Agro-morphological Characterization of Arabica Coffee Cultivars in Burundi

Abstract

Coffee is the most valued commodity among the stimulant crops. Its production is important in over 80 increasing countries including Burundi, for which it is the key foreign currency earner. The present study sought to contribute to the knowledge base required for the improvement of Arabica coffee in Burundi. The specific objectives of the study were to (i) determine diversity among Arabica coffee cultivars in Burundi using agro- morphological characters; (ii) assess the effect of locations on agro-morphological diversity among cultivated Arabica coffee cultivars in Burundi. In line with these objectives, two trials were conducted using two sets of coffee cultivars: Trial one with twenty accessions was conducted at Rukoba; while Trial two with fifteen genotypes was conducted at three locations (Kayanza , Rukoba and Nyange) in Burundi. For the first trial, data was collected for 11 qualitative and 17 quantitative morphological traits over one season and 10 quantitative traits for second trial using IPGRI coffee descriptors. Frequency distribution of the qualitative traits was assessed using visual counts. Analysis of variance (ANOVA) was conducted on the quantitative morphological traits for within and across location. The information on the relative importance of the morphological characters assessed was obtained using Principal Component Analysis (PCA) and the relationships among the accessions evaluated was determined using cluster analysis. Results from the Trial one showed variation in most of the traits assessed as evidenced by the coefficient of variation and frequencies of the qualitative and quantitative traits, respectively. The first and second principal components accounted for 44.79% and 19.61% of the total variability, respectively. PCA shown that Plant height (0.88), number of internodes per branch (0.89), internode length of primary branch (0.82), stem girth (0.69), length of primary branches (0.9), number of cherries per internode (0.75), hundred fruits weight (0.63), fruit length (0.68) and number of primary branches (0.63) were the chief characters to categorize the Arabica coffee genotypes studied. Cluster analysis clustered the Arabica coffee genotypes into four clusters. The diversity amongst the Arabica coffee accessions in both quantitative and qualitative traits revealed by this study can be used for trait enhancement through selection and germplasm conservation. Results from Trial two showed that the effects of environments were significant (p < 0.05) to highly significant (p < 0.01) for most of the traits. The accession and environment interaction (GEI) were also significant (p < 0.05) to highly significant (p < 0.01) for fruit and bean traits, demonstrating the presence of variability among the established materials for these traits. Phenotypic variations for fruit and bean quantitative traits were relatively different across environments indicating that these characters were under strong effect of environment. In contrast, phenotypic variation for leaf quantitative traits of tested coffee accession were not relatively different across environments. Cluster analysis based on the distance measures and PCA biplot graph grouped the fifteen accessions into three main clusters according to their genetic background. Evaluation of morphological traits showed that large phenotypic variation among fifteen coffee genotypes for most of the traits as expressed by the moderate to high plot-basis broad sense heritability (h2bs= 0.24 to 0.82). Highest heritability estimates were for leaf area, leaf width, fruit length and leaf length (h2 bs= 0.502 to 0.82). The character leaf area, hundred cherries weight, hundred beans weight, leaf width had relatively highest GCV compared with the others with respective value of 23.58%, 13.78%, 6.107 % and 5.340% with corresponding PCV of 26.70%, 16.37%, 9.88% and 6.702) indicating the relative importance of these traits for improvement of coffee in Burundi.