Previous PageTable Of ContentsNext Page

Desmanthus – a new pasture legume for the dry tropics

Chris Gardiner1, Leone Bielig1, Anthony Schlink 2, Ross Coventry1 and Michelle Waycott1

1 School of Tropical Biology, James Cook University, Townsville 4811, Queensland. Christopher.Gardiner@jcu.edu.au
2
CSIRO, Livestock Industries, Wembly 6913 Western Australia

Abstract

The legume genus, Desmanthus, comprises many species, some of which may be valuable in western Queensland grasslands where few, if any other sown pasture legumes persist. In the late 1980s, a selection of Alysicarpus, Centrosema, Chamaecrista, Clitoria, Desmanthus, Macroptilium, Stylosanthes and Vigna genotypes were sown in trials across semi-arid/arid western Queensland on clay soils in the Mitchell Grass Bioregion. In the early 1990s the trials ceased and the sites were abandoned but were reassessed 10 years later. Plant density data are presented for two sites from the establishment phase in 1989/90 to 2003. All genera other than several Desmanthus accessions failed in these harsh environments.

Key words

Desmanthus, clay soils, dry tropics, pasture legumes

Media summary

Desmanthus is a pasture legume with the potential to significantly enhance the sustainability and productivity of clay soil production systems in the dry tropics.

Introduction

Queensland’s Mitchell Grass Bioregion includes vast open grasslands and large areas of associated acacia ecosystems, which are used extensively for sheep and cattle production. An adapted pasture legume would contribute to the sustainability of the region by improving the quality of nutrition available to livestock as well as sustaining nitrogen levels for associated native and exotic grasses. The soils of the region have a high clay content, rainfall is summer dominant and very variable, mean annual rainfall (MAR) ranges from 250 mm - 550 mm/yr across the region (Orr 1975, Commonwealth of Australia 2001). The region lacks an adapted sown pasture legume. Leslie et al (1987) suggested that many of the original productive and palatable native herbaceous legumes of such regions were lost long ago due to grazing. Orr (pers. comm.) found that in similar regions where grazing utilization is > 10%, herbaceous legumes are a minor component of the available pasture. Native legumes also have been lost or reduced in other grasslands such as in Texas (Muir & Pitman 2003). In western Queensland, large areas of the native acacia trees have been cleared for buffel grass monoculture pastures (Butler & Fairfax 2003). Some years after tree clearing and grass pasture establishment, the grasses suffer from a “run down” effect, which is manifested as reducing productivity due to a decline in available soil N (Robertson et al 1997).

Methods

In 1989/90 CSIRO/DPI established experiments across Queensland’s Mitchell Grass Bioregion aimed at identifying potential pasture legumes for the region. Fifty-four accessions representing 8 legume genera were sown in the trials (Table 1). After initial evaluation, the trials ceased and the sites were abandoned. For several years since being abandoned, they have been unfenced and incorporated into grazed paddocks. Since planting, the sites have experienced periods of severe drought and floods (Table 2), locust plagues, frosts and often heavy grazing. A decade after planting, plot positions were located and plant density and soil seed bank data were collected. These attributes and others will be used as indices of the potential of the accessions for cultivar development. In this paper we focus on 2 sites in semi-arid rangelands: firstly at Blackall, a cleared gidgee (Acacia cambagei) light clay buffel (Cenchrus ciliaris) pasture with a MAR of 525 mm and secondly, at Isisford, a cleared gidgee very pebbly/stony light clay buffel/mitchell grass (Astrebla spp.) pasture, with a MAR of 451 mm. Table 4 describes the key soil attributes of each site.

In 1989/90, 10 grams of seed of each accession were planted into 5m x 4m plots with 3 replications. Seed was inoculated with appropriate rhizobia and seed was broadcast onto uncultivated plots, which were set out in a randomised plot design. Plant density (plants/m2) was recorded by counting all plants in each 20m2 plot. In 2003, seed bank data were collected from selected plots by inserting 3 steel cylinders (7cm diameter x 5 cm deep) per plot and washing the resulting soil cores through an Endecott 1mm gauge sieve.

Results

Table 1. Genera and accessions assessed, plants/m2 (mean of 3 replicates) for the establishment phase and for 10 and 14 years after planting at Blackall and Isisford. Isisford 2003 data is not shown as no plants were observed.

 

Accessions/

Blackall

Blackall

Blackall

Isisford

Isisford

Genus/Species

Cultivar

1989

1999

2003

1990

1999

Alysicarpis rugosus

TQ91

13

0

0

0

0

Chamaecrista rotundifolia

cv. Wynn

6

0

0

0

0

Clitoria ternatea

2 accessions

2 – 4*

0

0

0

0

Centrosema pascuorum

CPI 55697

1

0

0

0

0

Macroptilium atropurpureum

2 accessions

1 – 5*

0

0

1 – 2*

0

Stylosanthes scabra

2 accessions

22 – 23*

0

0

2 – 3*

0

S. hamata

7 accessions

0 – 26*

0

0

0 – 4*

0

Vigna trilobata

CPI 13671

6

0

0

6

0

Desmanthus bicornutus

CPI 81337

4

T

T

1

T

D. bicornutus

CPI 84508

9

0

0

1

0.3

D. bicornutus

CPI 90857

4

0

0

0

1.4

D. bicornutus

CPI 91162

17

0

0

2

T

D. covillei

CPI 90311

na

na

na

0

T

D. leptophyllus

TQ87

10

0

0

1

0

D. leptophyllus

TQ88

27

0

0

1

0

D. leptophyllus

TQ90

14

0

T

0

0

D. leptophyllus

CPI 37143

19

0

0

2

0

D. leptophyllus

CPI 38351

13

0

0

3

T

D. leptophyllus

CPI 55719

12

0

0

3

0

D. leptophyllus

CPI 76053

9

T

1

2

0

D. leptophyllus

CPI 92655

16

T

T

3

0

D. leptophyllus

CPI 92746

7

T

0.3

2

T

D. leptohyllus

CPI 92809

15

0

0

2

0

D. pernambucanus

CPI 40071

9

0

0.5

1

T

D. pernambucanus

CPI 49728

7

0.4

0.6

0

T

D. pernambucanus

CPI 83565

10

T

T

0

T

D. pubescens

CPI 92800

4

0

0

0

0

D. pubescens

CPI 92802

6

T

T

1

T

D. pubescens

cv. Uman

5

0

0

1

0

D. pubescens

CPI 92804

4

T

0

4

T

D. sp

CPI 33426

na

na

na

0

0.3

D. sp

CPI 70338

8

T

0.6

2

T

D. tatuhyensis

CPI 90362

3

0

0

0

0

D. virgatus

CPI 57960

13

1

1.5

1

0.5

D. virgatus

CPI 67643

28

1.3

1.5

0

1

D. virgatus

cv. Marc

10

3

9

0

0.4

D. virgatus

CPI 78372

11

1

4

0

0.3

D. virgatus

CPI 78382

6

0.8

2

0

0

D. virgatus

CPI 79653

4

1

2

0

0

D. virgatus

CPI 83563

13

0

0

0

T

D. virgatus

CPI 85173

3

1

1

0

0

D. virgatus

CPI 85178

6

2.5

9

0

0.3

D. virgatus

CPI 85182

13

1

0.7

1

0.6

D. virgatus

CPI 90751

6

2.7

5

0

0.5

D. virgatus

CPI 91181

8

2

3.5

0

0.4

* range of plants/m2 observed for 2 or more accessions of this species, na = not planted at site, T = trace (<0.3plants/m2)

Table 2 Total annual rainfall for Blackall and Isisford 1989 to 2003. Source: Bureau of Meteorology.

 

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

Blackall

698

741

388

410

455

418

345

540

703

756

787

706

480

275

260

Isisiford

740

873

397

399

412

313

260

422

522

691

860

546

558

164

284

Table 3 Seed bank (seeds/m2) recovered from plots of promising accessions at Blackall and Isisford in 2003.

Site

Desmanthus seeds/m2 in seed bank (mean of 3 cores/plot)

Accession

78373

85178

85182

90751

90857

Blackall

1044

261

-

783

-

Isisiford

522

-

174

174

87

Table 4 Soil properties of bulked top soil (0-10 cm) for the Blackall and Isisford sites

 

Texture
% clay

pH
(1:5 water)

Organic Carbon %C

Nitrate Nitrogen mg/kg

Phosphorus (Colwell) mg/kg

Potassium
(Amm.Ac.)
meq/100g

Electrical
conductivity
dS/m

Sodium (Amm.Ac.) meq/100g

CEC
meq/100g

Blackall

35

8.5

0.6

8.3

26

1.1

0.15

0.45

35.0

Isisford

34

8.8

0.7

44.5

20

1.0

0.20

1.75

44.0

Discussion

Of the species evaluated in this study, only Desmanthus genotypes remained in the trial plots 10 years after planting. Table 2 shows the extreme variability of the rainfall at the sites, including periods of severe drought. At Isisford for example, only 164 mm of rainfall was recorded in 2002 compared with 873 mm in 1990. The two sites have similar soil properties (Table 4) except that soil at the Isisford site has higher nitrogen and sodium levels and it also has a very pebbly/stony soil mantle.

Desmanthus is known to be nutritious (Schlink & Burt 1993; Jones et al 2000), palatable, productive, drought tolerant and adapted to clay soils (Cook et al 1993; Pengelly & Conway 1998). However, no Desmanthus cultivar has been released specifically for western Queensland where in fact, no suitable sown legumes are available and where an adapted legume would contribute to the productivity and sustainability of the grasslands. Elsewhere, Desmanthus cultivars have been released, for example in southern Queensland, a composite cultivar named Jaribu, made up of cvv Bayamo, Marc and Uman (Cook et al 1993) is available. Of these, Marc and Uman were included in our studies but of these only Marc survived and ranked highly at Blackall (Table 1) with 9 plants/m2 in 2003. A composite cultivar, which includes D. bicornutus CPI 90857, has been released in Texas, USA (Ocumpaugh et al 2003). D. illinoensis cv Sabine is used in temperate USA (Muir & Pitman 2003) and D. pernambucanus cv Chaland is used in SE Asia (Horne & Str 1999).

Assessment at the establishment phase of the trial indicated that Stylosanthes accessions (Table 1) were potential new cultivars for the region. However ten years later, only Desmanthus remains, demonstrating the value of long-term evaluation. In southern Queensland, Jones (1998) studying Desmanthus observed < 0.2 to 5 plants/m2 and suggested that it has promise for permanent pastures. In our trials, at Blackall, the superior accessions were all D. virgatus genotypes, including the Argentine lines, cv Marc (CPI 78373), and CPI 78372, and the Mexican lines, CPI 85178 and 90751, CPI 85178 had a mean plant density of 9 plants/m2 and up to 15 plants/m2 in one replicate in 2003. At Isisford in 1999, on the drier and stonier site, D. bicornutus CPI 90857 had a mean plant density of 1.4 plants/m2 with up to 3.25 plants/m2 in one replicate. This particular site will be a further test of the resilience and adaptation of Desmanthus in response to drought.

The 2003 seed bank data (Table 3), taken 2 years into very severe drought (Table 2), revealed Desmanthus seed banks ranging from 261 to 1044 seeds/m2 (mean of 696) at Blackall and 87 to 522 seeds/m2 (mean of 238) at Isisford. Elsewhere, Rangel & Gardiner (1996) found a mean Desmanthus seed bank of 729 seeds/m2 and Jones (1998) a mean of 1125 seeds/m2, while Burrows & Porter (1993) found a mean of 300 seeds/m2. Due to the severe 2002/03 drought (Table 2) at Isisford, all local grasses and forbs were dead by mid 2003. However, given the density of Desmanthus in 1999, the size of the seed bank and the current 2004 wet season Desmanthus is expected to regenerate as it did after the 1994/5 drought. Postscript: Desmanthus did indeed have recruitment in the 2004 wet season and was the sole sown legume to recover from the drought.

Conclusion

The data suggest that the population density of Desmanthus fluctuates considerably over time, illustrating the resilience of the genus and the importance of seedset, hardseedeness, seed bank longevity and recruitment. Of the 8 genera investigated, Desmanthus offers the best prospect for the development of cultivars adapted to dry tropical western Queensland clay soil environments. Selection and seed increase of the most promising accessions has commenced.

Acknowledgments

The Evans and Kent families; James Cook University; the Australian Wool Research Corporation, which funded the original CSIRO/QDPI project, are all gratefully acknowledged for their valuable support.

References

Burrows DM and Porter FJ (1993). Regeneration and survival of Desmanthus virgatus 78382 in grazed and ungrazed pastures. Tropical Grasslands 27, 100-107.

Butler DW & Fairfax RJ (2003). Buffel grass and fire in a Gidgee and Brigalow woodland: A case study from central Queensland. Ecological Management & Restoration 4(2), 120-125.

Commonwealth of Australia (2001). Mitchell Grass Downs. Australian Natural Resource Atlas http://audit.ea.gov.au/ANRA/rangelands/rangelands_frame.cfm?region_type=NT&region_code=MGD

Cook BG, Graham TW, Clem RL, Hall TJ, & Quirk MJ (1993). Evaluation and development of Desmanthus virgatus on medium to heavy textured soils in Queensland. Proceedings of the XVII International Grassland Congress, Rockhampton, 1993. 3, 2148 – 2149.

Horne PM & Str WW (1999). Developing forage technologies with farmers – how to select the best variety to offer farmers in SE Asia. ACIAR Canberra.

Jones RM (1998). Evaluation of a range of tropical legumes on two clay soils in SE inland Queensland. Tropical Agriculture Technical Memorandum no.2. CSIRO.

Jones RM, Bishop HG, Clem RL, Conway MJ, Cook BG, Moore K and Pengelly BC (2000). Measurements of nutritive value of a range of tropical legumes and their use in legume evaluation. Tropical Grasslands 34, 78-90.

Leslie JK, Mungomery VE, Cameron DG & Hogarth DM (1987). Plant genetic constraints in tropical agriculture. 57th ANZAAS Congress, James Cook University 24-28 August 1987.

Muir JP & Pitman WD (2003). Establishing bundleflowers in bumuda and switchgrass. Forage research in Texas 2003.CPR 5267. http://forageresearch.tamu.edu/2003/2003%20desmanthus%20establishment.pdf

Ocumpaugh WR, Kuntz D, Rahmes J, Martinez D, Grichar J, Hussey M, Reilly J, Abrameit A, Owens K, Reed R, Muir JP, & Bade D (2003). Desmanthus bicornutus: a new summer-growing perennial shrubby legume for south Texas and Mexico. Proceedings of the American Forage and Grassland Council. Vol. 12.

Orr DM (1975). A review of Astrebla (Mitchell Grass) pastures in Australia. Tropical Grasslands 9, 21-36.

Pengelly BC & Conway MJ (1998). The evaluation of tropical legumes for use in ley pastures in central and southern Queensland. Proceedings of the 9th Australian Agronomy Conference, Wagga Wagga, 1998. http://www.regional.org.au/au/asa/1998/1/278pengelly.htm#TopOfPage

Rangel JH deA & Gardiner CP (1996). Effect of fire and heat on the seed germination of Desmanthus virgatus accessions. Proceedings of the International symposium on the future of the tropical savannas, Townsville 1994.

Robertson FA, Myers RJ & Saffigna PG (1997). Nitrogen cycling in brigalow clay soils under pasture and cropping. Australian Journal of Soil Research 35, 1323 – 1339.

Schlink AC & Burt RL (1993). Assessment of the chemical composition of selected tropical legume seeds as animal feed. Tropical Agriculture 70, 169-173.

Previous PageTop Of PageNext Page