Florence Zapico, Catherine Hazel Aguilar, Marianita Eroy, Joey Real Marie Barbosa, Deocyth Sarsalejo and Sittie Rokaiya Banisil
Science Department, College of Natural Sciences and Mathematics, Mindanao State University- Fatima, General Santos City 9500 Philippines
Philippine Coconut Authority- Region 11, Bago Oshiro, Tugbok District, Davao City, 8000 Philippines
Key words: Musa textilis L. Nee, principal component analysis, GxE interaction, Pearson’s correlation, cluster Analysis.
Taximetric statistical methods were carried out on morpho-agronomic traits from three Manila hemp (Philippine abaca) varieties planted in three locations in Southern Mindanao, Philippines. Principal Component analysis (PCA) resulted in the extraction of five principal components which collectively accounted for 99.85% of total variance. Correlation analysis revealed highly significant relationships between % survival/number of mature plants at harvest time and pseudostem height and circumference. Three – way multi-environment trial (MET) analysis resulted in the formation of two biplots which jointly revealed groupings based on component loadings. Hierarchical cluster analysis sorted the varietal entries into two groups and highlighted the divergence of Glan abaca plants from those grown in the other two sites. Similar trends for Bongolanon obtained for % survival and number of harvestable plants in all three locations suggests that these traits may have a very strong genetic component, but pseudostem traits were discovered to be highly susceptible to environmental conditions and as such can be manipulated to increase fiber yield. Results rationalize the selection of Bongolanon for wide scale propagation in Southern Mindanao. Finally, these results will facilitate recommendation of abaca genotypes with broad adaptability and/or identify suitable genotypes for targeted agro-ecozones to prevent or minimize economic losses that farmers are most likely to incur if an unsuitable abaca variety is planted on their farms.
Get the original articles in Source: Volume 4, Number 2, February 2014 – IJAAR
Published By: International Network for Natural Sciences