RESEARCH WORK

Agronomic performance of seven Brachiaria humidicola accessions during the establishment stage

¹Instituto Universitario de la Paz, Calle 49 # 10-22, Avenida Santander, Barrancabermeja, Santander del Sur, Colombia
E-mail: emirocanchilas@yahoo.es
²EEPF "Indio Hatuey". Matanzas, Cuba
³CORPOICA, Colombia

ABSTRACT

The objective of this work was to evaluate the agronomic performance of seven Brachiaria humidicola accessions during the establishment stage, in order to select those with better adaptation to the conditions of acid soil and low fertility of the Santander region of Barrancabermeja and Magdalena Medio, Colombia. Twenty-one plots were used, with a completely randomized design and three replications for each accession. The evaluations were made since the seventh day after the seeds were planted and the plots were considered established when the covered area reached 80%. The emergence, cover, height, lateral displacement, dry forage production, incidence of pests and diseases and nutritional deficiencies of the plants were determined; the bromatological analysis was also made. The B. humidicola accessions had an average germination of 30%. The cumulative variance was 88,67% in the first three components. The total variability was concluded to be high, with regards to the measured and/or estimated indicators, which allowed the adequate differentiation of the accessions. The indicators of higher contribution in group formation and selection were: cover, dry matter, dry forage production, fat percentage, leafiness, energy and protein.

Key words: Acid soil, Brachiaria humidicola, evaluation.

INTRODUCTION

The pastures of tropical America are established, mostly, on soils characterized by their acidity and low fertility. The inadequate management of pastures and, in many cases, the lack of utilization of the germplasm adapted to these conditions, are the main conditioning factors of low productivity in those ecosystems (Nieuwenhyse, 2010).

During the last years grasses and legumes that adapt to such conditions have been evaluated and it was found that some species of Brachiaria showed a high productive potential in these ecosystems, due to their adaptation capacity to infertile soils and the efficient use of nutrients (Velásquez and Muñoz, 2006).

Among the most outstanding ones is Brachiaria humidicola, which shows good performance due to its aggressiveness, regrowth rate, tolerance to draught, resistance to pests and diseases, as well as adaptation to different soil types (Chacón, 2005).

As part of the research projects developed by the Colombian Corporation of Agricultural and Livestock Research (CORPOICA) and the Instituto Universitario de la Paz (UNIPAZ), for the enhancement of livestock production, this study was conducted in order to evaluate the agronomic performance of seven B. humidicola accessions during the establishment stage, and select those that adapt better to the conditions of acid and low fertility soil of the Santander region of Barrancabermeja and Magdalena Medio, Colombia.

MATERIALS AND METHODS

Site description. The work was conducted in the facilities of the Experimental Center Santa Lucía, property of the Instituto Universitario de la Paz, which is located in Vereda Zarzal. Barrancabermeja municipality, in the Santander region of Magdalena Medio, Santander del Sur Department, Colombia.

This Institution is located at kilometer 14 on the left margin of the Barrancabermeja-Bucaramanga road; it has an extension of 324 ha and it is located at 73º51'50" longitude west and 7º3'48"latitude north.

Its plant formation is tropical rainforest, of sloped and undulated topography, with average annual rainfall of 2 800 mm, mean temperature of 29ºC and relative humidity of 80%.

The soil is classified as Oxisol, extremely acid, with high aluminum and iron content, and loamy-clayey texture (CORPOICA, 1995).

Treatments and experimental design. Seven B. humidicola accessions planted in randomly distributed plots were evaluated, using a completely randomized design and three replications (plots) for each treatment (tabla 1).

The material for the evaluation was brought from the International Center of Tropical Agriculture (CIAT), by the Colombian Research Corporation, as part of the research agreement with the Instituto Universitario de la Paz.

Experimental procedure

Soil preparation and plot establishment. In order to eliminate the plant tapestry of native grasses and weeds present in the area where the research was conducted, a conventional preparation was made, with one pass of plough and two of light harrow. Between each labor 20 days were waited to control the regrowth. The last harrowing also allowed leveling the terrain.

A lot of 500 m2 (10 m wide and 50 m long) was used, in which 21 experimental plots were delimited, each 7 m long and 3 m wide, with an individual area of 21 m2; the separation between plots was 1 m. It was adopted as criterion considering the plots established when the covered area reached 80%, approximately after 110 days, moment in which a homogenization cutting was performed.

Planting. Planting was considered part of the evaluative process of the different accessions. For that purpose sexual seeds were used, which were put to germinate in a nursery, in plastic pots of 30 cm diameter by 25 cm high. When germination was verified and a sufficient number of seedlings were obtained, transplant was made in the research site. The planting distance was 0,5 m between plants and 1 m between rows.

Fertilization. The fertilizer application was based on the results of the soil chemical analysis (table 2) and on recommendations of the RIEPT (International Network of Tropical Pasture Evaluation). The following doses were used: phosphorus (P2O5- 40 kg/ha), potassium (K2O- 40 kg/ha), magnesium (Mg- 10 kg/ha) sulphur (S- 10 kg/ha) and nitrogen (N- 10 kg/ha). For the soil analysis the samples were taken at 25 cm of depth and were sent to the Laboratory of Soils of the International Center for Tropical Agriculture (CIAT).

Sample taking in the field. The field sample was taken from the center of each plot, after eliminating the edge effect. The cuttings were made within a 50 x 50 cm frame and 10 cm over the soil level. Six cuttings were performed every 42 days, at an equal time for all accessions.

Experimental measurements. Every week the emergence (%), cover (%), height (cm), lateral displacement (cm) dry forage production (t/ha/cutting), incidence of pests and diseases and nutritional deficiencies of plants, were evaluated. Bromatological analysis was also made.

Emergence (%). The germination percentage was determined from the emerged seedlings after a period of 40 days, moment in which this measurement was considered finished.

Cover. In order to determine the cover reached by the accessions in the plots after being transplanted, the following value scale was used: 1= 10-20% of area covered (very little covered); 2= 21-40% of area covered (little covered); 3= 41-60% of area covered (partially covered); 4= 61-80% of area covered (covered); 5= >80% of area covered (well covered).

Height. It was determined in four spots per plot, with a metric tape graduated in centimeters, from the soil, in perpendicular position.

Lateral displacement. Fifty percent of the plants were evaluated, which were randomly selected as a representative sample of the plot. The total stolons of each plant were measured using a metric tape, from the stem basis to the stolon apex.

Leaf percentage. In order to determine the leaf percentage, a subsample of 300 g was taken from the total green mass and the leaves were separated from the stems in fresh. It was placed in a ventilated stove at 80ºC until constant weight, to determine which percentage of the total dry weight belonged to each fraction.

Dry forage production. The production of each accession in green was expressed in tons of dry matter per cutting per hectare (DM/cutting/ha); for that purpose the weight of the green sample in a square meter was taken and this value was taken to its equivalence in DM production per hectare.

Bromatological analysis. From the green material cut to determine the dry forage production a subsample of 500 g was taken, which was sent to the laboratory for the analysis of DM, CP, fat, energy, non-nitrogen extract by the method of the AOAC (1990). Acid detergent fiber was also determined through the method proposed by Van Soest et al. (1991).

Disease incidence. The following scale was used: 0= 0-1% of the area affected (immune); 1= 2-10% of the area affected (resistant); 2= 11-20% of the area affected (tolerant); 3= >20% of the area affected (susceptible).

Incidence of potentially-pest insects. The plots were divided into four grids and in each one the affected plants were recorded. The affectation percentage was determined based on the average obtained in each of the grids and the same scale as in the incidence of diseases was used.

Mineral deficiency. It was determined by visual observation of the plants, identifying the main manifestations by nitrogen, phosphorus and potassium deficiencies, according to the methodology proposed by the RIEPT.

Statistical analysis

The results were processed through the principal component analysis (PCA) (Morrison, 1967), in which those main components that showed proper values higher than 1 and sum or preponderance factors higher than 0,60 were taken as criterion.

The cluster analysis was applied for the grouping and selection of the accessions, using as the Euclidian distance as criterion, from the results obtained in the PCA (Torres et al., 2006), and the stadigraphs mean and standard deviation were determined for the analyzed variables. Thus, groups of species were available that allowed making a simpler and more objective analysis of their performance.

For the case of the indicators potentially-pest insects and diseases the inverse value was taken, because the means higher than the population mean are equivalent to a higher affectation.

RESULTS AND DISCUSSION

The B. humidicola accessions had an average germination of 30%, with values of 53, 51, 32, 41, 10, 25 and 11% for CIAT-16866, 16888, 16867, 16886, 26159, 26427 and 16871 respectively.

Nevertheless, the accessions CIAT-26159 and 16871 showed higher cover, of 100%, 56 days after transplant, due to their stoloniferous growth (fig. 1). The ones with lower cover were CIAT-16886 and 16888, due to their turf growth. One hundred and four days after being transplanted, most of the accessions reached a total cover and only 16888 had 60%.

B. humidicola is characterized by its invading power, especially if it is established in areas with high rainfall. The displacement of the seven materials was higher than 80 cm in the first 40 days after planting. Regarding height (fig. 2) B. humidicola CIAT-16871 was the lowest, followed by CIAT-16866 and CIAT-16886. B. humidicola CIAT-26159 and CIAT-26427 were the highest.

Table 2 shows the results of the principal component analysis (PCA). The total variability was proven to be high, regarding the evaluated indicators, during this period. A cumulative variance of 88,67% was detected in the first three components. The variables that better explained the variance in the first component (31,24%) were: cover, dry matter, dry forage production, and fat percentage, which were positively related among themselves; the second component extracted a variance of 29,49%, which was explained by leafiness, energy and protein; while the third component extracted 27,37% of variability, explained by non-nitrogen extract and acid detergent fiber (the latter negatively).

The variability showed through the indicators can be due to the high relationship that existed among them in this stage, aspect which allows the grouping of the accessions and their later selection. The high percentage of the variance explained by the three components, suggests that it contains indicators that can discriminate well the accessions under study. For such reason, it could be stated that variability was related to the contrast in the performance of these accessions, and not to the growth habit which characterizes them.

Canchila et al. (2008 a; 2008 b) obtained similar results when evaluating a collection of 22 Brachiaria spp. accessions on an extremely acid Oxisol soil, with high aluminum and iron content, and loamy-clayey texture, in which three B. humidicola accessions were included and some of the indicators evaluated in this study were used.

These results coincide with the ones reported by Olivera and Machado (2004) when studying a collection of 20 Brachiaria spp. accessions on low soils of moderate fertility, where they found a variance of 81,6%. This indicates that independently from the edaphoclimatic conditions, the populations and particularly the accessions of the species of this genus, can express a marked variation among individuals for some indicators and group regarding those variables, which represents a positive element in the evaluation and selection work (Machado, 2006).

In this study it could be observed that variability was well distributed, because the proper value was higher than one, acceptable index for the variability corresponding to each indicator to be better related with each axis in correspondence with this type of analysis (Di Rienzo et al., 2001). For such reason these indicators as a whole were included in the cluster analysis and thus the differentiation or similarity among these accessions could be determined.

In correspondence with the high value of the cumulative variance, the phenotypical variability was assumed to be sufficiently propitious to make the cluster analysis based on the results of the PCA, which allowed the formation of four groups. The accessions belonging to each one of them are shown in table 3, as well as the mean and standard deviation of each of the groups formed.

According to the results it can be deduced that the accessions B. humidicola CIAT-16867, CIAT-16888 and CIAT-168886, belonging to group I were the plants with higher protein and dry matter content, higher dry forage production, and leafiness values.

The accession B. humidicola CIAT-16866, belonging to group IV, showed the highest values regarding: fat percentages, non-nitrogen extract and energy; while CIAT-26159 and CIAT-26427, from group II, had a good value of cover and acid detergent fiber.

On the other hand, B. humidicola CIAT-16871, belonging to group III, was characterized by having the lowest values in some of the evaluated indicators, as compared to the other accessions (cover, fat percentage, non-nitrogen extract, dry matter and dry forage production).

In this evaluation a remarkable superiority was observed in the variability of the accessions, which could have been influenced by the prevailing climate conditions in the period of highest rainfall (in which the study was conducted). In such environment, according to Del Pozo (2000), grasses can express a better response, particularly when temperatures are between 22 and 35ºC (which are considered optimum for their growth), and they also make a better utilization of sunlight with the predominance of long days.

In this sense, it should be mentioned that according to the studies conducted by Rao et al. (1998) and Rao et al. (2006) this performance could be related to: the capacity of these plants to maintain root growth at the expense of the aerial part growth; the use of both nitrogen forms (nitrate as well as ammonium); the attainment of nitrogen through associative fixation and P by long root systems and the association with vesicular-arbuscular mycorrhizas; in addition to Ca through widely ramified roots with a large number of root apices. It is valid to state that independently from the fact that Brachiaria species have much lower internal requirements (especially of P and Ca) than other grasses such as Panicum maximum, they show certain interspecific differences, which could be observed in this evaluation.

Regarding the incidence of potentially-pest insects and diseases, there was no representativeness of the lesions caused by them in the sampled plants, which would allow their quantification and later analysis. The same occurred with mineral deficiency; yet, the recommendation made by RIEPT is maintained, in order to prevent any affectation in this sense.

CONCLUSIONS

The accessions B. humidicola CIAT-26159 and 16871 were the ones with the best performance regarding cover percentage.

The total variability was high, with regards to the measured and/or estimated indicators, which allowed the adequate differentiation of the accessions.

The indicators with higher contribution in group formation and selection were: cover, dry matter, dry forage production, fat percentage, leafiness, energy and protein.

The accessions which stood out for showing higher values than the population mean were: B. humidicola CIAT-16-867, CIAT-16888 and CIAT-16886.