ABSTRACT

The study was carried out to assess the yield stability of some taro (Colocasia esculenta (L.) Schott) genotypes commonly grown by farmers in Nigeria as influenced by some cropping system management practices (two locations in two years). The experiment was laid out as a split plot in a randomized complete block design with three replications. The six genotypes used were: NCe 001, NCe 002, NCe 003, NCe 004, NCe 005 and NCe 010. Data were collected on the growth characteristics and yield attributes. There was significant effect of crop management system on most of the growth and yield attributes measured. The genotypes varied significantly (p<0.001) in their performance for almost all the traits studied. Location had significant effect on taro yield with Ishiagu giving the best performance for growth and most of the measured yield on attributes. Genotypes had strong influence all the characters considered in this study with NCe 002 and NCe004 consistently giving the best performance in yield and few of the studied traits. This indicates that these genotypes show good adaptation to the soil and climatic conditions of these agro-ecosystems. AMMI showed high significance on the taro yield. Effects of genotype and environment were highly significant. The main effects (genotypes and environments) captured 62.9% of the total sum of squares (TSS) while the genotype by environment interaction (GEI) contained 8.2% of the TSS for population density. Under propagule the main effects (genotype and environment) captured 52.1% of the total sum of squares (TSS) while the GEI explained 17.2% of the total variation. For fertilizer rates, the main effects (genotypes and environments) captured 87.6% of the total sum of squares (TSS) while the GEI explained 4.5% of the TSS. The linear correlation (r) coefficients between the various attributes and the yield as affected by fertilizer rates showed that the correlation in number of secondary suckers and cormel number per plant was positively and significantly associated with yield (t/ha), r = 0.223** and r = 0.169*, respectively. However, plant height, number of leaves, leaf area, cormel length and cormel circumference were negatively but significantly associated with yield (t/ha). With regards to the linear correlation (r) coefficients between the various attributes and the yield per hectare as affected by propagule sizes, the cormel number per plant and cormel weight per plant were positively and significantly associated with yield (t/ha), r = 0.549** and r = 0.330*, respectively. However, plant height, number of leaves, number of secondary suckers and leaf area were negatively but significantly associated with yield (t/ha). In 2016, plant h==[-eight was the largest contributor to yield accounting for up to 18 % of the total yield (B = 0.421), while cormel number per plant was observed to be the largest contributor to yield accounting for up to 38 % of the total yield (B = 0.619; P<0.001) in 2017.