Tải bản đầy đủ - 0 (trang)
V. Strategies to Enhance N2 Fixation

V. Strategies to Enhance N2 Fixation

Tải bản đầy đủ - 0trang

NITROGEN FIXATION BY TROPICAL LEGUMES



a



b



C



d



203



FIG. 8. Effects of fertilizer N and rhizobial inoculation on (a) nodulation, (b) relative

abundance of ureide-N in xylem sap, (c) P,and (d) shoot dry matter of field-grown plants of

Bragg soybean. Data derived from Herridge and Brockwell(l988).



204



MARK B. PEOPLES AND DAVID F. HERRIDGE



numbers and soil nitrate indicated that they were highly correlated. Thus,



P could be expressed as

P(%) = 40 + 4.4(z) - 0.4(Soil) + 0.001(Soi1)2



(r2 = 0.80)



(12)



where z = loglo number of B. japonicum in the seed zone (3 to 15 cm depth)

at sowing and Soil = amount of soil nitrate to 0.9 depth (kg N/ha) at

sowing.

Although this function described a specific crop and environment and

was not intended as a universal equation describing the relationship between soybean N2 fixation and soil N fertility and rhizobial status, it does

serve to illustrate the predictable and quantitative nature of N2 fixation. In

this instance, fixation was regulated essentially by only two variables, with

soil nitrate and rhizobial numbers accounting for 80% of the variation.

Plant (crop) yield is another determinant of N2 fixation, particularly when

the levels of soil N are moderate and there are sufficient numbers of

effective rhizobia present. Yields in these instances are related to cultivar,

mediated through growth rates or crop duration, or to nutritiodwater

availability. Thus, P can be described by

P(%) = (b + a/N1) x 100



(13)



The function is derived from

N2 fixed = a



+ b(N1)



(14)



(e.g., see legend to Fig. 9).

Examples of the strong relationship between N yield and N2 fixed, and

from that between N yield and P, are presented in Fig. 9. Crops involved

are soybean (Hardarson et al., 1984), pigeon pea (Kumar Rao and Dart,

1987), and faba bean (Viciafaba; Duc et al., 1988). The three experiments

involved assessment of N yield and N2 fixation of a range of cultivars.

Generally N yield varied with crop duration. Each data set is site specific,

with the values for N yield for unnodulated plants (i.e., values for x when y

= 0; Fig. 9a) being determined by the levels of soil nitrate. Modifying soil

nitrate will shift the line of best fit either to the left or to the right. The

reaction of the individual cultivars to nitrate affects the slope of the line.

Increasing the proportions and amounts of N2 fixed by legumes therefore will be achieved by: ( a ) maximizing legume yield within the constraints imposed by agronomic and environmental considerations; ( b ) reducing the legume’s sensitivity to the suppressive effects of nitrate on

nodulation and N2 fixation (nitrate tolerance); (c) optimizing the numbers

and effectiveness of rhizobia in the rooting zone, through strain selection

and inoculation techniques, and through plant breeding for promiscuous or

selective nodulation; and (d)reducing the amount of nitrate in the rooting

zone.



205



NITROGEN FIXATION BY TROPICAL LEGUMES



I

loo



Faba bean

Pigeon pea

Soybean



1. S



80

60



40



20

0



0



100



200



300



0



100



200



300



Crop N (kg/ha)



FIG.9. Relationships between N yield and (a) N2 fixed and (b) P for field-grown crops of

faba bean, pigeon pea, and soybean. Relationships in (a) are described by: (0)y = -32

( ? = 0 0 . 9 8 ) ; ( 0 ) y = - 6 5 + ].OX(?= 1.00);(.)~= -127+ 1 . 0 ~ ( 3 = 0 . 8 8 ) .



+0.9~



In the following sections, examples of programs to enhance N2 fixation

are examined. In each case, progress is achieved through one or more of

the preceding factors.

A. PLANTBREEDING

A N D SELECTION

The challenge to improve the N2 fixation capacity of the legumes

through selection and breeding is complex because there are two components to consider: the host plant and the rhizobia. Although host genotype x strain interactions have been shown to occur with a number of

agriculturally important legumes (Graham and Temple, 1984; Mytton,

1984), many of the programs selecting for enhanced N2-fixing potential

have chosen to ignore this complication. Instead, the programs have either

relied on nodulation by the indigenous rhizobia (Kueneman et al., 1984),

utilized commercial inoculants (Betts and Herridge, 1987; Herridge and

Betts, 1988), or prepared inoculants containing either single strains or

small numbers of strains (Kumar Rao and Dart, 1987; Neuhausen et al.,

1988). The focus, therefore, is entirely on the identification of variation in

N2 fixation attributable to the host.

1 . Legume Yield



Although it may be a simple matter to identify cultivars of legumes with

increased plant yields and therefore increased N2fixation (e.g., Hardarson



-&

a



206



MARK B. PEOPLES AND DAVID F. HERRIDGE



et al., 1984; Kumar Rao and Dart, 1987), the real challenge is to select

cultivars that maintain higher levels of N2 fixation and P at the same level

of yield. As we argued earlier, agronomic and environmental considerations may limit the size of individual plants and the duration of the crop

must be taken into account. With species such as common bean, low N

yield remains a major constraint to N2 fixation. In fact, of the commercial

crop legumes, the common bean is regarded as the weakest at fixing N2 and

is supplied with fertilizer N in most cases. There is some evidence that the

low N2-fixing potential of common bean may be associated with low

specific nodule activity (Felix et al., 1981; Pereira and Bliss, 19891, which

is associated with high levels of H2 production, that is, low relative efficiency (RE) of N2 fixation (Pacovsky el al., 1984; Piha and Munns, 1987a).

Results of Hungria and Neves (1987) endorsed these findings by showing a

strong inverse relationship between nodule H2 production, nodule specific

activity, and plant N yield. Hydrogen evolution and therefore nodule RE

were affected by both host cultivar and Rhizobium strain. In other programs, high levels of fixation were associated with late maturity and

climbing habit (Graham and Rosas, 1977; Rennie and Kemp, 1983; Piha

and Munns, 1987b). This implied a simple relationship between leaf area

duration and N2 fixation (see also Wynne et al., 1982), or patterns of

carbohydrate partitioning within the plant (Graham and Halliday, 1977).

A breeding program by Bliss and coworkers at the University of Wisconsin has produced new genotypes of common bean with increased plant

vigor, increased N yields, and higher levels of N2 fixation. Selected hybrid

lines have displayed between three- and sevenfold increases in total N

fixed and two- to fourfold increases in P relative to the commercial parent

cultivar Sanilac (Table XVII, Experiment 1). For all the hybrids, however,

fixation capacity was substantially less than the capacity of Puebla 152, the

high-fixing donor. In a second experiment, the superiority of hybrid line

24-21 was obvious. The line displayed higher rates of growth and N accumulation (+36%) and more total growth (+79%) than the commercial

parent, while retaining the short season and determinate characteristics

(Table XVII, Experiment 2).

2 . Nitrate Tolerance



Nitrate is one of the most potent inhibitors of N2 fixation (Table VII)

(Streeter, 1988). Development of symbioses in which P is maintained at

near maximum levels in the presence of high soil nitrate could provide the

biggest single advance in the improvement of N2 fixation by legumes.

Plant mutagenesis has been used to generate phenotypes exhibiting



207



NITROGEN FIXATION BY TROPICAL LEGUMES

Table XVII

Summary of Data from Two Experiments from a Breeding Program to Increase Nz

Fixation by Common Bean“



Experiment 1

N 2 fixation

Parent

or line



P



(mg N/plant)



Seed yield

(g/plant)



Maturity

(days)



Determinate



(mg N/plant)



Sanilac

24-17

24-21

24-55

Puebla



0.12

0.48

0.25

0.22

0.57



76

583

216

192

852



18

31

19

23

38



85

I10

91

94

120



Yes

No

Yes

Yes

No



591

1068

I045

668

1429



a



Amount



Experiment 2

Total N



Derived from Attewell and Bliss (1985).



greater nodule mass under field conditions (e.g., Rosaiah et al., 1987) or

forming high numbers of nodules in the presence of nitrate (e.g., “supernodulating” soybean; Carroll et af., 1985). Grafting experiments indicate

that the supernodulating trait identified in soybean is mediated through the

shoots, probably by the nonproduction of factors regulating the number of

successful rhizobial infections that develop into nodules (Delves et af.,

1986). Extreme supernodulating mutants can form up to 10-fold more

nodules in both the absence and presence of nitrate, but this trait results in

substantial reductions in root and shoot growth (Day et al., 1986), and

despite higher P values, the mutants are only capable of fixing more N than

the original wild-type parent under very high nitrate conditions (Hansen et

af., 1989).

Another approach that has been considered is the selection of mutants

with a reduced ability to utilize nitrate, that is, lowered nitrate reductase

activity (Nelson et al., 1983; Carroll and Gresshoff, 1986). From biochemical characterizations of such mutants it would appear that legumes can

contain several constitutive and nitrate-inducible nitrate reductase isoenzymes (Nelson et af., 1984; Streit and Harper, 1986), and a genotype

completely lacking nitrate reductase has yet to be reported. Nodulation

and N2 fixation are still apparently sensitive to nitrate in those mutants

already identified as having lowered nitrate reductase activities.

Species differ considerably in their symbiotic tolerance to mineral N and

sufficient natural variation may already exist among legume lines and

cultivars so that it might not be necessary to resort to mutagenesis procedures to induce further variation for breeding purposes (e.g., Hardarson el



208



MARK B. PEOPLES AND DAVID F. HERRIDGE



al., 1984; Harper and Gibson, 1984; Gibson and Harper, 1985). A program

was commenced in 1980 under this premise to screen 489 diverse genotypes of soybean for tolerance to nitrate (Betts and Herridge, 1987). In the

first two cycles of screening, all genotypes were grown in sand-filled pots

in the glasshouse and supplied with either nitrate-free nutrients or nutrients containing 2.5 mM nitrate. Plants were sampled at late flowering for

plant growth, nodulation, and N2 fixation (relative ureide-N in xylem sap

and plant parts). Results indicated that genotypes of Korean origin displayed higher-than-average levels of symbiotic activity in the presence of

nitrate. Of the original 19 Korean lines, 80% were included in the second

screening, and 47% were selected for subsequent field screening. Only 5%

of the remaining 470 genotypes were selected as high-fixing after the

second glasshouse screening. In the third year, 40 genotypes (including the

32 identified high-fixing lines) were sown into a high-nitrate soil (260 kg

N/ha to a depth of 1.2 m) in the field (Herridge and Betts, 1988). The

genotypes showing highest levels of nodulation and N2fixation under these

conditions were all Korean types (Table XVIII). They had shoot biomass

similar to the commercial cultivars Bragg and Davis, but had reduced

uptake of soil N as indicated by higher recoveries of soil nitrate from the

Korean plots immediately after seed harvest (Table XVIII). Seed yield of

the Korean lines was around 30% less than that of the commercial varieties. Correlation matrices among the indices of nodulation, N2 fixation,

plant growth, and seed yield revealed independence between the symbiotic- and yield-related characters. Therefore, the nitrate-tolerant Korean

Table XVlll

Measurements of Nodulation and N1 Fixation by, and Growth and Yield of, NitrateTolerant and Commercial Genotypes of Soybean in a High-Nitrate Field Soil. *

~~



Genotype

Nitrate-tolerant lines

Korean 466

Korean 468

Korean 469

Commercial cultivars

Bragg

Davis



Nodulation

weight

(mg/plant)



Nodulation

number



Shoot

Seed

Residual

dry matter yield soil nitrate'

@/plant) (tlha) (kg N/ha)



Pb



316

254

I76



34.5

16.8

19.5



0.31

0.18

0.22



45.9

43.3

41.6



I .6

1.7

1.4



64

19

16



24

40



2.0

1.3



0

0



39.7

48.5



2.2

2.2



45

n.a.



Data from Herridge and Betts (1985, 1988).

Calculated using the ureide method according to Herridge and Peoples (1990).

To a depth of I .2 m.



209



NITROGEN FIXATION BY TROPICAL LEGUMES



lines were used as high-fixing donor parents in a breeding program with

selection for both seed yield and N2-fixingcapacity.

The four outstanding Korean lines (K464, K466, K468, and K469) were

crossed with the commercial cultivars Forrest, Bossier, Reynolds, and

Valder. Approximately 1500 seeds from F2 combinations were sown into a

high-nitrate soil in the field in 1986. Seedlings were individually tagged and

assessed during growth for plant and seed characteristics, growth habit,

and agronomic type. Symbiotic N2 fixation of individual plants was evaluated during early pod fill using a nondestructive xylem ureide procedure

(Herridge ef al., 1988). The mean level of N2-fixingactivity of the Fz’s was

surprisingly constant for the 1 1 combinations and was between that of the

commercial parents (P = 0-0.20) and the high levels displayed by 3 of the 4

Korean parents (P = 0.40) (Fig. 10). Although the average levels of fixation

of the F2’s were below those of the best Korean lines, 35 individual F2’s

displayed equally high levels, that is, P > 0.40. Culling of the F2 population

for progression to the F3generation was made on the basis of N2 fixation

(xylem relative ureides > 32%, see Fig. lo), plant type, and seed color.

To establish the heritability of N2 fixation, the seed of 100 F2 lines,

selected to cover all 1 1 families and range of xylem ureide levels, was sown

in a replicated experiment in 1987. Levels of N2 fixation were again assessed for F3 plants by the ureide method and compared with data for the



selected for

F3 generation



Mean



634 F2‘s

rejected as

low-fixing



s

FIG.10. Ranges of relative abundance of ureide-N in xylem sap for the 1 1 F2 families and

for the commercial and Korean parents. Plants were grown in high-nitrate soil in the field. The

horizontal line indicates the cutoff value (32%) for selection of material for the F3generation.



210



MARK B. PEOPLES AND DAVID F. HERRIDGE



same lines the previous season. When data were divided into groups

designated by a single common parent, correlations (broad-sense heritabilities) between F2 and F3 xylem ureide data were higher (range r =

0.24-0.72) relative to the pooled (Le., all parents) data ( r = 0.27). This

implied that the separate populations behaved differently in the genetic

control of N2 fixation. The generally significant correlations between F2

and F3 relative ureides indicated that nitrate tolerance was under quantitative genetic control with a broad-sense heritability of between 0.24 and

0.72 (Rose et al., 1989).

B. RHIZOBIA,

INOCULATION,

AND PLANTNODULATION

Legume inoculation is a long-established and successful practice. Vincent (1965) and others (e.g., Allen and Allen, 1958) have argued that it is a

desirable practice in most agricultural soils throughout the world. Date

(1977), however, cautioned that the need to inoculate was not universal

and should be carefully determined for each individual situation before

investing in inoculant production and use.

There are three major groups of legumes that can be distinguished on the

basis of compatibility with a range of strains of Rhizobium (Table XIX). At

one extreme is a group of legumes that can form an effective symbiosis

with a wide range of strains. Members of this group were nodulated by

“cowpea-type” rhizobia and these Rhizobium spp. are so widespread in

tropical soils that such legumes seldom respond to inoculation. Yet even

within this supposedly “promiscuous” group that can be some host-strain

specificity in terms of the symbiotic effectiveness of the associations

formed (Gibson et al., 1982).

At the other extreme are legumes with very specific rhizobial requirements. These specificities are most relevant when the legume is introduced

to new areas. Response to inoculation of these legumes is usually successful provided that adequate numbers of rhizobia are applied at sowing. The

third and intermediate group of legumes nodulate with many strains of

Rhizobium, but effectively fix N2 with only a limited number of them. Thus

inoculation and nodulation failures are more frequent because the inoculum strain is unable to compete with the ineffective but established soil

populations of rhizobia.

There are a number of conditions under which soils may be devoid of

Rhizobium to form an effective symbiosis with a legume and which may

warrant inoculation: (a) the absence of the same or a symbiotically related

legume in the immediate past history; ( 6 ) poor nodulation when the same

crop was grown previously; (c) when the legume follows a nonleguminous



NITROGEN FIXATION BY TROPICAL LEGUMES



21 1



Table XIX

Legumes Grouped on the Basis of Nodulation and N2 Fmation with a Range

of Rhizobium Speciesa



Nodulate effectively with a wide range of strains.

Genera listed forming one loose group.

Albizia

Alysicarpus

Arachis

Calliandra

Calopogonium

Cajanus

Canavalia

Clitoria

Croialaria

Dolichos

Eryihrina



Galaciia

Gliricidia

Indigofera

Lablab

Lespedeza

Macropiilium

Macroiyloma

Mimosa

Pachyrhizus

Pongamia

Neonotonia



Psophocarpus

Pueraria

Rhynchosia

Siylosanihes (several

subgroups)

Tephrosia

Teramnus

Vigna

Voandzeia

Zornia



Nodulate with a range of strains but often ineffectively. Genera listed forming individual

groups with some crossing between groups. Subgroups distinguishable.

Acacia

Adesmia

Aeschynomene



Astragalus

Centrosema ( 2 subgroups)

Desmanihus

Desmodium (2 subgroups)



Psoralea

Sesbania(2 subgroups)



Nodulate effectively with specific strains only. Genera listed forming specific groups.

Cicer

Coronilla

Glycine max

Hedysarum

Lathyrus

Lens

Leucaena

a



Lotononis-Listia ( 3 subgroups)

Lotus ( 3 subgroups)

Lupinus (2 subgroups)

Medicago

Melilotus

Onobrychis

Ornithopus



Phase o1us

Pisum

Trifolium (many subgroups)

Trigonella

Vicia



After Peoples et al. (1989a).



crop in a rotation; (d)in land reclamation; and, (e) when environmental

conditions are unfavorable for Rhizobium survival (e.g., very acidic or

alkaline soils, under prolonged flooding, or very hot, dry conditions prior

to planting).

However, as a farming practice, inoculation generally remains the exception rather than the rule (Vincent, 1982). Exacting technology is essential for the production and distribution of inoculants (Date and Roughley,

1977),and the success of inoculation in the field depends on the procedure

used and operator competence (Brockwell, 1980; Brockwell et al., 1988).

In Australia and the United States, legume inoculation has played a funda-



212



MARK B. PEOPLES AND DAVID F. HERRIDGE



mental role in the establishment of legume-based pasture and cropping

systems, but less use has been made of inoculants elsewhere. In Latin

America, only two countries use inoculants to any extent and even in

Brazil, the largest producer of the seed legumes, common beans are fertilized with N rather than inoculated (Freire, 1982).

Inoculation responses in tropical soils appear to be confined to crops

such as soybean which have specific Rhizobium requirements (Ayanaba,

1977; Halliday, 1985). Typically responses are substantial when indigenous, infective rhizobia are absent and soil nitrate is low (Table XX). In a

high-nitrate soil, nodulation of, and NZfixation by, inoculated plants can

be suppressed. However, nodulation and NZfixation (but not necessarily

yields) could be increased by very high rates of inoculation ( 1 0 0 0 ~normal;

see Table XX; Bergersen et al., 1989).

When populations of infective rhizobia exist in high numbers in soils,

they present a formidable barrier to the successful exploitation of superior

Table XX

Effect of Inoculation on Nodulation, N2Fixation, and Productivity of Bragg Soybean

Grown in Various Backgroundsof Soil Nitrate and Bradyrhizobium japonicum"

~



Treatment

B. japonicum-free soil

Low nitrate

Uninoculated

Normal inoculation

High inoculation

High nitrate

Uninoculated

Normal inoculation

High inoculation



High B. japonicum

soil

Low nitrate

Uninoculated

Normal inoculation



Nodule

massb

(mglplant)



P'



Crop

dry

matte@

(tlha)



Crop

Nd

(kg N/ha)



Seed

yield

(t/ha)



66

I68

I95



I .7

3.3

3.0



0

72

334



0.11

0.61



4.9

6.8

7.0



0

4



0

0

0.17



8.9

8.0

8.5



205

I96

213



3.2

2.9

3.3



9.8

9.3



258

24 I



2.2

2.0



50



I29

146



0.44



Derived from Herridge et a!. (1987); Herridge and Brockwell(1988).

Sampled either 62 or 70 days after planting.

Calculated by the ureide method according to Herridge and Peoples (1990). Values

represent the mean of four sap samplings between days 70 and 109.

Sampled when shoot dry matter and N were at a maxima.

a



NITROGEN FIXATION BY TROPICAL LEGUMES



213



strains of Rhizobium used as inoculants (see Table XX) (Devine, 1984). In

the United States, large populations of soybean Rhizobium have become

established in soil with cropping so that now less than 10% of nodules are

formed on soybean by the inoculant and yield responses are rare (Berg et

al., 1988; Halliday, 1985). Research programs in several laboratories (e.g.,

Devine, 1984; Cregan and Keyser, 1986) currently aim to produce soybean

cultivars that bypass the resident rhizobia in the soil to be nodulated by

better, selected inoculant strains (this assumes that fixation and N supply

are limited by the effectiveness of indigenous rhizobia). A similar strategy

was employed for groundnut by Nambiar et al. (1984), who exploited

host X strain specificity with cultivar Robut 33-1 and strain NC 92 to obtain

consistent yield responses (mean of 16% over nine experiments) in soils

containing moderate to high numbers of infective rhizobia and where all

uninoculated plants were well nodulated.

A contrasting strategy is to develop varieties that can be effectively

nodulated by the resident soil rhizobia. Nangju (1980) observed that soybean genotypes from Southeast Asia nodulated successfully with the indigenous rhizobia in Nigeria, but U .S.-bred cultivars nodulated poorly

without inoculation. Hybridization of the Asian and U.S. types has resulted in high-yielding lines capable of fixing large amounts of N without

inoculation (Kueneman et al., 1984).

C. CROPA N D SOILMANAGEMENT



Both yield and P can be influenced by crop and soil management; we

examine two important practices.

1 . Tillage



Cultivation accelerates the oxidation of organic matter in soils (Doran,

1980) and generally results in higher nitrate-N in the profile (e.g., Thomas

et al., 1973; Dowdell et al., 1983). Cultivation may also decrease the rates

of denitrification (Doran, 1980; Rice and Smith, 1982), immobilization

(Rice and Smith, 1982), and leaching (Thomas et al., 1973) of nitrate

compared to in untilled soils. Additional fertilizer N may be required by

cereals under reduced tillage treatments, but for legumes, the lowered soil

nitrate levels should result in enhanced N2 fixation. No-till systems can

also modify and improve soil structure to create more favorable soil moisture and temperature regimes for plant growth (Lal, 1989).

In cropping systems research involving soybean in a moderate rainfall



Tài liệu bạn tìm kiếm đã sẵn sàng tải về

V. Strategies to Enhance N2 Fixation

Tải bản đầy đủ ngay(0 tr)

×