Soil Nitrogen And The Ecological Niche Of Legumes

Most Australian soils are low in nitrogen by world standards. Levels vary from around 0.05% for sands to around 0.7% for alpine humus soils with most soils, including those in south eastern Australian cropping areas, having organic nitrogen levels less than 0.15% in the top 10 cm (Spain, Isbell and Probert, 1983). A substantial increase in the supply of soil nitrogen can be economically achieved by growing pasture legumes. Legumes are pioneer plants, adapted for growth where soil nitrogen is low. In the natural order, after the legume-Rhizobium symbiosis has raised the quantities of soil nitrogen, the legume will be largely displaced by plants which do not fix nitrogen. Most of the soil nitrogen which accumulates under legume pasture is organic and unavailable to plants. It is cycled through the soil biomass which includes animals and micro-organisms. Generally, only 1-3% of organic nitrogen in the soil is mineralised each year and in pasture soils mineral nitrogen is usually found only in very small quantities because of uptake by grasses and nitrophilous weeds (Ellington, 1986). Losses of mineral nitrogen via leaching from pasture soils have been thought to be low (Ellington, 1986) but probably occur in sufficient magnitude to contribute to soil acidification (Helyar, 1976).

Nitrogen is also fixed by soil organisms not in symbiotic relationships with plants. The rates of fixation in grassland may be in the order of 50 kg N/ha/yr. However, fixation by free-living organisms is probably much less than that by legumes (Ellington, 1986).

The Benefit Of Legume Pastures To Soil Nitrogen And Wheat Crops

The level of soil nitrogen is generally directly related to the number of years of clover pasture and the clover content of the pasture (Kohn et al., 1977; Watson, 1963). This is demonstrated in Figures 1 and 2. The former illustrates nitrogen accretion under grazed subclover-barley grass pasture plots at Wagga Wagga. The latter shows the effects of numbers of years of clover pasture, grown on a very sandy soil at Kojonup, WA, on soil nitrogen, wheat dry matter and grain yields and grain protein. The second wheat crop in this experiment (Watson, 1963) also benefited from clover pasture with a two-fold increase in yield on 5 years of clover plots compared to no clover plots.

Lucerne, as a component of subclover pastures, not only provides a 10 to 20% increase in animal production (Mulholland, 1987), but will most likely result in a further increase in soil nitrogen because of its perennial growth and deep rooting habit. Data on the amount of nitrogen fixed by lucerne is scarce (Cregan, 1987). In the Mallee of Victoria, Wells (1970) found that wheat crops following lucerne were less responsive to added nitrogen than crops following volunteer or medic pastures. At Tamworth, in the northern wheatbelt of New South Wales, Holford (1980) showed substantial benefits from lucerne rotations in both grain yield (Table 1) and protein content (Table 2).

Figure 1. Changes in total soil nitrogen under pasture in a “high P” soil (Kohn et al., 1977).

Figure 2. Yields and grain nitrogen content of wheat sown after clover pasture, first crop: soil: Kojonup sand (Watson, 1963)

For each year of lucern prior to cropping, the average increase in total soil nitrogen (0-15 cm) was equivalent to about 140 kg N/ha in the black earth and about 110 kg N/ha in the red-brown earth (Holford, 1981).

Table 1. The effects of lucerne on wheat yields (Holford, 1980)

Location years, soil type	Rotation	Wheat yields (t/ha) in successive years after lucerne or wheat
		1	2	3	4	5	6	7	8
Tamworth,	Lucerne	1.51	3.22	2.70	1.59	2.18	3.07	1.20	2.13
1970 - 78 black	Wheat	2.04	1.90	2.79	0.58	1.14	1.21	0.61	1.92
Tamworth,	Lucerne	1.99	1.49	2.18	0.78	1.03	2.38	1.99
1973 - 78 red	Wheat	1.68	0.53	1.08	0.62	0.56	1.92	1.32

Table 2. The effects of lucerne on wheat protein (Holford, 1980)

Location, years, soil type	Rotation	Wheat Protein % in successive years after lucerne or wheat
		1	2	3	4	5	6	7	8
Tamworth,	Lucerne	14.8	15.8	14.3	16.7	11.6	10.9	10.2	10.6
1970 - 78 black	Wheat	10.1	10.8	12.6	14.8	9.4	8.6	9.5	10.4
Tamworth,	Lucerne	16.2	17.0	10.7	10.3	11.4	10.9	7.7
1973 - 79 red	Wheat	13.8	13.2	8.6	9.4	11.0	10.6	7.1

Factors Affecting Nitrogen Fixation By Pasture Legumes

The single most important factor that affects the rate of build-up or accretion of soil nitrogen under pastures is total pasture nitrogen production which is very closely associated with total dry matter production and clover dry matter production (Watson, 1963; Kleinig et al., 1974). The relationship between pasture nitrogen yield and soil nitrogen levels under annual (subclover and annual ryegrass) and perennial (white clover and perennial ryegrass) pastures is shown in Figure 3.

Figure 3. Relationship between soil and mat nitrogen accumulation and total herbage nitrogen produced over five years (Kleinig et al., 1974).

Therefore, to manage the ley phase for soil nitrogen accretion is to manage pastures for maximum pasture legume production. The factors affecting legume dry matter production are briefly reviewed below.

(i) Legume species

The amount of nitrogen fixed by the important pasture legume species is probably lucerne > subclover > medics (Ellington, 1986) with perennial clovers about equal to lucerne. Estimates for nitrogen accretion under a range of leguiles are shown in Table 3.

Table 3. Rates of nitrogen accretion under forage legumes and pastures

LEGUME	NITROGEN ACCRETION RATE (kg N/ha/yr
	Mean	Range
Lucerne and sweet clover	200	56-4631
Lucerne	206	148-2902
Perennial clovers	183	45-6731
White clover	191	128-2682
Annual clovers	190	45-3661
Subclover	155	21-2872
Subclover (+ grass)	85	47-1823
Annual medic (+ grass)	69	9-1173
Annual medic	0	-4

Sources: Data from many sources cited in the following reviews:

1. Ellington (1986)

2. LaRue and Patterson (1981)

3. Clarke and Russell (1977)

4. Scott (1985)

Data from references 1 and 2 refer to amounts fixed by legumes; those in 3 were changes under pasture which sometimes included substantial non-symbiotic fixation; those in 4 were from a comparison of medic and volunteer species

The greater amounts of nitrogen likely to be fixed by perennial compared to annual species, probably because of their longer growing season, is evident in Table 2. However, growing conditions, which dictate the choice of species anyway, would play a large part in the amount of legume nitrogen produced (Brockwell, 1988). This is illustrated by the comparison of data on nitrogen fixation by annual medics, from four increasingly arid locations (Table 4).

Table 4. Nitrogen accretion from annual medics in increasingly arid envi ronments

Location	Net change in nitrogen due to medic (kg/ha/yr)
Longerenong, Victoria	36	*
	29	*
Walpeup, Victoria	9	*
Merredin, WA	0	*
Condobolin, NSW	0	#

* cited by Clarke and Russel (1977)

# Scott (1985)

Table 5. Relationship between average annual rainfall and subclover variety mixtures (Dear, pers.comm.)

Annual rainfall	(mm)	Subclover varieties
380 - 425		Nungarin - Dalkeith
425 - 475		Dalkeith - Daliak
475 - 525		Seaton Park - Junee
525 - 650		Junee - Trikkala - Woogenellup
650 - 800		Trikkala - Karridale
800+		Larisa

(ii) Legume species and varieties for south eastern Australia There are three main groups of pasture legumes that may be sown in southern

New South Wales. Subterranean clover is the most important in areas which receive between 425 and 800mm annual rainfall. Full descriptions of varieties and characteristics are available in Dear (1986; 1987). Annual medics are important in the drier (300 - 500 mm average annual rainfall) areas of New South Wales, particularly on neutral to alkaline soils. Descriptions of annual medic varieties may be found in Dear (1987) and McDonald (1986). The third group is the array of more than 30 lucerne varieties currently available (McDonald et al., 1988).

For maximum production and long-term persistence of an annual legume (subclover and/or medic) pasture it is important that a mixture of varieties best adapted to the area be sown. Generally, for subclovers, as average annual rainfall becomes less and the season shorter or more unreliable, the varieties selected should flower earlier and have a higher proportion of hard seed. Dear (pers.comm.) suggested a range of possible mixtures for various rainfall zones (Table 5).

On the central western slopes and plains and upper Riverina where rainfall is non-seasonal and unreliable, annual medics (Paraggio and Sephi) may be sown as a mixture or with early maturing subclovers in the more eastern (Parkes-Condobolin) or southern (West Wyalong) areas. In marginal cropping areas (Euabalong-Cobar) a mixture of Sephi and the very early maturing variety Cyprus is recommended. In the southwest of the State onthe cropped Mallee soils, mixtures containing Harbinger strand medic, which is well adapted to sandy, alkaline soils, and one or more of the barrel medics Sephi, Paraggio or Parabinga are recommended (Young et al., 1987).

Lucerne, as a component of subterranean clover pastures, can be estabished on a large proportion of the southern cereal growing areas. In general the production of sheep and cattle grazing lucerne/subclover pastures will be 10 to 20% greater than that of subclover pastures (Mulholland, 1987). This greater production from lucerne pastures may be reflected in increased rates of nitrogen accretion.

(iii) Pasture establishment

To obtain maximum levels of legume dry matter production during the pasture phase, good initial establishment techniques are required. These are elaborated in Dear (1986), McDonald (1986) and McDonald and Waterhouse (1988), and are summarised below.

The basic aims of pasture establishment are to maximise the seed production of annual species and to maximise seedling establishment and early growth of lucerne.

(a) Pasture sowing rates should be adequate: 7 kg/ha for subclover alone and 5 kg/ha if mixed with lucerne; 1-2 kg/ha for lucerne; 4-6 kg/ha for annual medics.

(b) Pasture seed should be inoculated and lime pelleted to ensure good root nodulation with nitrogen fixing Rhizobium bacteria.

(c) Reduce cover crop competition by choosing a short-strawed, erect type of cover crop: two-row barleys are best, grazing oats the most competitive.

(d) Reduce cereal cover crop sowing rates to approximately 40% of those recommended for wetter areas and 75% of those recommended for drier areas.

(e) Sowing machinery should be set up so that pasture seed is sown shallow at a depth of 2-4 cm. Furrow sowing techniques may be employed in quite sandy soils or self-mulching clay soils.

(f) Pastures should be sown early in the season (April-early May) and, because of their shallow sowing, into good moisture.

(g) Red-legged earthmite is a serious problem in many areas and can decimate young pastures. It can be controlled with a range of insecticides.

(h) Where the yield of the cover crop is high and straw quantities large, use a straw spreader to scatter straw. Otherwise built-up straw will smother lucerne plants and prevent emergence of annual legume seedlings.

(iv) The importance of inoculating seed with Rhizobium

It is difficult to predict the need to inoculate seed to achieve effective nodulation (Roughley and Walker, 1973). However, unless there has been a recent history (less than 4 years) of a vigorous stand of the species to be sown, pasture legume seed should be inoculated with the appropriate peat culture. Inoculated seed should be lime pelleted, particularly if soils are acid or if the seed will come in contact with fertiliser. Lime pelleting has been found to be superior to slurry inoculation at soil pH (1:5 soil:

water) values below 5.5 (Roughley and Walker, 1973). The effectiveness of naturalised strains is highly variable; it has been estimated that the average levels of effectiveness of such strains is only about 50% of that of improved strains (Bergerson, 1970). Inoculation of seed beds with either solid or liquid inoculants was considered a practical alternative to seed inoculation by Brockwell et al. (1980). Though these alternative methods were largely developed for grain legumes, it was found that establishment of clover seedlings was generally better with seed bed inoculation compared to seed inoculation. The inoculant could be applied at much higher rates if required and the method was particularly advantageous when a fungicidal seed dressing toxic to the Rhizobium was applied. The seed bed inoculant was applied as a suspension of peat in water through spray nozzles.

(v) Grazing management

For persistence, lucerne requires a period of recovery after grazing to replenish reserves of energy in plant roots. In a largely lucerne-based system, this may be accomplished by rotational grazing. With smaller areas of lucerne, set stocking should be avoided and plants should be allowed to flower several times each year (see McDonald and Waterhouse, 1988, for detail).

Annual pastures are best set stocked. In young pastures with low seed reserves, grazing pressure should be reduced when flowers first appear. Annual medics are less tolerant of heavy grazing than subclovers and may suffer a greater loss in seed production if grazed heavily at around and after flowering.

(vi) Nutrition, fertiliser history

Phosphorus and sulphur

Apart from nitrogen, phosphorus deficiency is the most common and widespread nutrient deficiency in south eastern Australian cropping soils. Correction of this with single superphosphate also supplies sufficient sulphur and calcium to maintain adequate levels of these nutrients. The frequent use of compound nitrogen and phosphorus fertilisers low in sulphur may lead to sulphur deficiency in pastures (Kohn et al., 1977; Scott, 1985).

In soils low in phosphorus (<10 ppm), marked responses to fertiliser phosphorus by pasture legumes have been recorded. Barrel medic dry matter yields more than doubled at both Condobolin and Mt Hope when 126 kg/ha of superphosphate were applied (Scott, 1980). Furthermore, in these semi-arid environments it was found that for superphosphate to be effective, it must be drilled to a depth of 5-10 cm and not topdressed.

Research at Wagga Wagga has indicated that there is little or no response in pasture yields and livestock production to topdressing with superphosphate once the total amount applied exceeds 700-1000 kg/ha and soil phosphorus levels are greater than around 22 ppm (Kohn, 1974). However, on soils of medium phosphorus status (about 15 ppm) Ayres et al. (1977a) found that the increase in pasture availability conferred by topdressing provided a critical improvement to the feed supply in dry periods. Ayres et al. (1977b) found that nitrogen accretion was greater, and more responsive to superphosphate at medium compared to high phosphorus sites. This is shown below in Table 6.

Table 6. The relationship between nitrogen accretion and phosphorus application history (Ayres et al., 1977b)

Total superphosphate 1969-1972 (t/ha)	Total N, 0-10 cm
	Medium P sites		High P sites
	Site 1	site 2	site 3	site 4
0	0.148	0.145	0.120	0.123
0.5	0.170	0.157	0.124	0.130
1.0	0.174	0.145	0.120	0.121

This lack of response (in pasture, animal and soil N factors) to top-dressing with superphosphate on soils high in phosphorus recorded by these researchers, was associated with a rapid invasion and dominance of clover pastures by barley grass as a result of the build up of soil nitrogen.

Kohn (1974) suggested that phosphorus for pastures and crops was most efficiently applied in the cropping phase and not by topdressing. In a review of research on phosphorus, Batten and Osborne (1983) concluded that soil nitrogen and soil structure after good clover pasture will support up to two cereal crops without the need for fertiliser nitrogen. To do this, soil phosphorus must be maintained in the long term with above optimal applications of phosphorus with crops to provide residual phosphorus to offset the decline in soil phosphorus which will occur during a non-fertilised pasture phase. This is summarised in Figure 4 where changes in soil phosphorus at Wagga Wagga are predicted when crops are grown with 6, 12 or 24 kg P/crop/ha, pastures are not topdressed, and phosphorus under pasturedeclines by 1-2 kg P/ha for each pasture year. Twenty-four kg P/ha for each crop is required to maintain soil phosphorus levels at around 20 ppm. The largest losses of phosphorus from the system occur in the grain which contains from about 0.2 to 0.5% phosphorus. This rate of loss increases with yield and is more rapid when nitrogen is applied to crops (Osborne et al., 1977).

Figure 4. A summary diagram based on Wagga data: predicted changes in soil P when crops were grown with 6, 12 or 24 kg P/ha/crop and pastures not topdressed (Batten and Osborne, 1963).

The soil phosphorus measurements quoted here are from several different soil tests. Phosphorus requirement is also affected by soil pH. The phosphate fertilisation programme for a particular piece of country should be formulated with the use and proper interpretation of soil tests and strip trials.

The micronutrients or trace elements molybdenum and boron Molybdenum is one of the most recently recognised micronutrient elements considered to be essential for the growth of plants. It has a key role in nitrogen fixation by legumes and in the use of nitrates. Because of the extremely low requirement of molybdenum by plants, it may be applied as seed dressings and in the lime of lime-pelleted pasture legume seeds or mixed in the factory with superphosphate. Deficiencies occur mainly on acid soils in coastal areas and on much of the southern tablelands. In lucerne, clover and other pasture legumes the deficiency symptoms are associated with an inability to fix atmospheric nitrogen. The resultant stunting and yellowing is identical with nitrogen deficiency suffered by unnodulated legumes growing in poor soils (Weir, 1984). Though molybdenum is not readily toxic to plants, high levels of molybdenum for forages, in relation to the amount of copper present, can induce copper deficiencies in animals (Gupta and Lipsett, 1984).

Boron is closely associated with cell division, cell structure and the pollination process. It is important for good seed set in subclover (Dear and Lipsett, 1987). Boron deficiency is most likely on the tablelands and slopes, particularly where the soils are derived from granite or sandstone parent material. Boron is required only in small amounts and the range between deficiency and toxicity is very narrow (Dear and Weir, 1984).

(vii) Botanical Composition

The dry matter production of legumes in pastures is greatly affected by the presence of grasses and broadleaf weeds. This is because temperate legumes are relatively poor competitors for light, water and nutrients. After the legume-Rhizobium symbiosis, in the presence of sufficient soil phosphorus, has raised the level of soil nitrogen, there are several possible “sinks” for this increased nitrogen:

(i) invasion of the pasture by barley grass and broadleaf weeds;

(ii) a sown grass (annual ryegrass, perennial ryegrass, phalaris);

(iii) a non-legume crop.

Botanical composition in an existing pasture may be manipulated by several means. Increased stocking (7-10 ewes/ha/yr) has been shown to favour clover content (Ayres et al., 1977a). MCPA amine may be used in combination with grazing (“spray-grazing”). Paraquat can be used in winter (“winter cleaning”) for grass control to provide problem-free grazing through spring and summer with a penalty of reduced pasture production after spraying. Glyphosate and paraquat at low rates may be used at weed flowering time (“spray-topping”) to sterilise the seed. These methods are dealt with in detail by Sutherland (1987) and Dowling (1987).

For sown grasses, little work has been done to determine the ideal grass clover ratio. For white and red clover pastures it was found to be 70:30 (Walker et al., 1954). Dear (1986) suggested that the main advantages of a small 4~ass component (about 20%) are:

(i) Grass, as a “sink” for available nitrate nitrogen would reduce losses due to leaching and slow acidification (Helyar, 1976). In small proportions, grasses can increase the rate of nitrogen fixation by clovers.

(ii) Grass stubble, being more presistent than legume residues over summer, can provide additional summer feed and stabilise the soil surface.

(iii) Annual ryegrass-subclover pastures have been shown to be more productive in early autumn and winter than pure clover pastures in some environments.

Grasses are difficult to exclude, and if the proportion of clover is markedly decreased, nitrogen fixation will be reduced. Furthermore, they can be a weed in subsequent cereal crops and may act as hosts for crop diseases.

Lucerne, once established, is a strong competitor with many problem weeds, e.g. skeleton weed (Wells, 1969) and variegated thistle (Noble, 1970). Lucerne is particularly effective in controlling summer growing weeds that are often difficult to control in annual pastures, especially in years with significant summer rain (Cregan, 1987).

(viii) Soil Acidity

Soil acidification poses a major threat to the survival and productivity of legume pastures and the efficient use of fertiliser phosphorus (Batten and Osborne, 1983). Though the fixation and cycling of nitrogen is a neutral (non-acidifying) process, it plays a major part in soil acidification when nitrogen is lost from the system or accumulates in it in a different form from that which is added (Helyar, 1976). For some soils, the dominant factors leading to acidification were considered to be:

(i) nitrogen cycle effects (58-84% contribution);

(ii) accumulation of organic acids and anions (22-26%); and

(iii) the export of organic anions in products (12-15%) (Helyar, 1986).

Soil acidity is manifest in many ways:

1. Aluminium and manganese toxicity. The latter can be aggravated by waterlogging which is more likely under conditions of poor soil structure.

2. Several nutrients, including molybdenum, calcium, magnesium and phosphorus, are less available in very acid soils. Phosphorus applied as fertiliser is less effective. Trials have shown that considerably more fertiliser phosphorus is required to achieve any particular dry matter yield of clover in unlimed compared to limed acid soils. However, the phosphorus status of a soil may be immaterial if other factors are limiting productivity (Batten and Osborne, 1983).

3. Reduced survival of Rhizobium, R. melioti, which nodulate medics does not survive in soils more acid than pH 5.3. R. triftolium, which nodulates subclovers, has lower limits for reasonable survival at around pH 4.5. Recent trials conducted with subclover in southern New South Wales indicate that soil acidity can result in a reduction in nitrogen fixed of 20-60 kg/ha/yr (Evans quoted by Gammie, 1986).

The effects of soil acidity may be corrected or mitigated by a combination of:

(a) liming of acid soils;

(b) prudent use of molybdenum;

(c) inoculating and lime pelleting legume seed;

(d) the inclusion of grasses and perennial species in pasture mixes;

(e) under development are strains of Rhizobiurn more tolerant of acid conditions than those currently available (Roughley, pers.comm.).

Rotations And Length Of The Pasture Phase

Soil nitrogen levels increase with years of clover pasture (see also Figures 1 and 2) and run down with subsequent cereal crops. This combination of events is demonstrated by data from Rutherglen (Ellington et al., 1979) presented in Figure 5. After continuous wheat cropping, at least two years of pasture were required to substantially lift soil nitrogen above the previous low levels for following wheat crops.

Figure 5. The effect of length of subclover pasture (P) and following wheat crops (W) on soil nitrogen (Ellington et al, 1979).

The reduction in wheat yield with years of cropping is attributed to the decline in soil structure and nitrogen levels and disease. The Rutherglen work also showed trends in soil structure that were very similar to those for nitrogen shown in Figure 5. Work at Wagga Wagga (Osborne et al., 1977) showed that nitrogen fertilisers only reduced the rate at which wheat yields declined during the cropping phase, the remainder of the yield reduction being attributed to deterioration in soil structure (Batten and Osborne, 1983).

Results from an experiment carried out at Wagga Wagga where rotations were cycled three times between 1963 and 1981 (Kohn et al., 1986) indicated that soil nitrogen could be maintained with a cropping intensity of between 50 and 67% (Table 7).

Table 7. Total soil nitrogen change from cycle 1 to cycle 3 of six rotations at Wagga, 1963 to 1981 (Kohn et al., 1986).

	Rotation	Cropping intensity	Total soil nitrogen1 change (ppm)
1	WWPWWP2)		-13
2	WWWWPP )	67%	-24
3	WPWPWP )		48
4	WWWPPP )	50%	76
5	WPPWPP )		14
6	WWPPPP )	33%	9

1 Cycle 1 total soil nitrogen readings were in the range 961-988 ppm

2 W = wheat, P = pasture

Taylor (1986) considered that, though Australia-wide research carried out in the 1960s and 70s indicated that a pasture intensity of around 33% (cropping intensity of around 67%) will maintain soil nitrogen and crop yields, where modern wheat varieties are grown a nitrogen balance will only be achieved with a 50% legume intensity. The actual length of the pasture or cropping phase for a given cropping intensity does not appear to greatly affect changes in soil nitrogen when rotations are compared within pairs in Table 7.

Under commercial conditions, the number of years of crop that can be grown without bagged nitrogen after a pasture ley will be variable and may range from 1 to 3 (5. Sutherland, H. Turner, pers.comm.). Other factors such as cereal crop disease control and managing the clover content of pastures are also of considerable importance when considering the structure and length of rotations. Two broad options exist with a non-legume cropping intensity of around 50%:

1. Short rotations of 2 years each of crop and pasture. Superphosphate is supplied with the crop and pastures are sown to clover alone and are clover dominant.

2. Extended rotations with a four year pasture phase. Pastures may be mixtures of annual ryegrass and lucerne with subclover. They may require topdressing with superphosphate and weed control with herbicides. The cropping phase may begin with rape and contain a year of grain legume between cereal crops to break the cycle of cereal crop disease.

A range of similar rotations studied by Riley (1986) was found to have similar gross margins which were superior to those of continuous cropping or continuous pasture once the costs and returns from legume nitrogen had been accounted for.

References

1. Ayres, J.F., McFarlane, J.D., Gilmour, A.R. and McManus, W.R. (1977a). Superphosphate requirements of clover-ley farming. I. Productivity in the ley phase. Aust.J.Agric.Res. 28:269-85.

2. Ayres, J.F., McFarlane, J.D., Osborne, G.J. and Corbin, E.J. (1977b). Superphosphate requirements of clover-ley farming. II. The residual effects of topdressing. Aust.J.Agric.Res. 28:287-99.

3. Batten, G.D. and Osborne, G.J. (1983). Pastures in rotation with crops in relation to phosphorus in non-leaching situations. In ‘Phosphorus for Pastures’ ed. J.W. Gartrell, pp.22-46. Australian Wool Corporation, Ade1aide.

4. Bergerson, F.J. (1970). Some Australian studies relating to the long term effects of the inoculation of legume seed. Plant and Soil 32: 727-36.

5. Brockwell, J. (1985). The role of Rhizobirim meliloti in annual medic pastures in central western New South Wales: pragmatic appraisal. In ‘The ecology and agronomy of annual medics, ed. Z. Hochman. Technical Bulletin 30, pp.39-42, Department of Agriculture New South Wales.

6. Brockwell, J., Gault, R.R., Chase, D.L., Hely, F.W., Zorin, Margaret and Corbin, E.J. (1980). An appraisal of practical alternatives to legume seed inoculation: Field experiments on seedbed inoculation with solid and liquid inoculants. Aust.JAgric.Res. 31:47-60.

7. Clarke, A.L. and Russel, J.S. (1977). Crop sequential practices. In ‘Soil factors in crop production in a semi-arid environment’, eds. J.S. Russel and E.L. Greacen, pp.279-300. University of Queensland Press, St.Lucia.

8. Cregan, P.D. (1987). Lucerne agronomy reviewed. In ‘Realising the Potential of Pastures’, Proceedings of the 16th Riverina Outlook Conference, pp.47-56. Riverina-Murray Institute of Higher Education, Wagga Wagga.

9. Dear, B.S. (1986). Meeting cereal nitrogen needs with pasture legumes. In ‘Nitrogen Management on the Farms of Southern NSW’, pp.30-37. Department of Agriculture New South Wales, Wagga Wagga.

10. Dear, B.S. (1987). New pasture species evaluated. In ‘Realising the Potential of Pastures’. Proceedings of the 16th Riverina Outlook Conference, pp.73-83. Riverina-Murray Institute of Higher Education, Wagga Wagga.

11. Dear, B.S. and Lipsett, J. (1987). The effect of boron supply on the growth and seed production of subterranean clover (Trifolium subterraneum). Aust.J.Agric.Res. 38:537-46.

12. Dear, B.S. and Weir, R.G. (1987). Boron deficiency in pastures and field crops. Agfact P1.AC.1. Department of Agriculture, New South Wales.

13. Dowling, P. (1987). Herbicide manipulation for pasture rejuvenation -permanent pastures. In ‘Realising the Potential of Pastures’. Proceedings of the 16th Riverina Outlook Conference, pp.25-31. Riverina-Murray Institute of Higher Education, Wagga Wagga.

14. Ellington, A. (1986). Nitrogen inputs and utilisation in leguminous pasture:

15. a review of recent Australian literature. In ‘Nitrogen Cycling in Temperate Agricultural Systems’, eds. P.E. Bacon, 3. Evans, R.R. Storrier and A.C. Taylor, pp.58-78. Australian Society of Soil Science Inc. Riverina Branch.

16. Ellington, A., Reeves, T.G., Boundry, K.A. and Brooke, H.D.(1979). Increasing yields and soil fertility with pasture/wheat/grain-legume rotations and direct drilling. 49th ANZAAS Congress, pp.128-40, New Zealand.

17. Evans, 3. (1986). Quoted in Gammie, R. Factors limiting production other than nitrogen. In ‘Nitrogen Management on the Farms of Southern NSW. pp.67-71 Department of Agriculture New South Wales, Wagga Wagga.

18. Gupta, U.C. and Lipsett, 3. (1981). Molybdenum in soils, plants and animals. In ‘Advances in Agronomy’ 34:73-115, ed. N.C. Brady, Academic Press.

19. Helyar, K.R. (1976). Nitrogen cycling and soil acidification. J.Aust.Inst. A2ric.Sci. 42:217-221.

20. Helyar, K.R. (1985). Nutrition of plants on acid soils. In ‘Temperate Pastures - Their Production, Use and Management’. eds. J.L. Wheeler, C.J. Pearson and G.L. Robards, pp.159-71. Australian Wool Corporation/CSIRO Australia.

21. Holford, I.C.R. (1980). Effects of duration of grazed lucerne on long-term yields and nitrogen uptake of subsequent wheat. Aust.J.Agric.Res. 31:239-50.

22. Holford, I.C.R. (1981). Changes in nitrogen and organic carbon of wheat growing soils after various periods of grazed lucerne, extended fallowing and continuous wheat. Aust.J.Soil Res. 19:239-49.

23. Kleinig, C.R., Noble, J.C. and Rixon, A.J. (1974). Herbage production and accumulation of soil nitrogen under irrigated pastures on the Riverine plain. Aust.J.Exp.Agric.Aflim.Hu5b. 14:49-56.

24. Kohn, G.D. (1974). Superphosphate utilisation in clover-ley farming. I. Effects on pasture and sheep production. Aust.J.Agyjc.Res. 25:525.

25. Kohn, G.D., Osborne, G.J., Batten, G.D., Smith, A.N. and Lill, W.J. (1977). The effect of topdressed superphosphate on changes in nitrogen: carbon: sulphur: phosphorus and pH on red earth soil during a long term grazing experiment. Aust.J.Soil Res. 15:147-58.

26. Kohn, G.D., Raison, J., Furness, K. and Taylor, A.C. (1986). The response of total nitrogen, organic carbon and pH to intensity of cropping and length of rotation in clover ley farming systems. In ‘Nitrogen Cycling in Temperate Agricultural Systems’, eds. P.E. Bacon, J. Evans, R.R. Storrier and A.C. Taylor, pp.76-79. Australian Society of Soil Science Inc. Riverina Branch.

27. La Rue, T.A. and Patterson, T.G.(1981). How much nitrogen do legumes fix? In ‘Advances in Agronomy’ 34:15-38. Ed. N.C. Brady, Academic Press.

28. McDonald, K.J. (1986). Lucerne Varieties: Agfact P2.5.13. Department of Agriculture New South Wales.

29. McDonald, W.J. and Waterhouse, D.B. (1988). Lucerne for pasture and fodder. Information from Department of Agriculture New South Wales.

30. McDonald, W.J., Waterhouse, D.B., Falconer, G.W. and Williams, R. (1988). Lucerne varieties: Agfact P2.5.13. Department of Agriculture New South Wales.

31. Mulholland, 3. (1987). Animal production from lucerne-based pastures. In ‘Realising the Potential of Pastures’. Proceedings of the 16th Riverina Outlook Conference, pp.57-66. Riverina-Murray Institute of Higher Education, Wagga Wagga.

32. Noble, J.C. (1970). Competition between Heliotropium and dryland lucerne. Proceedings Australian Weeds Conference, Hobart.

33. Osborne, 6.3., Batten, G.D. and Kohn, G.D. (1977). Nitrogen and phosphorus requirements of wheat following subterranean clover pasture. Aust.J.Agric.Res. 28:971-9.

34. Riley, T. (1986). Profitable cropping rotations. In ‘Nitrogen Management of the Farms of Southern NSW’, pp.61-66. Department of Agriculture New South Wales, Wagga Wagga.

35. Roughley, R.J. and Walker, M.H. (1973). A study of inoculation and sowing methods for Trifolium subterraneurn in New South Wales. Aust.J.Exp.Agric. Anim.Husb. 13:284-291.

36. Scott, B.J. (1980). Phosphate placement studies in semi-arid Australia. M.Sc.Agr.Thesis, University of Sydney.

37. Scott, B.J. (1985). Nutrition of barrel medic in central western New South Wales. In ‘The ecology and agronomy of annual iliedics’ ed. Z. Hochman. Technical Bulletin 32, pp.35-37. Department of Agriculture New South Wales.

38. Spain, A.V., Isbell, R.F. and Probert, M.E. (1983). In ‘Soils - An Australian Viewpoint’, Division of Soils, CSIRO, pp.551-563. CSIRO,Melbourne/Academic Press, London.

39. Sutherland, 5. (1987). Herbicide manipulation of ley pastures. In ‘Realising the Potential of Pastures’. Proceedings of the 16th Riverina Outlook Conference, pp.57-66. Riverina-Murray Institute of Higher Education, Wagga Wagga.

40. Taylor, A.C. (1986). Comparison of Farming Systems. In ‘Nitrogen management on the farms of southern NSW’, pp.57-60. Department of Agriculture New South Wales, Wagga Wagga

41. Walker, T.W., Orchiston, H.D. and Adams, A.F.R. (1954). The nitrogen economy of grass/legume associations. J.Brit.Grassl .Soc. 9:249-74.

42. Watson, E.R. (1963). The influence of subterranean clover pastures on soil fertility. I. Short-term effects. Aust.J.Agric.Res. 14:796-807.

43. Weir, R.G. (1984). Molybdenum deficiency in plants. Agfact AC.4 Department of Agriculture New South Wales.

44. Wells, GAl. (1969). Skeleton weed (chondrilla junces) in the Victorian Malle. I. Competition with legumes. Aust.(fip.Aaric.Anim.HU5b. 9:521-27.

45. Wells, G.J. (1970). Skeleton weed (chondrilla juncea) in the Victorian Mallee. II. Effect of legumes on soil fertility, subsequent wheat crop and weed population. Aust.J.Exj.Ajric.Anim.Hu5b. 10:622-29.

46. Young, R.R., Croft, P.H., Lattimore, M.E. and Lucy, M. (1987). Annual medics: Varieties and establishment in south-western New South Wales. In ‘Proceedings of the pasture evaluation workshop, Cobar’, ed. B.H. Dowing. Miscellaneous Bulletin 30, pp.40-45. Department of Agriculture New South Wales.