Brachiaria decumbens 


Photo1.jpg (1333 bytes)



  • Brachiaria eminii
  • Brachiaria bequaertii


Author: Max Shelton


Common names

(English) Surinam grass, signal grass, Kenya sheep grass, sheep grass, (Spanish) braquiaria decumbens, pasto alambre, pasto braquiaria, pasto chontalpo, pasto de la palizade, pasto de las orillas, pasto peludo, pasto prodigio, zacate prodigio, (Portugese) australiano braquiaria, braquiaria comum, braquiaria de alho, capim brachiaria decumbens, (Malay) rumput signal, (Thai) ya-siknaentonnon, ya-surinam


Overview: Native to Africa and now widespread in the tropics and sub-tropics, the forage potential of the grass Brachiaria decumbens was first recognized about 40 years ago, mainly in restricted ecological niches in tropical Australia. Its full potential was realised only in the past 20-25 years, when a handful of cultivars, derived directly from naturally occurring germplasm, was widely sown in tropical America. Keller-Grein et al. (1996)suggest that "Brachiaria is now the most widely used tropical grass genus, especially in Central and South America. In Brazil alone, about 40 M ha of Brachiaria pastures exist, more than 85% of which consists of B. decumbens cv. Basilisk and B. brizantha cv. Marandu".

This rapid expansion did not occur without problems, and currently available cultivars are now recognized as having serious defects, the most important being susceptibility to spittlebugs particularly in B. decumbens and B. ruziziensis". Over the past decade, greatly enhanced germplasm collections and studies on the cytology and genetics of Brachiaria spp. have opened up new opportunities and challenges for the improvement of this forage.


Brachiaria decumbens is important because of its high productivity under intensive use, and its tolerance of low fertility and relative freedom from pests and disease, apart from spittlebugs. It also performs well under coconuts. It is a valuable grass for erosion control as it covers the ground well, it withstands heavy grazing and establishes on poor and rocky soils.

Brachiaria species produce high yields, show excellent response to fertilizer, are persistent, and remain green long into the dry season. Data on nutritive value indicate that forage from Brachiaria is highly palatable to stock, leading to high intake, whether fed fresh or grazed in situ(Ndikumana and Leeuw de 1996).

Despite the constraints, the few available cultivars play a major role in world livestock production systems.

Production statistics

Pizarro et al. (1996) report that Brachiaria is adapted to the savanna ecosystem of South America which is varied and extensive, covering about 250 M ha. The savanna "is characterized by a well-defined dry season and acid, low-fertility soils. A few Brachiaria species (principally B. decumbens and B. brizantha) have shown wide adaptation and are extensively used as pasture grasses in this ecosystem. These were introduced from Africa in the 1950s and 1960s, and spread, at first, vegetatively and then by seed, covering today an estimated 70 million hectares" (Pizarro et al. 1996).

Argel and Keller-Gren (1996) report that "Brachiaria species (principally B. decumbens and B. brizantha) have become important components of sown pastures in the humid lowlands of tropical America. This ecosystem occupies about 50% of Brazil; 60% of the area encompassed by Bolivia, Peru, and Ecuador; 14% of Mexico; and significant areas in other countries of the region".

In Asia, and the South Pacific "Brachiaria species occupy about 300,000 hectares and they are the most widely grown pasture grasses in the humid and subhumid tropics. In Australia, the area to which they are adapted is relatively small, but within it, Brachiaria occupies more than half the area of improved pastures. Since seed of B. decumbens became available in the early 1970s, this has become the most widely planted species. Brachiaria ruziziensis, introduced into Australia in the late 1960s, was soon replaced there by B. decumbens. The success of Brachiaria species can be attributed to their broad adaptation and to their aggressiveness and resilience, which enable them to persist even under unfavorable conditions"(Stur et al. 1996).

In Africa, use of Brachiaria is less widespread and "further research is needed to place selected ecotypes of Brachiaria and other perennial forage species in the context of farming systems and develop integrated crop-livestock production systems for sub-Saharan Africa" (Ndikumana and Leeuw de 1996).

General plant description

The review of Renvoize et al. (1996) indicates that "within the tribe Paniceae, the principal characters that identify the genus Brachiaria are ovate or oblong spikelets, arranged in one-sided racemes, with the lower glume adjacent to the rachis. These characters, however, are by no means consistent throughout the genus, and in those species in which the spikelets are paired and borne on a triquetrous rachis, the orientation of the spikelets is often difficult to determine".

"Brachiaria belongs to a small group of genera that includes Urochloa, Eriochloa, and Panicum. All have the PEP-CK (phosphoenolpyruvate carboxykinase) type of C4 photosynthetic pathway (Clayton and Renvoize, 1986) and, although they have been recognized for over 100 years, the precise boundaries of these genera are still in doubt. Urochloa is scarcely separable from Brachiaria, differing in little but the orientation of the spikelets" (cited from Renvoize et al. 1996).

Renvoize et al. (1996) group Brachiaria into 9 groups, B. decumbens belongs to Group 5 (containing six species), but is allied to Group 6 (containing nine species), as both are African, and have elliptic oblong spikelet shape. They are "maintained as two separate groups on the basis of (1) the length of the lower glume, which may be very short and cuff-like (as opposed to almost as long as the spikelet), and (2) the presence or absence of reticulate venation in the upper glume and lower lemma. Rachis shape ranges from triquetrous to winged".

"This group includes the three most important species currently used for pasture development: B. brizantha, B. decumbens and B. ruziziensis. The first two species are closely related and at times difficult to distinguish. Brachiaria ruziziensis may be identified by its exceptionally broad, winged rachis, 2.0-3.5 mm wide; B. decumbens likewise has a winged rachis, 1.0-1.7 mm wide. In both these species, the spikelets are borne in two rows, and the glumes and lower lemma are membranous in texture. Brachiaria brizantha has a crescentic rachis that is seldom more than 1 mm wide; the spikelets are borne in a single row; and the glumes and lower lemma are cartilaginous in texture. Furthermore, it is distinguished from the other two species by its erect, tufted habit and often much longer leaf blades. Brachiaria decumbens and B. ruziziensis are both decumbent in habit and have lanceolate leaf blades" (Renvoize et al. 1996).

"Brachiaria brizantha and B. decumbens are inclined to intergrade in their vegetative features, and ambiguous specimens may be distinguished only by the rachis shape and spikelet arrangement and texture. The most widely grown genotype, cv. Basilisk (CIAT 606), which is currently being used in hybridization programs at CIAT and EMBRAPA, was originally identified as B. decumbens, but after reassessment was reidentified as B. brizantha. Brachiana eminii is also closely related and has been confused with B. decumbens; it is most readily distinguished by its annual habit. Other differences are the 3-10 racemes in the inflorescence; a slightly longer lower glume, one-half to two-thirds the length of the spikelet; and a mucronate upper lemma" (Renvoize et al. 1996).


Roots: A stoloniferous base and roots developping from the lower nodes producing a dense sward.

Stems: The erect stems arise from a long stoloniferous base.

Leaves: Heavy lanceolate leaf-blades 8-10 mm wide.

Flowers: Two to five racemes, 2-5 cm long with broad ciliate rachis and 4 mm long spikelets.


Spikelets: The are born in two rows, and the glumes and the lower lemma are membranous in texture.



Brachiaria decumbens has as its natural habitat the open grasslands and partial shade on the Great Lakes Plateau in Uganda and adjoining countries of Eastern and Central Africa.


Latitudinal range: about 27°N and S.


Altitude range: sea-level to 1,750m.


Temperature for optimal growth of B. decumbens is 30-35°C. It is readily frosted, but its winter production is better than Digitaria decumbens in frost free areas. Low temperature depresses growth; therefore, Brachiaria in general, performs poorly at altitudes above 1,800 m(Ndikumana and Leeuw de 1996).


B. decumbens is essentially a grass of the wet tropics, but it has good drought tolerance and is adapted to a dry season of four or five months. However, it prefers 1,500 mm or more of rain. It does not do well where the dry season is more than five months, but is more productive than Digitaria decumbens, Panicum maximum and Brachiaria mutica in the late dry season.

In Australia "Loch (1977) assessed B. decumbens as better adapted to the humid tropics, with a dry season of less than 4 months and an annual rainfall of more than 1,400 mm". "In the strongly seasonal climate of the isothermic savannas of the Brazilian Cerrados, however, B. decumbens cv. Basilisk is grown in areas where the dry season is as long as 7 months and rainfall as low as 1,300 mm. It extends further into drier zones than B.humidicola. Brachiaria brizantha is reputed to tolerate drought better (Thomas and Grof, 1986) than either B. decumbens or B. humidicola. All three species grow well throughout the year in the piedmont of the eastern Cordillera of the Andes in Colombia, where rainfall is more than 4,000 mm" (cited from Fisher et al. 1996).

Fisher and Kerridge (1996) noted that "at two sites, with contrasting soils, on the Caribbean island of Martinique, Gayalin (1994) compared the performance of B. decumbens with Panicum maximum, Pennisetum pureureum, and Tripsacum laxum as forages for deferred grazing in the dry season. Although B. decumbens is widely grown by farmers for its drought resistance, it was outyielded by T. laxum, which retained three to seven times more green leaf; P. pureureum yielded two to three times more total forage than B. decumbens and had between one and three times as much green leaf dry matter (DM). Obviously, farmers' perception of drought resistance, probably on the basis of the apparent proportion of green leaf in the standing forage, was not a good guide to the amount of green leaf actually present in the forage".

Radiation: Fisher and Kerridge (1996) reported that "Brachiaria species are used as soil covers in many plantation crops, such as rubber and coconut, in Southeast Asia and the Pacific Islands. Their tolerance of shade is therefore of interest". In one experiment, thirty-five forage grass accessions were grown under coconut on fertile soils in North Sulawesi and Bali, Indonesia, with light transmissions of 73% or 58%, respectively. Rainfall (amount and distribution) was confounded with light transmission: higher total rainfall and more even distribution occurred at the site with greater light transmission. Brachiaria decumbens cv. Basilisk was the top performer at the site with the higher rainfall, less shade, and a 12-month growing season (Kaligis and Sumolang, 1991). It was also one of the better performers at the site with the lower rainfall, more shade, and a more marked dry season (Rika et al., 1991)" (cited from Fisher and Kerridge 1996).

Photoperiodism: B. decumbens is inferred to be a quantitative short-day plant, flowering everywhere in the longer days of the year, and more vigorously at high than at low latitudes (Hopkinson et al. 1996).


Rao et al. (1996), in their review suggest that "Brachiaria species adapt to a wide range of soil types, from Oxisols and Ultisols (low-fertility acid soils) to {{Alfisoils}} and Mollisols (high-fertility neutral soils). They perform, much better on acid soils than other grasses, such as Panicum species. They also perform well on moderately fertile to very fertile soils".

"Several commercially grown Brachiaria species are well adapted to low-fertility, acid soils of the tropics. Research to identify plant attributes that contribute to efficient acquisition and use of nutrients for plant growth is recent. Several root and shoot attributes have been shown to contribute to the adaptation of Brachiaria species to acid soils; these include their ability to change the partitioning of fixed carbon to favor root growth, to acquire N through associative fixation, to acquire P through extensive root systems and mycorrhizal association, and to acquire Ca through highly branched root systems. Differences in adaptation to acid soils among Brachiaria species cannot be attributed to Al toxicity. Internal requirements of P, Ca, and K for growth of B. humidicola are much lower than those for other species. Greenhouse and field studies have demonstrated sticking responses in forage yield to P, but no response to lime nor to micronutrient applications. Rapid and reliable screening procedures are urgently needed to improve the efficiency of evaluation and genetic improvement of Brachiaria germplasm" (Rao et al. 1996).


Within three months a complete ground cover can be obtained. It has a long growing season with productive active growth from spring to late autumn.

Cropping systems

It needs to be stocked heavily. Added nitrogen is required to keep it in active leafy growth. Frequent applications of nitrogen-up to six times per season-keep the grass in a very nutritious condition and improve live weight gain especially under high rainfall conditions. Grazing at 42-day rotations gave the best balance with Brachiaria decumbens and Pueraria phaseoloides at Carimaqua, Colombia. At 56-day rotations the Brachiaria seeded and new seedlings invaded the Pueraria. At 28-day rotations the Brachiaria maintained good stands.
It is compatible with legumes such as Stylosanthes sp., Centrosema sp. and Pueraria sp. for a short time but these are soon suppressed by the vigorous grass (Brachiaria decumbens) to leave a pure grass sward. Hetero (Desmodium heterophyllum) has similar tolerance (Loch, 1978 Gutteridge and Whiteman, 1978).

Land management

It will establish in rough seeds-beds but gives better results on a well-prepared seed-bed.

Minimum tillage systems should be used. When established, Brachiaria decumbens will suppress weeds very effectively.

Propagation material

It spreads naturally from seed. Initially planted from vegetative material, it is now established mostly from seed, after dormancy has been broken.


It will establish in rough seeds-beds but gives better results on a well-prepared seed-bed. Minimum tillage systems should be used. When established, Brachiaria decumbens will suppress weeds very effectively. Within three months a complete ground cover should be obtained. It has a long growing season with productive active growth from spring to late autumn.

Seed should be sown no deeper than 1 cm and rolling is recommended after sowing. Best results are obtained when seed is sown during the wet season at 2-5 kg seed (number of seed per kg 220,000 to 225,000) per hectare.

The seed needs after-ripening for 10 to 12 months to break dormancy. Treating the seed for 10 to 15 minutes with commercial sulphuric acid will improve germination of recently harvested seed from 0-33% (Grof, 1968). Seedling vigour is excellent.

Grazing management

B. decumbens needs to be stocked heavily. Grazing at 42-day rotations gave the best balance with Brachiaria decumbens and Pueraria phaseoloides at Carimaqua, Colombia. At 56-day rotations, the Brachiaria seeded and new seedlings invaded the Pueraria. At 28-day rotations the Brachiaria also maintained good stands.

Other workers also report that "Brachiaria species show rapid regrowth and good persistence under heavy or frequent defoliation (Rika et al., 1991). Brachiaria decumbens ranked higher over 10 harvests under cutting, compared with the other species; under grazing, B. decumbens persisted longer than Panicum maximum and Digitaria setivalva (now D. milanjiana) under increasing stocking rates and heavy grazing pressure (Chen et al., 1981). As grazing pressure increased, the pasture of P. maximum was gradually invaded by Paspalum conjugatum , unlike that of B. decumbens pasture, which maintained high ground cover until it "crashed" under prolonged and heavy grazing at a stocking rate of 10 head/ha" (cited from Fisher et al. 1996).

Fertility management

Added nitrogen is required to keep B. decumbens in active leafy growth. Frequent applications of nitrogen, up to six times per season, keep the grass in a very nutritious condition and improve live weight gain especially under high rainfall conditions. However, a selection of Brachiaria decumbens yielded over 4,000 kg DM/ha without nitrogen but with adequate phosphorus at Quilichao, Colombia (CIAT, 1978) and it was one of the better grasses at low nitrogen and low phosphorus. In Honduras, mean annual production was 23,072 kg DM/ha with added nitrogen as 450 kg sulphate of ammonia per hectare (Romney, 1961). At Korovinia, in Fiji's wet zone, it yielded 34,126 kg DM/ha over an 11 month period, while in Solomon Islands it produced up to 30,000 kg DM/ha per year under coconuts. On an oxisol at Carimagua, Colombia, it gave maximum yields at 50 kg P2O5/ha, responding much more to phosphorus than Panicum maximum and Andropogon gayanus. On an ultisol at Quilichao, Colombia, it gave a linear response to more than 400 kg N/ha (CIAT,1978).

Boddey et al. (1996) reported that productivity has declined over much of the B. decumbens pastures in South America. They suggested several reasons, "a major one of which is the immobilization of plant-available nitrogen by large quantities of grass litter of very high C-to-N ratio. For the extensive beef cattle pastures of tropical America, N fertilization is usually not economical, and legumes have been introduced to provide biologically fixed N. The plant litter of mixed grass-legume swards contains more N, thus decreasing the C-to-N ratio and favouring pasture sustainability, according to studies in Brazil and Colombia".

Weed control

It tolerates pre-emergence application of atrazine at 2.5 kg of 80% product per hectare which gives good control of a wide range of weeds when establishing the grass in red latosolic soils on the Atherton Tableland, Queensland (Loch, 1978; Hawton, 1979).

Land management

The work of Boddey et al. (1996) also suggested that the "environmental impact of a vigorously growing Brachiaria pasture is only positive: the grass provides good soil cover, facilitating water infiltration and preventing erosion; leaching of soluble nutrients is minimal; N2 fixation associated with some Brachiaria genotypes may slow down pasture decline; and the grass's deep, dense rooting system also sequesters much more atmospheric C than native pasture. The large area occupied by Brachiaria pastures suggests that this may be a sink of global significance for atmospheric CO2. Sustaining the productivity of pastures while improving animal production requires strategies for maintaining or even improving soil fertility. One effective strategy is to introduce forage legumes".

Compatibility with legumes

Fisher and Kerridge (1996) report that Brachiaria species are very aggressive and there have been difficulties in them forming long-term, stable associations with legumes. Desmodium heterophyllum and D. ovalifolium were reported to be more compatible with Brachiaria than species of Centrosema and Pueraria, which, in turn, were more compatible than Stylosanthes guianensis . Of the common cultivars of Brachiaria, B. humidicola is regarded as being the most aggressive, followed by B. dictyoneura CIAT 6133, then B. decumbens. There are no data, nor even an objective measure, for this ranking. Usually, few weeds invade a Brachiaria pasture unless it is grossly mismanaged through overgrazing".

"Some legumes, grown in association with B. decumbens, have persisted over the long term, for example, at Carimagua, associations of B. decumbens-Stylosanthes capitata on a sandy soil. Brachiaria decumbens-Pueraria phaseoloides persisted for more than 10 years on a clay loam, even though, in some years, the vigor of B. decumbens was reduced by spittlebug infestation. At Pucallpa, Peru, assciations of B. decumbens-D. ovalifolium have persisted for 8 years under lenient grazing" (Fisher and Kerridge 1996).

"Research in Costa Rica confirms that A. pintoi is more persistent than other legumes in associations with Brachiaria. The proportion of A. pintoi increased with heavy grazing pressure, and fewer weeds grew in these than in the other Brachiaria-legume associations. One mechanism for greater persistence was the greater half-life of A. pintoi plants" (Fisher and Kerridge 1996).

"In summary, it is difficult to assess whether the poor persistence of legumes (other than A. pintoi) with Brachiaria species results from competition with the grass or from inherent characteristics of the legumes. Most other legumes have characteristics that do not favour persistence under heavy defoliation; that is, easily removed growing points and low seed production. Brachiaria decumbens tolerates heavier grazing than other grasses and thus tends to be grazed more heavily; this in turn puts greater pressure on the legumes. On the contrary, where growing points are not removed, as with A. pintoi, the proportion of legume can increase under heavy defoliation" (Fisher et al. 1996).

Other crop management practices

It withstands heavy grazing and trampling. CIAT (1978) recommend waiting for a month after the opening rains before establishing rainy-season stoking rates. It is affected by burning but if the environment is dry enough, Brachiaria decumbens will take a fire and recovery after fire is usually satisfactory.


A selection of Brachiaria decumbens yielded over 4 tonnes of DM/ha without nitrogen but with adequate phosphorus at Quilichao, Colombia (CIAT, 1978). It was one of the better grasses at low nitrogen and low phosphorus. In Honduras, mean annual production was 23 tonnes DM/ha with 450 kg sulphate of ammonia per hectare (Romney, 1961). At Korovinia, in Fiji's wet zone, it yielded 34 tonnes DM/ha over an 11 month period, while in Solomon Islands it produced up to 30,000 kg DM/ha per year under coconuts.

"Seed of six commercial Brachiaria species is extensively produced for pasture sowing. Production is restricted geographically and seasonally by photoperiodic flowering reactions. It also requires prior control of vegetative tiller production, and therefore a reliable dry season. The necessary conditions are most readily found at high tropical latitudes.

Seed crop management is mostly conventional. Vigorous synchronized tillering is stimulated by decapitation and use of nitrogenous fertilizer at times when rainfall, temperature, and sunshine are expected to favor unrestricted development. Ripe seed sheds readily and, coupled with imperfect synchronization of crop ripening, tends to make conventional direct harvesting inefficient and its timing critical. Where possible, seeds are let to fall and accumulate, and then recovered. Seed yields range from more than 1,000 kg/ha of pure seed to less than 100 kg/ha.

Seed quality is heavily influenced by vitality and dormancy. Vitality depends mostly on maturity of seed at harvest, being higher in accumulated fallen seed and much lower in directly severed seed. Dormancy is strongly developed in the genus and persists in most taxa at least into the season after harvest. This creates problems for germination testing and in the field use of fresh seed. Breaching the husk, most commonly by sulfuric acid, provides a partial solution." Hopkinson et al.

Genetic resources

"The genus Brachiaria, tribe Paniceae, includes about 100 species, which occur in the tropical and subtropical regions of both eastern and western hemispheres, but mostly in Africa. Seven perennial species of African origin: B. arrecta, B. brizantha, B. decumbens, B. dictyoneura, B. humidicola, B. mutica, and B. ruziziensis have been used as fodder plants, particularly in tropical America, and less so in Asia, the South Pacific, and Australia" (Keller-Grein et al. 1996).

As previously mentioned, it is difficult to separate closely related species and there is confusion regarding identification. Bogdan (1977) stated that B. ruziziensis and B. decumbens were regarded as closely related. At the Kitale Research Station, Kenya, B. ruziziensis and B. brizantha were not distinguished from each other for a long time. Maas (1996) suggested that "the widely used cv. Basilisk, commonly identified as B. decumbens (Oram, 1990), is in fact B. brizantha". Because much of the B. brizantha germplasm overlaps morphologically with accessions of B. decumbens, Maas (1996) suggested that we "continue identifying cv. Basilisk as B. decumbens, until the taxonomic status of the whole agamic complex, including B. brizantha, B. decumbens, and B. ruziziensis, is clarified".

Keller-Grein et al. (1996)reported that while "Brachiaria brizantha is widespread in tropical Africa, occurring in open and wooded grasslands, along margins of woodlands and thickets, and in upland grasslands, Brachiaria decumbens has a narrow natural distribution". "However, except for cv. Basilisk, no germplasm is available from Uganda, where the species is very common and where numerous herbarium specimens were collected. In addition, no germplasm has been collected from westem Tanzania or Zaire, where the species also occurs naturally. It is found in deciduous bushland, grasslands, and at forest edges" .

"Four major and three minor collections of Brachiaria exist ex situ, holding a total of 987 distinct accessions of 33 known species. About 40% of existing accessions are of B. brizantha, and another 39% are of B. humidicola, B. decumbens, B. nigropedata, B. jubata, or B. ruziziensis" (Keller-Grein et al. 1996).

"To minimize genetic drift, shift, or erosion, proper maintenance of germplasm is essential. But conservation of Brachiaria germplasm as seed is extremely difficult because of problems in seed production and processing. Hence, most major genetic resource centers maintain live field collections" (Keller-Grein et al. 1996).

"Because most Brachiaria species are apomicts, problems of contamination of accessions by outcrossing during rejuvenation are limited. Current practices to avoid mechanical mixtures are adequate. Sexually reproducing biotypes would be better conserved in vitro; however, no conservation method has yet been developed for this" (Keller-Grein et al. 1996).

Variability and cultivars

"Brachiaria decumbens and B. brizantha are tetraploid (2n = 4x = 36) and apomictic, that is, the embryo is produced without fusion of male and female gametes. Sexuality has been found at the diploid level in these species and in B. ruziziensis, and is generally associated with regular chromosome pairing and division. As a breeding tool, apomixis offers several advantages, because it associates fixation of hybrid vigour with seed propagation. Apomictic hybrids breed true and superior genotypes can be rapidly increased by seed." (Valle do and Savidan 1996).

"Until 10 years ago, genetic improvement of Brachiaria species depended entirely on selection among naturally existing genotypes, genetic recombination being impossible because of prevailing apomictic reproduction. However, recent advances in understanding the genetics and cytogenetics of Brachiaria, have opened the way for controlled genetic manipulation. Genetic recombination is now possible in the B. ruziziensis/B. brizantha/B. decumbens pool. Present breeding initiatives aim to create apomictic genotypes combining spittlebug resistance and edaphic adaptation with other desirable attributes" (Miles and Valle do 1996).

"Probably the best known and most widely used Brachiaria cultivar is B. decumbens cv. Basilisk (signal grass). It derives from seed (CPI 1694) introduced into Australia from the Ugandan Department of Agculture in 1930. It was approved for commercial release in Australia in 1966 and registered in 1973 (Oram, 1990). This cultivar is well adapted to infertile acid soils, and forms an aggressive, high-yielding sward that withstands heavy grazing and trampling. It is a palatable grass of good forage quality and gives good animal performance. However, its susceptibility to spittlebugs reduces its value as a pasture plant in areas where this pest is a major constraint, for example, in the neotropical savannas" (cited from Keller-Grein et al. 1996).

"Brachiaria brizantha cv. Marandu (IRI 822; BRA-000591), released in 1984 in Brazil by EMBRAPA, originates from germplasm introduced to the Ibirarema region, Sao Paulo, Brazil, from the Zimbabwe Grasslands Research Station, Marandella (now Marondera) (Nunes et al., 1984). Its antibiotic resistance to spittlebugs (Ferrufino and Lapointe, 1989) has led to the rapid adoption of cv. Marandu throughout the American tropics. This grass provides a palatable forage of nutritional quality similar to that of B. decumbens cv. Basilisk. However, cv. Marandu does not tolerate poor soil drainage and requires higher soil fertility than cv. Basilisk; thus, it does not persist on the low-fertility Ultisols and Oxisols that are widespread in tropical America. In mixed pastures, this cultivar seems to exert an allelophathic effect on several legume species (Rodrigues and Reis, 1994)'' (cited from Keller-Grein et al. 1996).

"Improvement of protein concentration in Brachiaria seems feasible, as indicated by the variation in CP among accessions of some important Brachiaria species grown at different sites. However, based on results with other grass species, GxE variance for CP in Brachiaria is likely to be greater than genetic variance (Vogel and Sleper, 1994). But, the large variation in IVDMD and the stability of this attribute across environments suggest that digestibility of Brachiaria species can be improved. Even small changes in digestibility in B. decumbens or B. brizantha could have a significant impact on animal performance, as was shown with bermuda grass (Hill et al., 1993)" (cited from Lascano and Euclides 1996).

"Advanced biotechniques offer the potential for greater precision and efficiency in genetic manipulation of crop species. Applications of molecular markers in Brachiaria include assessment of genetic variation, gene tagging, and genetic mapping. In particular, recent applications of molecular-marking techniques to apomixis provided strong evidence of single-gene inheritance of reproductive mode in Brachiaria. Regeneration protocols for Brachiaria have been worked out and plants of four Brachiaria species have been regenerated. These tools provide a basis for a project aimed at fine-mapping the apomiids gene, and its possible cloning and transfer to other crop species" (Tohme et. 1996).

Improved evaluation and selection methodologies for nearly all traits of agronomic importance, particularly spittlebug resistance and edaphic adaptation are needed.


The seed is harvested with an all-crop harvested. All seed recovered directly from the crop must be dried after harvest because moisture contents can be as high as 60% and the usual target for storage is about 10%. The minimum germination and quality required for commercial sale are 15% germinable seeds with 50% purity.

Nutritional Quality and Animal Production

The quality of B. decumbens (signal grass) "has been measured in many cutting experiments and feeding trials, together with other well-known grass species. Results summarized show that the in vitro (IVDMD) and in vivo digestibility of immature (leaf) and mature (whole plant) signal grass is as high as or higher than that of other tropical grasses, such as Panicum maximum . Values of IVDMD in signal grass have ranged from 60% to 70% in immature forage, and from 50% to 60% in mature forage - higher than the average (55%) for tropical forage grasses found by Minson (1990) from a review of the world literature" (cited from Lascano and Euclides 1996).

Lascano and Euclides (1996) summarize results from experiments designed to measure liveweight gain (LWG) of steers grazing signal grass under different types of management. In the humid environment of South Johnstone, Australia. Live weight gain/ha/year on unfertilized signal grass was 50% higher than that recorded earlier at the same location with unfertilized common P. maximum (Grof and Harding, 1970), whereas on fertilized signal grass, LWG was almost 1000 kg/ha (Harding and Grof, 1978). In a drier environment (Cape York Peninsula, Australia), signal grass in association with legumes, produced more LWG in heavily stocked pastures than a common P. maximum association (Winter et al., 1977). In southeastern Queensland, Australia, signal grass fertilized with N and the improved P. maximum cv. Hamil produced similar LWG at both medium and high stocking rates (Whiteman et al., 1985). However, in Campo Grande, Brazil, animal gains on unfertilized signal grass were 20% to 30% lower than on improved P. maximum cultivars" (cited from Lascano and Euclides 1996).

A long-term grazing experiment is being conducted on an Oxisol on the Colombian Llanos with signal grass fertilized every 2 years (10 P, 13 K 10 Mg, and 16 S kg/ha), under continuous grazing (1979 to 1994), and with seasonal adjustment of stocking rate (1 h/ha in the dry season and 2 hd/ha in the wet season). "During the first 9 years, the average seasonal LWG was 125 kg/hd and 225 kg/ha, with minimum values of 48 kg/hd and 86 kg/ha in 1986 and maximum of 182 kg/hd and 328 kg/ha in 1981 (Lascano and Estrada, 1989). The low LWG in 1986 coincided with a season of unusually heavy rains (over 3,000 mm) and a severe spittlebug attack on the grass. Maximum LWG was recorded in a year with abnormal dry season rainfall (300 mm)" (cited from Lascano and Euclides 1996).

"An analysis of the 16-year LWG data indicates that this pasture does not show signs of degradation, despite periodic spittlebug attacks. The annual LWG recorded in 1994 (140 kg/hd) is similar to that recorded in the first year after establishment. However, this should not be interpreted to mean that spittlebug is not a problem in B. decumbens. Results clearly indicate the large losses in LWG caused in a given year by spittlebug. Nevertheless, signal grass has been persistent in this long-term grazing experiment, probably because of careful management, that is, seasonal stocking rate adjustment and maintenance fertilizer, which is not practised for most commercial pastures" (cited from Lascano and Euclides 1996). Overgrazing and lack of fertilization coupled with spittlebug infestation are probably responsible for the degradation and consequent low animal production of large areas of the Cerrados".

Liveweight gains on signal grass increase with improved management. "Results from Mato Grosso do Sul, Brazil, showed that LWG on signal grass could be doubled (from 120 to 250 kg/ha/year) with P fertilization (44 kg/ha) (Schunke et al., 1991). On less fertile soils in the Cerrados, LWG on signal grass increased only by 60 kg/ha per year when the amount of P applied to the pasture was doubled (29 to 58 kg/ha) but animal performance was probably limited by low N levels in the soil" (cited from Lascano and Euclides 1996).

One effective way to increase the quality of Brachiaria pastures is through grass-legume associations. "However, because of the inherent aggressiveness of most Brachiaria species, few associated legumes have persisted. Results from a short-term grazing study in Campo Grande, Brazil, showed an 18% advantage in LWG from the association of signal grass with the legume Calopogonium mucunoides. In a long-term grazing trial in the Colombian Llanos, signal grass associated with Pueraria phaseoloides (kudzu) produced, on average, 40% more LWG than the grass alone over the first 9 years of grazing (Lascano and Estrada, 1989). Over a longer period (16 years), the signal grass-kudzu pasture has produced an average of 34% more LWG than the grass alone (C. E. Lascano, unpublished data). The grass-legume pasture has shown a clear advantage over pure grass, giving a 67% higher LWG in in the dry season and a 24% higher LWG in the wet season, indicating both a direct (i.e., animal selection) and an indirect (i.e., N cycling) contribution of the legume to animal performance (Lascano and Estrada, 1989)" (cited from Lascano and Euclides 1996).

"The excellent compatibility and persistence of A. pintoi cv. Amarillo with Brachiaria was documented in the Colombian Llanos by Grof (1985). Following these initial results, grazing experiments were established in the Llanos and in the humid tropics of Costa Rica to measure LWG on Brachiaria with and without A. pintoi. Associations gave substantially higher LWG than the pure grass pastures: 67% higher for B. humidicola, and 52% higher for B. dictyoneura. In the more favourable environment of a humid forest site at Guapiles, Costa Rica, LWG on heavily stocked pastures of B. brizantha cv. Marandu/A. pintoi has been exceptionally high, producing 30% more LWG over a 3-year period than the pure grass. At this site, the pure grass pasture is showing signs of N deficiency (R. J. Argel, 1994, personal communication), which, in the long term, will undoubtedly result in greater advantage in LWG because of the legume.

Another legume compatible with Brachiaria is Desmodium ovalifolium (now D. heterocarpon subsp. ovalifolium). In a humid environment of Bahia, Brazil, pastures of B. humidicola and D. ovalifolium were relatively stable over time with adequate grazing management (Pereira et al., 1992b). However, because of the low quality of this legume (i.e., tannins), LWG on the grass-legume association did not differ from that on the pure grass (Pereira et al. 1992a)" (cited from Lascano and Euclides 1996).

"Few studies have measured milk yields of cows grazing Brachiaria pastures with or without legumes. In short-term grazing experiments at CIAT's experiment station at Santander de Quilichao (Cauca, Colombia), daily milk yield was greater on signal grass (8 kg/cow) than on B. dictyoneura (6 kg/cow) (C. E. Laseano, unpublished data)" (cited from Lascano and Euclides 1996).


"A widespread, but sporadic, toxicity syndrome associated with B. decumbens is hepatogenous photosensitization, which can cause losses in LWG of up to 40% in severe cases (Fagliari et al., 1991). Toxicity symptoms, mainly in young sheep, goats, and cattle, include skin lesions, facial edema, liver damage, ruminal stasis, neurological disorders, and even death if animals are not removed from the pasture (Abas-Mazni et al., 1983; Salam Abdullah et al., 1988; 1989). In most reports from Africa, Southeast Asia, or South America, photosensitization has occurred in animals on signal grass. However, toxicity has also occurred in sheep grazing mainly B. brizantha cv. Marandu (Magalhaes et al., 1988) and in horses grazing B. humidicola (Schenk et al., 1991)" (cited from Lascano and Euclides 1996).

"Photosensitization with signal grass has been related to infestation of the grass by the saprophytic fungus P. chatarum, which produces spores thought to contain toxic sporidesmin (Andrade et al., 1978; Dobereiner et al., 1976; Nobre and Andrade, 1976). However, the cause-effect relationship between P. chatarum and photosensitization in signal grass has been challenged by researchers from Malaysia (Abas-Mazni and Sharif, 1986; Salam Abdullah et al., 1992) and New Zealand (Smith and Miles, 1993). The arguments put forward against the theory that P. chatarum alone is responsible for photosensitization in signal grass are:

1) In South America, different strains of P. chatarum were isolated from Brachiaria pastures where cattle had exhibited toxicity, but no strain had produced sporidesmin (Brewer et al., 1989).

2) In New Zealand, a high correlation between spore count in the pasture and toxicity (facial eczema) exists in sheep (Brook, 1969), but this does not appear to be the case in South America (E. Aycardi, 1994, personal communication).

3) In Malaysia, sheep developed photosensitivity when fed hay or fresh signal grass. However, liver damage was not observed when grass litter, ideal for fungal growth, was fed (Abas-Mazni and Sharif, 1988).

4) The pathology of animals poisoned with signal grass is similar to that found in animals grazing Panicum species not infected with P. chatarum (Graydon et al., 1991).

5) Steroidal saponins were isolated from the rumen contents of poisoned sheep fed signal grass (Lajis et al. 1993; Salam Abdullah et al. 1992). In addition, steroidal saponins were identified in several plants known to cause photosensitization (Miles et al. 1993), and the toxicity symptoms were reproduced in sheep through oral dosing of crude saponins extracted from the plant Tribulus terrestris (Zygophyllaceae) (Kellerman et al. 1991)" (cited from Lascano and Euclides 1996)

"Clearly, more study is needed on the role of steroidal saponins in hepatogenous photosensitization in livestock and their possible interaction with P. chatarum and other fungi (endophytes) that may be present in B. decumbens and other Brachiaria species. Even if photosensitization in ruminants and horses is not caused primarily by P. chatarum, the presence of fungal spores may exacerbate the toxicity (Smith and Miles, 1993)" (cited from Lascano and Euclides 1996).


Links for the genus:


Keller-Grein et al. (1996); Ndikumana and Leeuw de (1996); Pizarro et al. (1996) ; Stur et al. (1996); Renvoize et al. (1996); Clayton and Renvoize, (1986) ; Fisher et al. (1996); Kaligis and Sumolang, (1991); Hopkinson et al. (1996); Rao et al. (1996); Loch, (1978); Gutteridge and Whiteman, (1978);Rika et al. (1991); Valle do and Savidan (1996) ; Miles and Valle do (1996); Lascano and Euclides (1996); Hawton, 1979; Boddey et al. (1996); Romney, (1961); Bogdan (1977) Maas (1996); Tohme et al. (1996)