactios of the genetics of the host, the microsymbionts and environmental conditions (Binkley and Giardiana, 1997). In plantations four approaches are used for estimating the rates of N-fixation. These are 1) N accretion, 2) Acetylene reduction, 3) ¹⁵N natural abundance, and, 4) ¹⁵N labelling (Silvester, 1983; Reporter, 1985; Caldwell and Virginia, 1991).

N accretion is determined by examining changes in the total N content of a stand, or contrasting the total N content of an N-fixing stand with a non-N-fixing stand. Estimates of accretion may under represent the actual rate of nitrogen fixation if losses of N are substantial through leaching losses, denitrification or volatilization etc. These estimates gauge the total input of N across the life of the stand and may not represent current activity.

The Acetylene Reduction approach takes advantage of the nitrogenase enzymes preference for reducing acetylene to ethylene (C₂H₂ + 2H ® C₂H₄) over reducing dinitrogen to ammonia. In this approach, nodules are excavated from the soil, incubated in a closed jar with acetylene and then the ethylene produced is measured by gas chromatograohy. A conversion from acetylene reduced to N2 reduced may be applied. But, this approach is subject to many uncertainties, including the effect of acetylene on nitrogenase activity and the uncertainties in variation in rates and nodule mass (Silvester, 1983). Acetylene reduction may or may not provide quantitative estimates of N fixation, but this approach is well suited for examining patterns of variation (Srivastava and Ambasht, 1994).

Nitrogen from fixation has a ratio of stable isotopes (¹⁵N:¹⁴N) that is similar to the ratio in the atmosphere, whereas the natural abundance of these isotopes in soils often shows enrichment of ¹⁵N. Where the isotope ratio of newly fixed N differs from the ‘prior’ N, these ratio differences can be used to estimate the rate of N fixation, as well as the addition of newly-fixed N to various ecosystem pools. The small differences in isotope ratios are usually reported as s¹⁵N, which is the parts-per-thousand enrichment in ¹⁵N of the sample relative to the ratio in the atmosphere.

The ‘prior’ soil pool of N may also be labelled through addition of ¹⁵N and any additional N provided by N fixation then dilutes the ratios of ¹⁵N:¹⁴N. This approach is expensive, but not beyond the reach of many research budgets (Baker and Mullin, 1992).

2.1.11 Biological Nitrogen Fixing (BNF) Species

The plants in the Leguminosae family are found in temperate and tropical environments. They represent a diverse group, ranging from small plants of clover to large Albizia species. The group is estimated to contain between 16,000 and 19,000 species in about 750 genera and most of those examined fix nitrogen in symbiosis with Rhizobia (Allen and Allen, 1981). Approximately 3,000 species of legumes examined, more than 90 percent form root nodules (in which nitrogen fixation presumably occurs in symbiosis with rhizobia). Nitrogen fixation by rhizobia is of great importance in agriculture and forestry in several ways.

Symbiotic nitrogen fixation in legumes allows them to grow well without the addition of nitrogen fertilizer. However, it may be necessary to apply phosphorus and other deficient nutrients, as well as lime to alleviate soil acidity (MacDicken, 1994: Binkley and Giardina, 1997). However, the successful biological N fixation for soybeans, beans, pea nuts, sugarcane and forage legumes are reported elsewhere (Dobereiner, 1997).

2.1.12 Biological Nitrogen Fixation in Agricultural Crops

Nitrogen fixation by rhizobia is of great importance in agriculture in several ways. Because few have been exploited for food, there is the prospect that the utilization of legumes could be expanded substantially. It is estimated that about 20 percent of food protein worldwide is derived from legumes. The researches on BNF of dietary used legumes are done by various authors dry bean (Phaseolus vulgaris), dry pea (Pisum sativum), chickpea (Cicer arietinum) (Solaiman, 1999), broad bean (Vicia faba), pigeon pea (Cajanus cajan), cowpea (Vigna unguiculata) (Singleton et al., 1990), and lentil (Lens culinaris) (Agostini and Khan, 1986). Peanut (Arachis hypogaea) and soybean (Glycine max) (Mpepereki et al., 2000; Hungria and Vargas, 2000) are dominant sources of cooking oil and protein. They are also major food sources in some regions.

2.1.13 Biological Nitrogen Fixation in Forestry Programmes

Biological N fixation studies, especially on fundamental processes have been restricted mainly to a limited number of agriculturally important annual crops like peas and beans (Akkermans and Houwers, 1983). However, recently there has been increasing interest in nodulated leguminous trees and shrubs because of their promising economic importance in the tropical countries (NAS, 1977, 1980).

Several nitrogen-fixing tree and shrub species are particularly advantageous for reforestation and agroforestry in having the ability to grow fast, tolerate soil acidity and withstand regular coppicing. Sesbania sesban, Gliricidia sepium, Leucaena leucocephala and Erythrina poeppigiana are commonly recommended for agroforestry systems in the humid tropics. They provide a variety of products including fruit, forage, fuelwood, fodder, fence and timber material and are used in soil improvement, land reclamation, and forestry programs (Allen and Allen, 1981; Giller and Wilson, 1991)

.

Nitrogen fixing trees in tropical environments appear to offer both high growth rate and soil enrichment (Binkley and Giardina, 1997). Nitrogen fixing trees may increase the supply of available N in the soil, benefiting both N-fixing and non-N fixing trees (Binkley et al., 1992). There is a wide range of nitrogen fixing plants that have been used in forestry with an objective of raising soil nitrogen levels and subsequently improving the growth of the non-nitrogen fixing forest species (Turvey and Smethurst, 1983). But still research on nitrogen-fixing forest trees in general have lagged behind that on food, feed, and forage crops. Cleveland et al. (1999) found that for tropical evergreen forests estimates of N fixation are extremely rare and highly variable. Some shrub and tree species form a symbiosis with Frankia, a nitrogen-fixing actinomycete that only relatively recently has been cultured in vitro. The other nitrogen-fixing trees are legumes and form a symbiosis with Rhizobia. The major research need to quantify nitrogen fixation by trees is still at the methodological stage, though this is very important for developing countries.

A wide variety of N-fixing trees are available for use in plantations, but only few species have received much attention for silviculture and management (Giller and Wilson, 1991; Gutteridge and Shelton, 1994; Binkley and Giardina, 1997). Allen and Allen (1981) well covered the many uses of legume trees. However, there are some reports regarding the ecological and silvicultural aspects of biological nitrogen fixation of tropical tree species (Werner and Muller, 1990; MacDicken, 1994) and temperate species of Alnus rubra (Hibbs et al., 1994). There is now increasing interest in non-timber uses of trees, including nitrogen-fixing legumes, many of these uses are medicinal (Johnston, 1998).

2.1.14 Importance of Biological Nitrogen Fixation (BNF)

Biological nitrogen fixation is an essential natural process that supports life on this planet. Nitrogen fixation occurs both biologically and non-biologically. Asymbiotic and symbiotic biological systems fix an estimated 100-175 million metric tons of nitrogen annually (Burns and Hardy, 1975). Non-biological nitrogen fixation occurs through the effects of lightning, and in industry primarily by the Haber- Bosch

Figure 2.3 : The Importance of Biological Nitrogen Fixation (Bezdicek, DF & Kennedy, AC, 1998)

process which requires high levels of fossil fuel. Worldwide, lightning may fix 10 million metric tons of nitrogen year^-1. Industrial fixation for fertilizer nitrogen has increased from 3.5 million tons in 1950 to 80 million tons in 1989 in response to the needs of high-yielding crops. Burns and Hardy (1975) estimated the annual amount of nitrogen fixed biologically in various systems (Table 2.3).

Table 2.3: Estimated annual amount of nitrogen fixed biologically in various systems (Burns and Hardy, 1975).

System	Nitrogen fixed (t –1 yr x 106 )
Legumes Non- legumes Permanent grassland Forest and Woodland Unused land Total land Sea	35 9 45 40 10 139 36
TOTAL	175

2.1.15 Benefits of Using BNF

2.1.15.1 Economics

Greater use of BNF can help lower production costs and thus increase profit margins for producers. The use of crops that fix nitrogen in crop rotations can significantly reduce N fertilizer needs for crops in the rotation. Thus, the development and adoption of new biological N fixation technology can become a real technology for economic growth and the society, as a whole will be the beneficiary. Some nitrogen fixation estimates for tropical plantations are shown in the Table below.

2.1.15.2 Environment

Greater use of BNF can help enhance environmental quality by reducing problems with air and water pollution. BNF is part of responsible natural resource management.

2.1.15.3 Efficiency

Greater use of BNF will reduce society's current dependence on synthetic N fertilizers. The production of widely used synthetic N fertilizers such as anhydrous ammonia requires the use of relatively large amounts of energy from non-renewable energy sources such as natural gas. Distribution and application of these fertilizers also requires relatively large amounts of non-renewable energy sources such as diesel fuel.

Table 2.4 : Nitrogen fixed by some common legume tree species.

Species and location	N fixation kg ha-1 yr-1	References
Albizia falcataria, Hawaii Albizia lebbeck, Puerto Rico Alnus jorullensis, Columbia Alnus nepalensis, India Casuarina equisetifolia, Puerto Rico C. equisetifolia, India C. equisetifolia, Senagal C. equisetifolia, Senagal C. equisetifolia, Senegal Leucaena leucocephala, Puerto Rico L. leucocephala, Tanzania L. leucocephala, Nigeria	65 - 140 260 280 30-120 80-90 in pure Casuarina 40-60 in Casuarina/ Eucalyptus mix 90 in 5 yr plantation 60 in 8 yr plantation 60 35-40 75 50-70 in mixed 100 in monocultures 110 ± 30 150 - 180	Binkley and Giardina, 1997 Parrotta, 1992 Carlson and Dawson, 1985 Sharma and Ambasht, 1988 Parrotta et al. 1994 Srivastava and Ambasht, 1994 Dommergues, 1963 Mariotti et al., 1992 Mailly and Margolis, 1992 Parrotta et al., 1994 Hogberg and Kvarnstrom, 1982 Sanginga et al., 1989, 1994

2.1.15.4 Increased fertility:

Greater use of BNF can help enhance sustainable food production by improving soil fertility and tilth. Some producers have found that the use of so called green manure crops is a more sustainable fertilizer alternative than purchased, synthetic fertilizer. Green manure crops are crops grown specifically to be incorporated into the soil rather than for harvest. Using green manure crops that fix atmospheric N₂ may potentially increase soil N levels and organic matter content over time. Additional organic matter in soils generally improves the tilth of a soil, where tilth refers to desirable physical properties of soil such as proper drainage, water holding capacity, aeration, and structure.

2.1.15.5 Sustainability

Using BNF is part of wise management of agricultural systems. The economic, environmental, and agronomic advantages of BNF make it a cornerstone of sustainable agricultural systems.

2.1.15.6 Livelihoods

For many poor farmers, BNF is an essential, cost-effective alternative or complementary solution to industrially manufactured N fertilizers, particularly for staple crops. Many grain legumes are major sources of protein for the subsistence of poorest farmer households. When legume production exceeds household requirements, it can be readily traded to generate income, making significant and direct contributions to livelihoods.

2.2 The Nitrogen Cycle

Figure 2.4: The basic Nitrogen cycle (Internet)

The diagram below shows an overview of the nitrogen cycle in soil or aquatic environments. At any one time a large proportion of the total fixed nitrogen will be locked up in the biomass or in the dead remains of organisms (shown collectively as "organic matter"). So, the only nitrogen available to support new growth will be that which is supplied by nitrogen fixation from the atmosphere (pathway 6 in the diagram) or by the release of ammonium or simple organic nitrogen compounds through the decomposition of organic matter (pathway 2). Some of other stages in this cycle are mediated by specialized groups of microorganisms and are explained below.

Figure 2.5: An overview of the nitrogen cycle in soil or aquatic environments (Bezdicek, DF & Kennedy, AC, 1998)

2.2.1 Nitrification

The term nitrification refers to the conversion of ammonium to nitrate (pathway 3-4). This is brought about by the nitrifying bacteria, which are specialised to gain their energy by oxidising ammonium, while using CO₂ as their source of carbon to synthesise organic compounds. Organisms of this sort are termed chemoautotrophs - they gain their energy by chemical oxidations (chemo-) and they are autotrophs (self-feeders) because they do not depend on pre-formed organic matter. In principle the oxidation of ammonium by these bacteria is no different from the way in which humans gain energy by oxidizing sugars. Their use of CO₂ to produce organic matter is no different in principle from the behavior of plants.

The nitrifying bacteria are found in most soils and waters of moderate pH, but are not active in highly acidic soils. They almost always are found as mixed-species communities (termed consortia) because some of them - e.g. Nitrosomonas species - are specialized to convert ammonium to nitrite (NO₂^-) while others - e.g. Nitrobacter species - convert nitrite to nitrate (NO₃^-). In fact, the accumulation of nitrite inhibits Nitrosomonas, so it depends on Nitrobacter to convert this to nitrate, whereas Nitrobacter depends on Nitrosomonas to generate nitrite.

The nitrifying bacteria have some important environmental consequences, because they are so common that most of the ammonium in oxygenated soil or natural waters is readily converted to nitrate. Most plants and microorganisms can take up either nitrate or ammonium (arrows marked "1" in the diagram). However, process of nitrification has some undesirable consequences. The ammonium ion (NH₄⁺) has a positive charge and so is readily adsorbed onto the negatively charged clay colloids and soil organic matter, preventing it from being washed out of the soil by rainfall. In contrast, the negatively charged nitrate ion is not held on soil particles and so can be washed down the soil profile - the process termed leaching (arrow marked 7 in the diagram). In this way, valuable nitrogen can be lost from the soil, reducing the soil fertility. The nitrates can then accumulate in groundwater, and ultimately in drinking water. There are strict regulations governing the amount of nitrate that can be present in drinking water, because nitrates can be reduced to highly reactive nitrites by microorganisms in the anaerobic conditions of the gut. Nitrites are absorbed from the gut and bind to haemoglobin, reducing its oxygen-carrying capacity. In young babies this can lead to respiratory distress - the condition known as "blue baby syndrome". Nitrite in the gut also can react with amino compounds, forming highly carcinogenic nitrosamines.

2.2.2 Denitrification

Denitrification refers to the process in which nitrate is converted to gaseous compounds (nitric oxide, nitrous oxide and N₂) by microorganisms. The sequence usually involves the production of nitrite (NO₂^-) as an intermediate step is shown as "5" in the diagram above. Several types of bacteria perform this conversion when growing on organic matter in anaerobic conditions. Because of the lack of oxygen for normal aerobic respiration, they use nitrate in place of oxygen as the terminal electron acceptor. This is termed anaerobic respiration and can be illustrated as follows:

In aerobic respiration (as in humans), organic molecules are oxidized to obtain energy, while oxygen is reduced to water:

C₆H₁₂O₆ + 6 O₂ = 6 CO₂ + 6 H₂O + energy

In the absence of oxygen, any reducible substance such as nitrate (NO₃^-) could serve the same role and be reduced to nitrite, nitric oxide, nitrous oxide or N₂.

Thus, the conditions in which we find denitrifying organisms are characterized by (1) a supply of oxidisable organic matter, and (2) absence of oxygen but availability of reducible nitrogen sources. A mixture of gaseous nitrogen products is often produced because of the stepwise use of nitrate, nitrite, nitric oxide and nitrous oxide as electron acceptors in anaerobic respiration. The common denitrifying bacteria include several species of Pseudomonas, Alkaligenes and Bacillus. Their activities result in substantial losses of nitrogen into the atmosphere, roughly balancing the amount of nitrogen fixation that occurs each year.

2.2.3 The Nitrogen Cycle and the environment

In most natural systems available or "fixed" nitrogen is usually the limiting factor in plant growth. This realization led to the invention and massive use of nitrogen fertilizers during the 20th century and ever-increasing crop yields per acre of farmed land. Without this use of nitrogenous fertilizers, the Earth could not support its current population of 6.1 billion people.

At the same time, the widespread use of fossil fuels release not only carbon dioxide, but also nitrogen oxides as well. These nitrogen oxides contribute to urban photochemical smog and acid precipitation. The combined effect of these two anthropogenic processes, agriculture and fossil fuel combustion, is similar to natural nitrogen fixation. These substantial modifications to the global nitrogen cycle have important implications in a number of areas including global warming, stratospheric ozone destruction, photochemical smog, and regional eutrophication.

This trend will result in increased levels of nitrogen in soils, natural waters, crop residue, and municipal and agricultural wastes, and there is national and international concern about its potential adverse effects on environmental quality and public health.

Briefly, the concern is that nitrogen fertilization and livestock wastes may lead to a significant increase in nitrous oxide (N2O) concentration in the troposphere and thereby cause the destruction of the stratospheric ozone layer and add to the greenhouse effect associated with the continuing increase in atmospheric concentrations of CO2 and CH4.

Also, there is concern that the intensive use of fertilizer and the application of livestock wastes will lead to increased N levels in ground water and surface waters, and that this in turn will lead to increased eutrophication of lakes and stream, and to health hazards to humans through nitrate (NO3 -) enrichment of drinking water.

However, nitrogen is essential to life. It is a fundamental component of the class of molecules called amino acids. These are the vital 'building blocks' of proteins. Proteins have a variety of functions including those of natural catalysis, chemical messengers within an organism, and.the transport and storage of small molecules, such as oxygen. Nitrogen is an essential component of the bases that make up DNA. This is the molecule that carries the genetic code of all living creatures. Nitrogen has other biologically significant roles. For instance, there are organisms that are capable of utilizing nitrogen in its oxidized forms as a substitute for oxygen during respiration, while there are others that can oxidize reduced nitrogen with oxygen, to liberate energy.

In this way, nitrogen is a double-edged sword that can be used for enrichment of human life as well as cause environmental destruction.

2.3 Nitrogen Fixation by Legumes

Nitrogen (N) is an essential element for all living organisms and nitrogen is the main component of the atmosphere on earth. Nitro-genmoves in a cycle from the atmosphere into the soil and plants, through the food chain of animals and man, and back to the atmosphere. Protein is the principle nitrogen compound that has an important role in the growth and development of plants and animals. One group of bacteria called rhizobia are able to convert the atmospheric nitrogen (N₂ gas) to ammonia (NH₃) within root nodules of legumes. This process is called biological nitrogen fixation (BNF). Biological nitrogen fixation is a type of symbiosis resulting in benefits to both plants and bacteria.

Figure 2.6: How BNF works in legumes (Elevitch and Wilkinson, 1998)

2.3.1 Legume Nodules

Legume nodule is an organ of primary N assimilation, comparable in rate of metabolic activity to the leaf, the organ of primary C assimilation (Walsh, 1995). Legume nitrogen fixation starts with the formation of a nodule (Anon, 2002). A common soil bacterium, Rhizobium, invades the root and multiplies within the cortex cells. The plant supplies all the necessary nutrients and energy for the bacteria. Within a week after infection, small nodules are visible with the naked eye (Anon, 2002). In the field, small nodules can be seen 2-3 weeks after planting, depending on legume species and germination conditions. Nodules are small, knob-like outgrowths on roots of some legumes. In nodules, the ammonia is converted to amino acids. These amino acids are then transported out of the nodules to other parts of the host plant where the N compounds undergo further changes, mainly to produce proteins (Marutani et al., 1994). Legume BNF are assessed by nodule color, observations of plant growth, and field and pot trails (Singleton et al., 1990).

2.3.2 Observation of stem and root nodules

Stem nodulation occurs in the genera Aeschynomene (several species) and Sesbania (S. rostrata only), two genera of legumes that have a capacity to grow in waterlogged conditions. Nodulation can be induced along the whole length of the stem, sometimes up to three meters above the ground. Each of these plants has the capacity to form root nodules. Stem nodulation was earlier reported to occur also with Neptunia oleracea, but in fact the observed nodules from on adventitious roots that arise from the stem of waterlogged plants Giller and Wilson, 1991).

The nodules can be detached from the roots, unlike the galls, which are actually root swellings (Singleton et al., 1990). Root nodules have characteristics shapes, depending on the legume species. Nodules are generally observed by dissecting them by a razor blade to establish whether they are red, pink (effective nodulation), brown, dark or green in color ineffective nodulation). Effective nodules are usually large and clustered on the primary and upper lateral roots. In contrast, ineffective nodules are often small, numerous, and distributed throughout the root system (Singleton et al., 1990) When nodules are young and not yet fixing nitrogen, they are usually white or grey inside. Legume nodules that are no longer fixing nitrogen turn green and may be

discarded by plants as the contents of the nodules are used up or may be the result of an inefficient Rhizobium strain or poor plant nutrition leading to the collapse and

of the nodules (Anon, 2003; Tchoundjeu, 2003). Collapsed, brown and black nodules

Figure 2.7 : Root nodules (Right) and Stem nodule (Left)

are regarded as dead (Fownes and Anderson, 1991). As nodules grow in size they gradually turn pink or reddish in color, indicating nitrogen fixation has started. The pink or red color is caused by leghemoglobin (similar to hemoglobin in blood) that controls oxygen flow to the bacteria (Anon, 2002; Lindermann and Glover, 1998).

2.3.3 Shape of Leguminous nodules

Classification of nodules for shape was based on mature nodules, small globose nodules in the presence of other shapes being ignored as juvenile (Corby, 1971).

Figure 2.8 Shapes of Leguminous Nodules (Source: Corby, 1971)

Table 2.5: Shape of leguminous root nodules

Subfamily And tribe	Numbers examined			Percentage of records within description	Shape of mature nodules
	Genera	Species	Records
Papilionatae
Galegeae	9	84	135	89	Globose to elongate, rarely two-branched, delicate
Genisteac	8	71	108	98	Coralloid
Hedysareae	12	40	68	99	Semi-globose to globose; globose forms often thick-necked
Phaseoleae	22	84	118	97	Globose to oblate, usually thin-necked
Dalbergieae	6	15	21	95(100)	Globose (to elongate)
Sophoreae	4	4	4	75(100)
Loteae	1	2	3	100	Globose to prolate
Mimosoideae					Globose to elongate, with or without branching, robust
Acacieae	1	27	38	100
Adenanthereae	2	2	4	100
Eumimoseae	1	1	1	100
Ingeae	1	11	17	88
Piptadenieae	2	3	8	100
Caesalpiniodeae
Cassieae	1	9	17	94
Swartzieae	1	1	2	100
Dimorphandreae	1	2	6	100	Globose

Source: Corby, 1971

2.3.4 Root nodule formation

Sets of genes in the bacteria control different aspects of the nodulation process. One Rhizobium strain can infect certain species of legumes but not others e.g. the pea is the host plant to Rhizobium leguminosarum biover viciae, whereas clover acts as host to R. leguminosarum bivour trifolii. Specificity genes determine which Rhizobium strain infects which legume. Even if a strain is able to infect a legume, the nodules formed may not be able to fix nitrogen. Such rhizobia are termed ineffective. Effective strains induce nitrogen-fixing nodules. Effectiveness is governed by a different set of genes in the bacteria from the specificity genes. Nod genes direct the various stages of nodulation.

The initial interaction between the host plant and free-living rhizobia is the release of a variety of chemicals by the root cells into the soil. Some of these encourage the growth of the bacterial population in the area around the roots (the rhizosphere). Reactions between certain compounds in the bacterial cell wall and the root surface are responsible for the rhizobia recognizing their correct host plant and attaching to the root hairs. Flavonoids secreted by the root activate the nod genes in the bacteria that then induce nodule formation. The whole nodulation process is regulated by highly complex chemical communications between the plant and the bacteria.

Once bound to the root hair, the bacteria excrete nod factors. These stimulate the hair to curl. Rhizobia then invade the root through the hair tip where they induce the formation of an infection thread. This thread is constructed by the root cells and not the bacteria and is formed only in response to infection. The infection thread grows through the root hair cells and penetrates other root cells nearby often with branching of the thread. The bacteria multiply within the expanding network of tubes, continuing to produce nod factors, which stimulate the root cells to proliferate, eventually forming a root nodule. Within a week of infection small nodules are visible to the naked eye. Each root nodule is packed with thousands of living Rhizobium bacteria, most of which are in the misshapen form known as bacteroids.

Portions of plant cell membrane surround the bacteroids. These structures, known as symbiosomes, which may contain several bacteriods or just one, are where the nitrogen fixation takes place.

Figure 2.9 Formation Root nodules

2.3.5 Size of leguminous nodules

Nodules on annual legumes such as beans, peanuts, and soybeans are round and can reach the size of a large pea (Lindermann and Glover, 1998). Nodules on annuals are short-lived and will be replaced constantly during the growing season. At the time of pod fill, nodules on annual legumes generally lose their ability to fix nitrogen because the plant feeds the developing seed rather than the nodule. Beans will generally have less than 100 nodules per plant, soybeans will have several hundred per plant, and peanuts may have 1,000 or more nodules on a well-developed plant (Lindermann and Glover, 1998).

2.3.6 Nodulation Status of tropical woody legumes

There are about 650 genera and 18,000 species in the Leguminosae, the third largest family of flowering plants (Polhill et al., 1981). Although taxonomists are not entirely in agreement, Sprent (1995) followed the classification of Polhill et al. (1981), which devides the legumes into three sub-families. The Caesalpinioideae, with about 1900 species, are mainly woody and mainly tropical; the Mimosoidae (2713 species) also are mainly woody and extend into sub-tropical and temperate regions. The largest sub-family, the Papilionoideae has about 13,000 species and is largely herbaceous, but about 4000-5000 species are woody and of these, most are tropical (Southerland and Sprent, 1993).

The nodulation status, nodule ultrstructure and nodule nitrogen nitrogen fixing activity of wild legumes are very important characterstics which should be determined when evaluating the contribution of these legume s to soil fertility. Of the family as a hole, about 57% of genera and 20% of species have been examined for nodulation (de Faria et al., 1989); most of those which have not been examined are tropical. Within the three sub-families, nodule morphology and anatomy are generally consistent within tribes (Corby, 1981; Sprent et al., 1989; de Faria et al., 1989).

Allen and Allen (1981) published a classic and fascinating book on legumes summarizing not only what was known about nodulation at that time, but also the history and uses of the various genera. Since then, the most systematic search new nodulating genera and species has taken place in Brazil ( de Faria et al., 1989; Moreira de Souza et al., 1992). The fundamental position that nodulation is least common in the Caesalpiniodeae; most common in the Papilionoideae, with the Mimosoideae being intermediate remains unchanged. These sub-families will be considered in turn.

The Caesalpinioideae

Nodulation is rare (23% of species examined) within the Caesalpinioideae, but consistent within certain groups (Sutherland and Sprent, 1993). Of the 8 confirmed nodulating genera within this sub-family, 7 are found in the tribe Caesalpinieae. This tribe contain 47 genera, 19 of which are unknown nodulation status and the other 21, including the type genus (with >100 spp) Caesalpinia, apparently non-nodulating (de Faria et al., 1989. Ecological data on nodulation are very rare. Barrios and Herrera (1993) reported abundant nodules on Campsiandra laurifolia in seasonally-flooded areas of the Orinoco basin. Nodules were perennial and had active nitrogenase, with the typical structure of other caesalpinioid nodules (de Faria et al., 1987), in which bacteria are retained in infection threads throughout the N₂-fixing period. The other nodulating caesalpinioid genus is Chamaecrista, in the tribe Cassieae. All the nodulated members of the genus Cassia listed in Allen and Allen (1981) have been re-assigned to Chamaecrista, using the usual taxonomic criteria such as flower characters.

The Mimosoideae

Nodulation in this sub-family is far more common than in the Caesalpinioideae (90% of species examined), but by no means universal. Several related groups in the tribe Mimoseae appear not to nodulate (de Faria et al.,1989). These include Adenanthera (not to be confused with Anadenanthera from the same tribe, which does nodulate) a genus of about 8 species in tropical Asia and the Pacific which includes large trees.

By far the largest genus (about 1200 species) in the Mimosoideae is Acacia. At present several sub-genera are recognized (Ross, 1981; Vassal, 1981). Whilst nodulation is very common, there are several well-authenticated non-nodulating species including the American A. greggii (Eskew and Ting, 1978) and African A. brevispica (Odee and Sprent, 1992).

Nodules examined in the Mimosoideae are all indeterminate, with distinct apical meristems: their central tissue contains a mixture of infected and uninfected cells. Infection by rhizobia may be via a classical root hair pathway [e.g. Prosopis (Barid et al., 1985)] by direct epidermal entry [e.g. Mimosa scabrella (de Faria et al., 1988)] or via sites of emergence of lateral roots [e.g. Neptunia plena (James et al., 1992)]. However, in the latter, infection threads are formed shortly after infection, unlike the situation in wound infections of papilionoid species.

The Papilionoideae

In this sub-family 97% of the species examined can nodulate. In the tribe Swartzieae, only the type genus has been found to nodulate. Of the 11 remaining genera, 6 have been examined and no nodules found (de Faria et al., 1989; Moreira de Souza et al., 1992). Barnet (1988) summarized the reports of nodulation among Australian native legumes: 52 genera are listed, 49 from the Papilionoideae, 3 from the Mimosoideae (Acacia, Albizia, Neptunia) and none from the Caesalpinioideae. Although Norris (1959) reported nodules on Intsia bijuga from Queensland, this and other species of the caesalpiniod genus Intsia have been reported as non-nodulators (Allen and Allen, 1981).

2.4 Agroforestry : Nitrogen Fixing trees in integrated agriculture

2.4.1 Agroforestry

“Agroforestry is a collective name for land-use systems in which woody perennials (trees, shrubs, etc.) are grown in association with herbaceous plants (crops, pastures) and/or livestock in a spatial arrangement, a rotation or both, and in which there are both ecological and economic interaction between the tree and non-tree components of the system” (Nair et al., 1984).

Nitrogen fixation is a pattern of nutrient cycling which has successfully been used in perennial agriculture for millennia. The legumes, which are nitrogen fixers of particular importance in agriculture. Specifically, tree legumes (nitrogen fixing trees, hereafter called NFTs) are especially valuable in subtropical and tropical agroforestry. They can be integrated into an agroforestery system to restore nutrient cycling and fertility self-reliance (Elevitch and Wilkinson, 1998).

2.4.2 Why N₂ fixing trees

Nitrogen fixing plants are key constituents in many natural ecosystems in the world. They are the major source of all nitrogen that enters the nitrogen cycle in these ecosystems. Many nitrogen-fixing plants are woody perennials, or nitrogen fixing trees (NFTs), most of these being found in the tropics. In temperate areas, the nitrogen fixers tend to be herbaceous (NFTA, 1989).

In the context of sustainable agricultural, the use of NFT may be considered as a biological strategy directed at sustainability (Budowski and Russo, 1997). In an agroecosystem this means profitable production without damage to the environment.

NFTs have been removed of reduced in most man-made ecosystems, such as agricultural and forest lands and urban environments. These lands require expensive chemical fertilizer inputs in order to maintain their productivity. Manmade systems can be improved be learning and adopting from natural ecosystems. For example, the reintroduction of NFTs, with appropriate management, can increase and sustain productivity. Agroforestry land-use practices do this (NFTA, 1989).

Why nitrogen-fixing trees? Brewbaker (1985) highlighted the reasons as follows: (1) many species with many uses; (2) nitrogen fixation; (3) rapid growth; (4) ecological significance; (5) multiple purposes; (6) potential for genetic improvement; and (7) easy propagation.

Woody perennials in agroforestry have two main roles, one productive and the other protective (Nair et al., 1984). The productive role includes production of food, fodder, firewood and various other products; the protective role stems from soil improving and soil conserving functions (Fig. 2.9). Systems in which food or cash crops are grown in combination with N₂-fixing leguminous trees is an ancient agricultural practice whose origins reach back to the domestication of such perennial crops. Most of the tree species used and associated with shade or perennial crops are legumes.

Figure 2.10 Roles of nitrogen-fixing trees in agroforestry (Nair et al., 1984)

2.4.3 How to Use NFTs in a System In the tropics, most of the available nutrients (over 75%) are not in the soil mbut in the organic matter. In subtropical and tropical forests, nutrients are constantly cycling through the ecosystem. Aside from enhancing overall fertility by accumulating nitrogen and other nutrients, NFTs establish readily, grow rapidly, and regrow easily from pruning. They are perfectly suited to jump-start organic matter production on a site, creating an abundant source of nutrient-rich mulch for other plants. Many fast-growing NFTs can be cut back regularly over several years for mulch production. The NFTs may be integrated into a system in many different ways including clump plantings, alley cropping, contour hedgerows, shelter belts, or single distribution plantings. (See figure below). As part of a productive system, they can serve many functions: microclimate for shade-loving crops like coffee or citrus (cut back seasonally to encourage fruiting); trellis for vine crops like vanilla, pepper, and yam; mulch banks for home gardens; and living fence and fodder sources around animal fields.

Figure 2.11 How N₂ fixing trees can be integrated in agriculture (Elevitch and Wilkinson, 1998)

2.4.4 N₂-fixing trees in agroforestry

N₂-fixing trees fall into main groups- the nodulated legumes and the actionrhizal trees. Within the Leguminosae, the majority of the trees used for agroforestry are nodulated members of the Papilionoideae and the Mimosoideae, but there are a few species in the Caesalpiniodeae (many of which do not form nodules) which havd been used experimentally in agroforestry (Giller and Wilson, 1991). Detailed information in many more species is given by Allen and Allen (1981).

A resource document with a master list of woody species under consideration as NFT was prepared for the 1981 Bellagio workshop on nitrogen-fixing tree germplasm (Halliday and Nakao, 1982). The authors considered 1054 species: 88% legumes and 12% non-leguminous trees. Later, Brewbaker et al. (1990) in a validation and prioritization of NFT compiled 648 species representing nine families. The families were: legumes (519 species); Betulaceae (38 species); Casuarinaceae (20 species); Coriariaceae (16 species); Myricaceae (14 species); Rhamnaceae (14 species); Elaeagnaceae (10 species);Ulmaceae (8 species); and Rosaceae (5 species).

Within the leguminoseae or Fabaceae, the Mimosoideae and the Papilionoideae both are thought to have evolved from the Caesalpinoideae, the latter being more primitive and less nodulated compared to the other two groups (Polhill et al., 1981).

2.4.4.1 Nodulated legume trees for Agroforestry

Acacia spp:

The genus Acacia is one of the largest genera in the legume family containing about 1200 species and distributed throughout the tropics and sub-tropics (Brewbaker, 1987; Binkley and Giardina, 1997). Acacia trees host nitrogen fixing Bradyrhizobia (Galiana et al., 1990) or Rhizobia (Turk and Keyser, 1992) and are infected by VA mycorrhizae. Galiana et al. (1990) reported that Acacia mangium and Acacia auriculiformis nodulated effectively with Bradyrhizobium strains but formed ineffective nodules with Rhizobium strains.

In Bangladesh, Acacia auriculiformis and A. mangium introduced in early 1980's and getting priority in plantation programmes. Acacia auriculiformis grows in a wide range of deep and shallow soils including sand dunes, clay, limestone and lateritic soils. An increment in height of 2-4 m per year in the first few years is common even on soils of low fertility (Boland, 1989). A mean annual increment of 15-20 m³ ha^-1 is obtainable on relatively fertile soils, but on less fertile or highly eroded sites the increment is reduced to 8-12 m³ ha^-1 (Wiersum and Ramlan, 1982). Recommended rotation for A. auriculiformis is 4-5 years for fuelwood, 8-10 years for pulp and 12-15 years for timber productions. A. auriculiformis produces profuse bundles of nodules and can survive on land very low in nitrogen and organic matter (Domingo, 1983). In Indonesia and Philippines A auriculiformis has been successfully planted on steep and unstable slopes for erosion control (NAS, 1980).

Acacia mangium is extensively used in Sabah for the conversion of degraded Imperata grasslands to productive forests (NAS, 1979). A. mangium annual increments in plantations typically range from 20-40 m³ ha^-1, and up to 50 m³ ha^-1 on the best sites (Binkley and Giardina, 1997).

Albizia spp:

Albizias are nodulated by Rhizobia and Bradyrhizobia bacteria (Turk and Keyser, 1992). About 100 species comprise the genus of Albizia and Paraserianthes, but a few of them (Albizia procera, A. lebbeck, A. chinensis, P. falcataria) have been widely used in tropical plantations (MacDicken, 1994).

Albizia lebbeck (L.) Benth. is a moderate to large deciduous tree that reaches 30 m in height in rain forests of India, Bangladesh, Myanmar and Andaman Islands and naturalized in many other tropical and sub-tropical areas. In plantings corresponding to 25,00, 10,000 and 40,000 trees ha^-1 in Puerto Rico, above ground biomass per unit area increased with density during the first 24 months, yielding 12.6, 14.5 and 17.4 ton ha^-1 respectively (Parrotta, 1987). A. lebbeck is not Rhizobium specific, and native strains are always capable of producing an abundance of nodules.

Albizia procera (Roxb.) Benth. is usually 60-70 cm in diameter and 25 meters in height. The species is a component of tropical and subtropical moist and wet forest types where rainfall is 1000-5000 mm yr^-1. The native range is South and Southeast Asia (Nielsen, 1979). The species is used for timber, fuel wood, charcoal, fodder or shade in plantation and agroforestry programmes. The species is considered a promising source of pulp for high quality paper (Parrotta, 1987), traditional medicines (Venkataramany, 1968) and also for vegetables (Hensleigh and Holaway, 1988).

The species grows best on moist alluvial soils, well-drained loams or clay soils (Brandis, 1906; Venkataramany, 1968). Good survival and rapid early growth reported in afforestation trials on both saline and alkaline soils (Ghosh, 1976). A. procera form symbiotic association with Rhizobium bacteria enabling it to fix nitrogen and thrive on infertile soils. Phosphorus applications improve nodulation and nitrogen fixation, particularly on infertile soils.

Albizia saman: The huge umbrella-shaped canopy of shady leaves of Albizia saman (Samanea saman), the rain tree is well known as an ornamental throughout the tropical and sub-tropical world (FAO, 1987). It produces pods with an edible pulp, and its wood is excellent for boats, furniture and crafts.

Calliandra spp:

This genus has about 100 species of shrubs and trees in tropical and warm temperate regions. Fast growth, excellent coppicing ability and ability to tolerate infertile soils make it an excellent fuel wood tree (FAO, 1987). Calliandra calothyrsus, a native of Central America was introduced into Java in 1936 and then some other Asian countries (Domingo, 1983). C. calothyrsus is a tall shrub (4-6 m) and recently introduced in Bangladesh (Hossain and Khan, 2003).

By nitrogen fixation and heavy litter production, this species improves soil productivity so rapidly that farmers in Indonesia (Java) often rotate or intercrop agricultural crops with Calliandra plantations (Domingo, 1983). It is also widely used by farmers for fuelwood and animal fodder. Calliandra nodulates readily with indigenous rhizobia in soils in Java and is nodulated by both fast- and slow growing strains (Giller and Wilson, 1991).

Dalbergia sissoo:

About 300 species of tropical and sub-tropical plants belong to the genus Dalbergia (FAO, 1987). D. sissoo is a good quality timber, cabinet and furniture making wood native to the Himalayan region from the lowlands to 1300 m. Other uses of the species are fuelwood, charcoal, shade, windbreaks, erosion control and fodder (young foliage and branches).

Gliricidia sepium:

The genus Gliricidia is indigenous to tropical America (Simons and Stewart, 1994), where the small trees are commonly found up to altitudes of 2000 m. G. sepium is the most widely known species and has been used as a shade tree for coffee and cacao (Giller and Wilson, 1991). The leaves of Gliricidia are toxic to horses but can provide high quality feed for cattle and goats. The dense wood is excellent firewood. Nitrogen rich leaves are useful in alley farming, shading and fertilizing coffee and cacao. G. sepium can form nodules with both fast- and slow growing rhizobia and was found to fix N₂ most effectively with fast growing strains originally isolated from Gliricidia nodules (Akkasaeng et al., 1986). The species was found successful in an international provenance trial in Bangladesh (Hossain et al., 1996).

Leucaena:

Leucaena leucocephala is the most widely used agroforestry species because of its fast growth and large biomass production, deep rooting and high nitrogen fixing ability (NAS, 1977; Pound and Martinez-Cairo, 1983; Anon, 1984; FAO, 1986). It is the most famous nitrogen fixing trees in the tropics, often hailed as a miracle tree (NAS, 1977; Parrotta, 1992; MacDicken, 1994; Shelton and Brewbaker, 1994). The annual stem wood increments range from about 10-60 m³ ha^-1 yr^-1 (Domingo, 1983).

Growth of Leucaena is improved by rhizobial inoculation on some soils of Australia (Diatloff, 1973), Colombia (Halliday and Somasegaran, 1983), Africa (Savory and Thomas, 1977; Sanginga et al., 1986), Malaysia (Chee et al., 1989) and Thailand (Homchan et al., 1989), though in several cases indigenous rhizobia are present which can nodulate Leucaena (Bushby, 1982; Homchan et al., 1989).

Table 2.6 : Nitrogen fixing tree species commonly used in agroforestry plantation programmes of Bangladesh.

Family/ sub-family	# Species	% N fixer	Species used in plantation programmes of Bangladesh
Leguminosae- Sub-family Caesalpinioideae	1900	23	Senna siamea Cassia fistula Delonix regia
Sub-family Mimosoideae	2800	90	Acacia auriculiformis Acacia mangium Acacia nilotica Albizia chinensis A. lebbeck A. lucida A. odoratissima A. procera A. richardiana A. saman Calliandra calothyrsus Enterolobium ….. Leucaena leucocephala Mimosa pudica Paraserianthes falcataria Xylia dolabriformis
Sub-family Papilionoideae	12300	97	Cajanus cajan Dalbergia sissoo Erythrina indica Flemingia ….. Gliricidia sepium Sesbania sesban Tephrosia candida
Casuarianaceae	90	100	Casuarina equisetifolia

Source: Hossain, M.K. (2003)

L. leucocephala nodulates profusely and nodule masses of 51 kg ha^-1 have been recorded in the top 70 cm of soil with the majority of the nodules at a depth of 10-30 cm (Hogberg and Kvarnstrom, 1982). The young indeterminate nodules are spherical but these elongate and branch to form coralloid nodules up to 3 cm in diameter (Halliday and Somasegaran, 1983). The Leucaena-Rhizobium partnership is capable of annually fixing more than 500 kg of nitrogen per hectare, which is equivalent to 2500 kg ammonium sulphate ha^-1 yr^-1 (Domingo, 1983). However, nitrogen fixation occurs only if the correct Rhizobium strain is present in the soil, since Leucaena is highly specific in its inoculation requirements (Hill, 1971). Leucaena is nodulated by fast-growing rhizobia isolated from several tropical hosts but generally not by most temperate Rhizobium species or by slow growing rhizobia (Trinick, 1968; Lewin et al., 1987). Rhizobia bacteria nodulate Leucaena, and VA mycorrhizae also infect Leucaena roots. About half of the nitrogen required by Leucaena comes from N fixation with a typical rate of 100-150 kg N ha^-1 yr^-1 (Binkley and Giardina, 1997), but the establishment and growth on acid soils in the tropics are poor (Halliday, 1981). Aluminium toxicity and phosphorus deficiency have been identified as the main factors inhibiting L. leucocephala growth in acid soils (Duguma et al., 1988; Shelton and Brewbaker, 1994), although improvement of acidity tolerance has been achieved by crossing with acid tolerant accessions of L. diversifolia (Hutton, 1990). N and P are partly responsible for the poor establishment of L. leucocephala in Alfisols of South-western Nigeria (Cobbina et al., 1992).

Paraserianthes falcataria:

Paraserianthes falcataria is one of the most fast growing species in Bangladesh and other South East Asian countries. On good sites, it may reach 7 m in height in a year, 13-18 m in 3 years and 21 m in 4 years (Domingo, 1983). Diameter may increase at about 5-7 cm yr^-1 (NAS, 1979). Its amazingly fast growth is due partly to nitrogen additions through its root nodules and its heavy leaf litter production (Domingo, 1983). The annual increment in plantations range from 20-50 m³ ha^-1 yr^-1 for average sites and up to 55 m³ ha^-1 yr^-1 for best sites (Domingo, 1983; Binkley et al., 1992; Seorianegara and Lemmens, 1993).

Sesbania species:

About 60 species of Sesbania are distributed throughout the tropics with representatives native to all continents (Evans and Rotar, 1987; Evans, 1990) and vary from annuals to short lived perennials. Sesbania grandiflora is more widely distributed in Southeast Asia, used mainly in agroforestry systems (Domingo, 1983). Extra ordinary nodulation capacity of the species helps restore fertility of the poor soils. Flowers and fruits of Sesbania grandiflora are eaten as a vegetable in Indonesia (FAO, 1987; Giller and Wilson, 1991). Major advantage is that many species of Sesbania are both tolerant of water logging and of saline and alkaline conditions (Giller and Wilson, 1991). The stem nodulating S. rostrata has excited interest due to its ability to grow and fix N₂ in water-logged conditions and its fast growth.

Rhizobia nodulating roots of Sesbania spp. are generally fast growing strains, and often exhibit rapid growth, producing large (>4 mm) colonies in less than 24 hours, though a few slow growing isolates have been recorded (Odee, 1990). The young root nodules are round, but older root nodules are multilobed (Harris et al., 1949). Stem nodules on S. rostrata are formed by Azorhizobium caulinodans.

2.4.4.2 Non-Nodulating Legume Trees for Agroforestry

The non-nodulating trees which have been used in agforestry mainly belong to the Caesalpiniodeae- and more specifically to what was formerly the genus Cassia. Taxonomic revision of the genus Cassia has led to the recognition of three separate genera; Cassia, Sanna and Chamaecrista (the latter contains only nodulating species) (Irwin and Barneby, 1981). The species Cassia siamea and C. spectabilis are now placed within the genus Senna (and will therefore be referred to as Senna siamea and S. spectabilis), in which no species have been found to nodulate. Despite the lack of nodulation, these two species have been tested in agroforestry and other species now placed within this genus (e.g. S. cobanensis) have been used as shade plants in tea and coffee plantations (Allen and Allen, 1981). The vigorous growth of Senna tora was once the source of speculation that they might be able to fix N₂ in symbiosis with bacteria without forming nodules, but all attempts to demonstrate bacteria in root tissues proved negative (Allen and Allen, 1933). There are few reports of nodulation within the species of the tribe Cassiea that have been used in agroforestry.

2.4.4.3 Actinorhizal Trees for Agroforestry

Of the actinorhizal trees, the alders (Alnus spp.) are used at high altitudes for timber in some countries (e.g. Costa Rica (Holdriidge, 1951)) but these trees are not adapted to the

Table 2.7 : Nitrogen fixing tree species commonly used in road side and forest plantation, agroforestry, community forestry and homestead plantation programmes in Bangladesh.

Species	Family /sub-family	Origin
Acacia auriculiformis Acacia catechu Acacia mangium Acacia nilotica Albizia chinensis Albizia lebbeck Albizia lucida Albizia odoratissima Albizia procera Albizia richardiana Albizia saman Calliandra calothyrsus Casuarina equisetifolia Dalbergia sissoo Erythrina variegata Gliricidia sepium Leucaena leucocephala Paraserianthes falcataria Sesbania grandiflora Sesbania sesban Xylia dolabriformis	Mimosoideae Mimosoideae Mimosoideae Mimosoideae Mimosoideae Mimosoideae Mimosoideae Mimosoideae Mimosoideae Mimosoideae Mimosoideae Mimosoideae Casuarianaceae Papilionoideae Papilionoideae Papilionoideae Mimosoideae Mimosoideae Papilionoideae Papilionoideae Mimosoideae	Australia Native Australia Native Native Native Native Native Native Native Naturalized Indonesia Native Native Native S. America Philippines S. America Native Native Myanmar

Source: Hossain, M.K. (2003)

climates of the lowland tropics. Species of Casuarina are the most important non-leguminous N₂-fixing trees in the lowland tropics.

Casuarian species:

Casuarina species belong to the family Casuarinaceae host nitrogen fixing Frankia actinomycetes (filamentous bacteria), vesicular-arbuscular mycorrhizae (VAM), and ectomycorrhizae (Reddell et al., 1986). Most of the 80 species in the genus originate from Australia and widely used species is Casuarina equisetifolia. C. equisetifolia is naturally found in the coastal area of Chittagong, Bangladesh (Hossain et al., 1995). The species is mainly used for fuelwood, land reclamation, sand dune stabilization and wind breaks. The species has been called the best fire wood in the world (FAO, 1987). On good sites, annual wood increments are 15 - 20 m³ ha^-1 yr^-1 (NAS, 1980).

All species of Casuarina have the ability to form root nodules and fix atmospheric nitrogen (NAS, 1980), though there are some negative reports (Khan, 1982). Rates of N-fixation have been reported of 60-95 kg N ha^-1 yr^-1 in pure stands to 40-60 kg N ha^-1 yr^-1 in stands mixed with Eucalyptus (Binkley and Giardina, 1997). Long term measurements of C. equisetifolia on tropical sandy soils showed that fixed N accumulated at about 58 kg N ha^-1 yr^-1. Another figure of about 229 kg N ha^-1 yr^-1 reported for trees growing on sand dunes (Aspiras, 1981). Marked differences in effectiveness of N₂ fixation have been found both between different strains of Frankia (Reddell and Bowen, 1985) and between different accessions of Casuarina (Sanginga et al., 1990).

2.4.5 Many Uses of Nitrogen Fixing Trees

Many nitrogen fixing trees are important to rural households throughout the country providing a variety of products and services. Nitrogen fixing trees can provide many of the products and services that people need such as firewood, forage, timber, green manure and erosion control (NFTA, 1989).

Fire Wood and Charcoal: Fire wood and charcoal are the primary energy sources for almost one half the world’s population. Fast growing, high density nitrogen fixing trees make excellent firewood and charcoal. Many species re-sprout or coppice vigorously after cutting and allowed repeated harvests without replanting.

Fodder: Fodder to feed animals is a constant concern to many farmers in developing countries, specially in Bangladesh. The highly nutritious and digestible leaves and pods of some nitrogen fixing trees (Leucaena, Albizia, Gliricidia) make them excellent feed for animals.

Soil Fertility: Soil fertility is critical to crop production, but many poor farmers in Bangladesh can’t afford chemical fertilizers. Leaves of many nitrogen-fixing trees are high in nitrogen and other plant nutrients and can be a renewable, free source of fertilizer. Atmospheric nitrogen fixation is also increased the nitrogen level in degraded forest lands.

Timber and Poles: Timber and poles are needed all over the world for furniture, house and other general construction purposes. Nitrogen fixing trees include both fast growing trees for rough wood to some of the most valuable luxury timbers, e.g. Acacia, Albizia, Dalbergia sissoo.

Human Food: Human food is harvested from several species of nitrogen fixing trees in various parts of the world, in some instances supplying important seasonal staples. Leaves, pods or flowers of Leucaena, Sesbania, Acacia, Parkia, Erythrina and Prosopis species are eaten by peoples.

Shade and Support: Coffee, cacao and tea often benefit from the shade and nitrogen rich litterfall provided by nitrogen fixing trees, such as Leucaena, Gliricidia, Erythrina etc. In Bangladesh, Albizia procera, A. chinensis are commonly planted as shade tree in the tea gardens.

Other Uses: Many other nitrogen-fixing trees are planted for erosion control, watershed protection, windbreaks, living fences, ornamentals and for production of tannins, gums and medicine.

2.5 Prospect of Biological Nitrogen Fixation (BNF) in Agroforestry

and Plantation Forests of Bangladesh

Increased use of legumes offers the potential for a significant decrease in the need for chemical nitrogen fertilizer, and therefore is a key component of sustainable agricultural systems. Site- and situation-specific research will be necessary to develop new and varied crop rotations, to optimise nitrogen fixation and other nutritional resources and also for pest control.

Bangladesh has an immense scope for expanding forest plantations in its barren degraded hilly lands and the plantation programme is increasing day by day. With deforeststion, many degraded sites are available for plantation, though very little is known about how to manage such sites economically. There has also been a net loss of forest cover over the years and the net yield per ha of original planted area has dropped significantly (Anon, 1993). The mean annual increment of present planting stock is extremely low by regional and international standards. Low fertility of the degraded area is also creating a major problem for the successful establishment of some important timber species (Hossain et al., 2001). The situation has been further aggrieved by existing silvicultural system of clear felling followed by artificial regeneration. It is well establish fact that intensive harvesting of forest resources decreased the nitrogen and organic matter content from the forest floor. It is essential to add chemical fertilizers for maintaining the soil fertility in plantation areas, but chemical fertilizers are very expensive and scarce. But, the other way is to plant nitrogen fixing tree species that may enrich the soil nitrogen status (Binkley and Giardina, 1997; Aryal et al., 2000; Chaukiyal et al., 2000). Most of the leguminous plantation species in Bangladesh form nodules in their roots with symbiotic association of Rhizobium and fix atmospheric nitrogen biologically, which is cheaper and renewable sources of nitrogen (Chaukiyal et al., 1999). In India Tewari (1998) has also emphasized that the afforestation might be accomplished by using nitrogen-fixing trees in barren areas.

Among the Acacia species introduced so far in Bangladesh, Acacia mangium and A. auriculiformis have shown promising growth in poor afforestation sites (Latif et al., 1985; Ara et al., 1989; Hossain et al., 1997). Some Acacia plantations showed better survival and growth in different areas of the country and the yield is 15-20 m³ ha^-1 yr^-1 at 10-12 years rotation. Some other successful nitrogen fixing species, such as Gliricidia, Calliandra are still in the experimental plantations, whereas Leucaena leucocephal is growing well in North and South of the country. Other promising nitrogen fixing species of Albizia (A. procera, A. saman and A. lebbeck) are common in both the plantation and homestead forests. But, the nodulation and nitrogen fixation potential of many tree legumes has not been examined so far in India (Purohit et al., 1997). Similarly, in Bangladesh information on comparative seedling growth and nodulation of legumes are very scanty (Aryal et al., 2000; Hossain et al., 2001), though multipurpose tree legumes are a major tree component in agroforestry and other participatory forestry practices.

The present need is to evaluate the nitrogen fixing tree species available or used in plantation programmes in Bangladesh and also to determined their nitrogen fixing ability with environmental conditions.

2.6 Fertilizers and biological nitrogen fixation as source of plant nutrients: Perspective for future agriculture and agroforestry

One of the triumphs of human endeavour is the increase in agricultural productivity that has taken place in the last 50 years. World human population is still increasing. Global population has grown from 2.5 billion in 1950 to 5.7 billion in 1995. During this time, yields of cereals have increased in harmony with the population boom, providing enough food for the majority of the people and which even exceeded the demand (Bockman et al., 1990). While malnutrition and hunger are still widespread, they are due to political factors (e.g., wars, poverty), and not because agriculture has already reached an absolute biological or resource ceiling for food production.

The world’s population is increasing by about 1 billion people every 12 years. In 1980 the world population was 4.4 billion. In 2000 it was about 6.2 billion. In 2020, the population is projected to be about 7.7 bullion. It is expected to reach about 8.3 billion by 2025 before attaining a stable level later in the next century (United Nations Population Division, 1996). This spells out the need for increasing rice yields further and expanding the areas for intensive rice production.

The increase in food production is due to the interaction of factors such as plant breeding, irrigation, use of multiple cropping, crop protection with agrochemicals, and increased availability of plant nutrient. The most common nutrient that affects the yield is N.

Biological nitrogen fixation (BNF) and mineral fertilizers provide the principle input to agriculture soils which is N. the atmospheric deposition of nitrogen oxides and ammonia also contributes to the supply of N. This deposition originates mainly from pollution.

On the global scale, BNF provides the largest input of N to soils. Various estimates of this contribution have been published, ranging from 44 to 200 Tg N/year (Soderlund and Rosswall, 1982). In general, about 140 Tg N/year is the estimate mostly used.

In 1991-92 about 11% of all arable land was used for producing legume oil seeds (soybeans and groundnuts) and pulses (Peoples et al., 1995). In addition grazed grass-clover swards are commonly rotated with cereals in regions where yields are uncertain and generally low due to water deficiency, such as in some parts of Australia. Other sources of BNF for agriculture are Azollae and Sesbania sp., which are used as green manure in rice production (Ladha and Garrity, 1994).

Past and present, this proves that BNF has an assured place in agriculture. However, the steady increase in agriculture productivity would not have been possible without mineral fertilizers. Fertilizer N use has increased from 9.6 Tg N in 1960 to 77 Tg N in 1995.

Nitrogen fertilization is sometimes needed to achieve a substantial yield of legumes (e.g., soyabean) when the symbiotic N₂ fixation is unable to provide enough nitrogen (Buttery and Dirks, 1987). However, fertilizer rates exceeding those exerting a “starter nitrogen” effect generally reduce nodulation and N₂ fixation (Afza et al., 1987). The response of the Rhizobium-legume symbiosis to added nitrogen fertilizer is definitely determined by time of application (growth stage), level and form of N, and the legume species (Abdel-Wahab et al., 1996; Imsande, 1996; Kucey, 1989).

Food is first. A growing population requires additional food, and over 90% of food production requires arable land (Brown et al., 1998). Until the mid-20^th century the amount of arable land kept approximate pace with population growth. In the last 30 years the amount of new arable land farmed each year has leveled off and not kept pace with population growth. After 10,000 years of growth the world is now running out of new arable land (Gardner, 1996). In contrast, grain production has generally kept pace with population increases.

The examination of population, arable land, and grain production on a per capita basis sets the stage for discussing what has to occur in the future relative to food production, if only to keep pace with population growth.

Food is first; energy is second. A growing population needs energy, again not only because there are more people, but because of increases in per capita use. Up until the late 19^th century, wood was a primary fuel. With the industrial revolution, fossil fuels became increasingly important (Mackenzie, 1998).

Increased resources are required for survival of the growing population. Further increases are needed if the standard of living is to improve. This is especially true in less developed countries where per capita resource use is significantly lower than in more developed regions. In the latter, there is more of an aim to stabilize per capita resource use. In less developed regions, if resource use increases linearly with population, per capita use of resources will remain constant, and there will be little change in the quality of life. In other words, more energy and food needs to be produced solely to supply more people. Only it resource use increases at a faster rate than population will there be substantial increases in the quality of life (Erisman et al., 1998).

2.7 The Need of Legume Improvement

The legume, with its symbiotic root nodule bacteria, is the most commonly used biological nitrogen fixation system in agriculture and forestry. However, it generally has a secondary role in cropping systems because its yield potential, food energy, and often its profitability is lower than that for cereals. Furthermore, in some cases, legume production is often not maximized because of limiting nutritional factors, disease, insect predation, and environmental stresses to which the plants may be more susceptible than those of other crops. If biological nitrogen fixation is to be effectively exploited, factors that constrain nodulation, nitrogen fixation, and the growth and yield of legumes must be identified and addressed.

High yield, disease and insect resistance, and tolerance of environmental stresses are required in legumes that are high in nitrogen-fixing ability. Improved cultivars should be developed using both conventional and molecular plant-breeding techniques. All legume breeding must be done on low- nitrogen soils without nitrogen fertilizer application to ensure preservation of optimal nodulation and nitrogen-fixing capacity. To exploit the advances in molecular biology, routine transformation systems for legumes will be needed.

The yield potential of locally important legumes should be measured in the field using quality seed, controlling disease and insects, applying sufficient nutrition (fertilizer), employing irrigation, controlling weeds, and using optimum plant spacing and the best management practices known. The yield potential with effective inoculation of superior strains should be compared with and without a high rate of nitrogen fertilization. It is important to develop legume cultivars that obtain most of their nitrogen from fixation without sacrificing actual and potential yield levels. Cultivars that are disease and insect resistant as well as tolerant of environmental stresses should be developed. This will entail the use of the best combination of conventional and molecular plant-breeding techniques with trials on low-nitrogen soils and including rhizobia inoculation.

2.8 Promising Research on BNF of NFT (Budowski and Russo, 1997)

v Uptake of N₂: how much from the air and from the soil?

v Improving the delivery of N₂

v Retrieving information from promising traditional field practices

v Planting techniques and their effect on BNF

v Nitrogen fixing nurse trees in tree plantations for timber