RESEARCH WORK

 

 

 

Morphoagronomic characterization and protein content evaluation in two genotypes of Clitoria ternatea L. cultivated on a trellis system

 

 

 

Hallely Suárez1, W. Mercado1, Maribel Ramírez1, Belkys Bracho2, J. Rivero3 y D. E. García4,5

1Departamento de Botánica
2Departamento de Estadística
3Departamento de Zootecnia,Facultad de Agronomía, Universidad del Zulia (LUZ), Apdo. 15205, ZU4005, Venezuela
4FMF-Freiburg Materials Research Center, D-79104, Freiburg i. Br., Germany
5Institute of Forest Utilization and Work Science, D-79085, Freiburg i. Br., Germany
E-mail: hsuarez@fa.luz.edu.ve, mcramire@fa.luz.edu.ve

 

 

 


ABSTRACT

The objective of this work was to evaluate the morphoagronomic characteristics and the protein percentage of two genotypes: blue (BG) and white (WG), of the forage legume butterfly pea (Clitoria ternatea), cultivated on a trellis system, in Venezuela. Thirty-eight morphological and growth characteristics were evaluated, with a completely randomized experimental design. At 15 days of age the plants showed equal main stem length, number of nodes (Nn), number of leaves in the main stem (NLMS) and number of leaves in the plant, indicators which increased with age. Nn had a similar performance as NLMS and the vegetative stage lasted 30 days, when the first-order lateral branches developed and after 45 days the second-order branches developed. Pentafoliolate leaf shooting indicated the onset of flowering. Both genotypes showed a similar performance in the development stages: vegetative, flowering and production. BG showed higher vigor, high branching capacity and maximum values in leaf width and flower length, and WG had high pod and seed production; these characteristics constitute important morphological indicators in both genotypes. BG showed a higher content of CP (32,05%) at 90 days of age. To use this herbaceous legume as short-term protein source, because of its fast growth and high dry matter (89,48-90,35%) and CP (32,05-29,50%) contents is recommended; as well as to conduct other studies related to the management of this crop.

Key words: agronomic characteristics, Clitoria ternatea, protein content.


 

 

INTRODUCTION

Cattle production is one of the sectors of high socioeconomic importance in Venezuela, because it generates income, jobs and food for the population. Most milk and beef production in the country is from double-purpose cows, fed forage and balanced supplements (concentrate feeds), which are generally elaborated with imported and highly costly raw materials. The quantity of supplement to be used depends on the availability and quality of fresh forage, which varies with rainfall distribution (Villalobos and Tobías, 2008).

Tropical zones have the highest genetic diversity worldwide; however, the animal feeding models have been mainly based on the use of few plant species (Rosales, 1999). Hence the feeding value of the species has been identified for only a limited number, for which the evaluation of promising local plant genetic resources is essential. One of them is the legume Clitoria ternatea L. (butterfly pea), which stands out for its excellent adaptation to tropical and subtropical conditions (Polo, 2004; Villanueva, Bonilla, Rubio and Bustamante, 2004; López et al., 2011), good forage production, huge potential for improving animal productivity with lower cost (González and Chow, 2008) and high crude protein content (Ramos et al., 2008). The incorporation of this legume in agricultural and livestock production systems is an excellent choice, which would contribute to increase sustainability (Rojas, Olivares, Jiménez and Hernández, 2005; Faría and González, 2008; Hurtado, Vega, Ramos and Álvarez, 2008; Bugarín et al., 2009; Morris, 2009).

The morphological characterization and agronomic evaluation of plant genetic resources are important activities, because they allow describing and differentiating qualitative and quantitative attributes of several individuals of one species, based on their usefulness, as well as generating basic technical knowledge for management as crop in certain agroecological zones (Ramos et al., 2008). The use of butterfly pea in Venezuelan production systems is scarce, due to the little seed availability and the lack of technical knowledge about its cultivation and utilization. For such reasons this work was conducted, in order to evaluate the morphoagronomic characteristics and protein content of two genotypes cultivated on a trellis system, under highly dry tropical forest conditions.

 

MATERIALS AND METHODS

Study location. The trial was conducted at the university nursery of the School of Agronomy, University of Zulia (LUZ), Zulia state, Venezuela, which is located at 10º41'12" latitude North and 71º38'05" longitude West, at an altitude of 25 masl; it is framed in an ecological zone of very dry tropical forest, with rainfall of 500-600 mm/year and annual average temperature of 29ºC, relative humidity of 79% and evapotranspiration of 2 500 mm (Sánchez and Ramírez, 2006). The environmental conditions during the essay are indicated in table 1.

Plant material and seeding. The plant material was obtained from butterfly pea seeds, blue and white genotypes, which remained stored at 10ºC during a year at the laboratory of plant propagation, of the School of Agronomy-LUZ. For obtaining the seedlings, the seeds were planted in black polyethylene bags of 1 kg capacity, which contained a substratum of sand and organic matter (leached cattle manure) in a 2:1 ratio, previously treated with 5 mL of Ridomil® fungicide (3 g.L-1) per bag to disinfect the substratum. During this stage irrigation took place every two days; but seven days before the transplant the field was not irrigated, in order to guarantee seedling hardening. After 30 days they were transplanted to the field on a fine loamy Typic haplargids soil, of low natural fertility, low moisture retention and pH which varies from 6,26 to 4,95 (Larreal, Jiménez, Wilhelmus and Noguera, 2004).

Seedling establishment in the field was carried out on a vertical trellis system, of thirteen rows, 5 m long and 1 m tall. In each row two wire lines, caliber 18, were placed, at a distance of 40 and 80 cm from the soil. Wood stakes separated at 2,5 m were used as supports. Holes were made in the rows, in which 1 kg of sand and cattle manure was placed in the same proportions; afterwards, ten seedlings were planted at a distance of 0,50 m, for a total of 65 seedlings of each genotype. After the transplant the following practices took place: irrigation and manual weeding every three days, sanitary inspections every 15 days and fertilization with cattle manure (1 kg per plant) once a month. A cover of dry plant material was also placed between rows in order to decrease weed incidence, and to protect and maintain soil moisture.

Variables used. Fifteen days after transplant morphological evaluations were conducted in the plants; for such purpose 12 plants of each genotype were randomly selected, without considering those at the end of each row. The following quantitative and qualitative variables were evaluated, including some of those indicated by IPGRI descriptors (1984): main stem length (MSL), length of first-order branch (LB1), length of second-order branch (LB2), length and width of trifoliolate leaf (L3FL, W3FL), length and width of pentafoliolate leaf (L5FL, W5FL), length and width of the heptafoliolate leaf (L7FL, W7FL) of the fully developed or expanded leaves-, pod length and width (PL, PW), seed length and width (SL, SW), leading shoot length (LSL) and length of the internode previous to the leading shoot (IPLS). The length was measured from the proximal part to the apical portion of the organ, and for width (W3FL, W5FL, W7FL and PW) the equatorial plane of the organ was considered and five were measured per plant. In MSL, LB1 and LB2 a wool piece was placed near the apex, indicating the last measurement. All these variables were expressed in centimeters.

The following variables were also evaluated: number of leaves in the main stem (NLMS), number of leaves in the plant (NLP), number of leaves in first-order branch (NLB1) and in second-order branch (NLB2), number of nodes in the main stem (Nn), number of nodes to the first flower (NNFF), number of nodes to the first leaf with five and seven folioles (NNL5F, NNL7F), number of nodes before the presence of the leading shoot (NNBLS), number of nodes in the leading shoots (NNLS) as well as the number of buds and flowers per plant (NB, NF), by counting each of the organs. The number of pods per plant (NPP) and the number of seeds per pod (NSP) were counted 90 days after planting. The main stem diameter (MSD), and the flower diameter and length (FD, FL) were measured with a millimetric calibrator. The FL was determined through polar diameter; MSD and FD were determined through the equatorial diameter of the organ.

The percentage of plants with flowers (PPF) and flower buds (FB) was determined through the relation of the number of plants with flowers and of floral buds at 30 days of age between the total number of plants, respectively. The flower color (FC) was established according to the Pantone® table (2004). The weight of a thousand seeds (WTS) was evaluated 90 days after seeding, for such purpose a thousand seeds were harvested from dry pods and they were weighed. The percentage of dry matter (DM) and crude protein (CP) was determined at 90 days of age, according to the AOAC (1995), in six plants.

Experimental design and statistical analysis. The design was completely randomized; a plant was taken as experimental unit and there were twelve repetitions per genotype, value higher than the one used by Romero (2012). In the case of the variables MSL, MSD, NNMS, NLMS, NLP, NBP, NFP, LB1, LB2, NLB1 and NLB2 an analysis of measures repeated in time was applied, considering its linear and quadratic trend; the studied factor was genotype, and first and second-order polynomial models were used. In general, the model with best adjustment to the data was selected according to the Bayesian information criterion (BIC), the Akaike information criterion (AIC) corrected for small samples (AICC)- and the REMLlog likelihood criterion (REMLlogL), generated by the procedure MIXED from SAS® (SAS, 2010). To explain the performance of L3FL, L5FL, L7FL, W3FL, W5FL, W7FL, NNFF, NN, NN5FL, NN7FL, FD, FL, NPP, PL, PW, SW, SL, NSP, LNAG, NNBLS, LSL, IPLS, NNLS, DMP, CP and DM a variance analysis was made through the GLM procedure from the Statistical Analysis System program. When significant differences were found, Tukey's test was used for mean comparison (SAS, 2010).

 

RESULTS AND DISCUSSIONS

The MSL (fig. 1A) was similar in both genotypes after 15 days, and at the moment of transplant the white genotype showed the highest MSL, until the end of the evaluation stage. The MSDs in the two genotypes were similar at 45 days of age of the plant; nevertheless, since day 60 the diameter increased until day 75, especially in the blue genotype (fig. 1B). As the genotypes advanced in age stem growth increased, performance also indicated by Piña and Bautista (2006). The NLMS was similar in the genotypes during the evaluated time intervals (fig. 1C). This variable had a fast growth until day 45, stage in which the plants showed their highest potential, with the highest values in the blue genotype. Since that moment the increases were lower and this could be associated to the fact that during this time the plant showed emergence of lateral buds and formation of new branches, which produced variations in biomass distribution among its different organs. Regarding Nn (fig. 1D), a similar performance as that of NLMS was detected, although the Nn was higher in the white genotype after 30 days of age and similar to the number of leaves, because one leaf per node was always observed (figs. 1C-D).

In the NLP, during the first 30 days, the plants were found to show a slow growth phase (fig. 2A), associated to the little existence of meristematic centers or spots (Raven, Evert and Eichhorn, 2005; Barceló, Rodrigo, Sabater and Sánchez, 2005; Taiz and Zeiger, 2006). Afterwards, they started the formation of lateral branches, which increased NLP, leaf area and aerial part, in which metabolic processes occur at higher rate and carbohydrates are destined to the formation of leaves and other vegetative organs (Medina et al., 1996; Azcón and Talón, 2008). The number of leaves was higher in the blue genotype due to its high branching capacity.

The vegetative stage lasted approximately 30 days and 88,46% of the blue plants and 53,84% of the white ones were found to show flower buds; 53,84%, blue flowers and 15,38%, white flowers. These results contrasted with the ones reported by Villanueva et al. (2004), who indicated that flowering started at 45 days of age. The white genotype surpassed the blue one regarding the number of flower buds and number of flowers per plant (figs. 2B and 2C). The butterfly-pea genotypes showed early flowering (30 days); this was associated to the genotype-environment mechanism, because during the initial nursery stage the plants were irrigated three times per week and they were then subject to a hardening period without irrigation (seven days), after transplant to the field; in addition, in this period there was little rainfall (table 1), conditions which could have induced or anticipated flowering. Medel (2008) indicated that a short period of water stress before flowering onset increases flowering. Thus, the opposition between the vegetative and reproductive periods causes vegetative growth, produced by water stress, to favor the expansion of the flower bud and the distribution of absorption for inflorescence development. In citrus fruit trees the practice of drought periods is frequently used to induce flowering, in order to plan fruit production and harvest in the year (Cruz, Siquiera, Salomao and Cecon, 2006; Rodríguez et al., 2007).

The development of first-order branches in both genotypes started at 30 days of age of the plants, and the development of second-order branches, at 45 days (fig. 3A). Both branch types showed an increase in length until the end of the evaluation. In the two genotypes the growth of LB1 was similar until 90 days of age; while that of LB2, until 75 days, moment after which the blue genotype reached maximum length. When analyzing the curves of the NLB1, both genotypes were observed to show equal number towards 30 days of age, and then the white genotype reached the highest values (fig. 3B). In the blue genotype leaf formation in second-order branches started at 45 days of age, and in the white one, by 75 days of age, with a higher number of leaves in the blue genotype (fig. 3B). With the formation of first- and second-order branches the plant reached higher amount of foliage and higher leaf and photosynthetic organ surface, which allows an increase in the photosynthetic capacity (Portillo, Razza, Marín and Araujo, 2009).

After the analysis of the morphological characteristics of the two genotypes significant differences were detected only in the variables: W3FL, W5FL, W7FL, FL, NNBLS, LSL, NPP, NSP, PL and CP (table 2). Fifteen days after planting, both genotypes showed cotyledonal leaves which remained until 45 days after seeding-, followed by two simple leaves located on the first node (fig. 4A). Afterwards, compound, imparipinnate, trifoliolate leaves were observed up to the third node (fig. 4B). From the fourth node several pentafoliolate leaves developed (fig. 4C), which indicated the flowering onset, and the first flower buds appeared 30 days after seeding, which remained until the end of the evaluation. The pentafoliolate leaves represent an indicator of flowering, valuable aspect which should be considered in crop management planning, such as (organic) fertilization and irrigation. The blue genotype showed higher FL values (table 2). In the flowers of this genotype violet-bluish color prevailed (Pantone® 2746C) in the vexillum and yellow-greenish in the central part of the vexillum (Pantone®386C) and the calyx (Pantone®373C), and it was greener in the calyx; while in the other genotype the white color prevailed (Pantone®649) in the vexillum and yellow-greenish (Pantone®373C) in the calyx (fig. 4D and 4E). In the blue genotype few plants were observed to show flowers with more developed wings (fig. 4F).

At 45 days of age the plants produced heptafoliolate leaves (fig. 4G) which indicated the emission of long leading shoots, which tangle in the trellis system; the leading shoot length was higher in the white genotype (table 2). At this moment the presence of green pods was observed (fig. 4H) in both genotypes; branching was also remarkable, more profuse in the blue genotype. Regarding the width of the trifoliolate, pentafoliolate and heptafoliolate leaves, the blue genotype showed the highest values, which suggests that this variable is a possible morphological indicator of high vigor or plant growth. The apex of the folioles in the leaves was classified as acute, of truncated basis, and smooth edge, according to the general leaf classification indicated by Lindorf, Parisca and Rodríguez (1999).

With regards to the production characteristics, at 90 days of age, the blue genotype was found to have around 45 pods per plant and was surpassed by the white one (70). The NSP was significantly higher in the white genotype; however, no differences were observed regarding SL and SW. Differences were found between the two genotypes in the CP content, which was higher in the blue one (table 1). The high CP content may be associated to the organic management of the plants during their establishment (fig. 4I-L) and to rainfall distribution in the different crop stages, which could have contributed to the fact that they expressed their maximum development potential, even on a poor soil, of low biological activity and with presence of argillic.

The CP contents in this study largely surpassed the contents reported by Ramos et al. (2008), who found between 21,31 and 26,56% CP in the blue accessions, and between 23,81 and 24,56% in the white ones; as well as 23% of CP indicated by Marín, Carías, Cioccia and Hevia (2003). They were also higher than the ones reported for forage trees and shrubs recommended as feeding alternative for cattle rearing (García and Medina, 2006; Delgado and Ramírez, 2008). In this sense, the described results are relevant, if it is considered that higher importance has been ascribed to the nutritional profile in terms of protein- of the biomass of trees and shrubs, as compared to other herbaceous or creeping species. Regarding the DM percentage, no differences were observed between the genotypes (table 2); the values were considered high and close to those reported by Ramos et al. (2008), who obtained 90,58 and 96,23% in the blue accession, and 95,57 and 96,11% in the white one.

The analysis of the morphological and growth characteristics of the blue and white genotypes allowed concluding that both showed a similar performance in the different plant development stages: vegetative, flowering, and pod and seed production. The blue genotype showed higher vigor or growth and high branching capacity, and the white one had high pod and seed production, which constitute important morphological indicators. Based on the DM and CP contents and on the fast growth of this herbaceous legume, it is recommended as a short-term supplementary protein source; and, in addition, the application of (leached) cattle manure or any other manure from the same production units is suggested. The obtained results contribute to the characterization of this species and its management, which has not been reported in Venezuela, and lay the foundations for continuing studies related the management of this crop.

 

ACKNOWEDGEMENT

To the CONDES-LUZ for the funding granted under the research projects No. 0216-12 and to the university nursery of LUZ for providing its facilities.