PHENOTYPIC CORRELATION AND REGRESSION ANALYSIS OF YIELD AND FIBRE TRAITS IN UPLAND COTTON (GOSSYPIUM HIRSUTUM L.)

Abstract

A field experiment to determine correlation and regression analysis between yield and fiber traits in cotton genotypes of Gossypium hirsutum L. was conducted at the experimental area of the Botanical Garden, Department of Plant Breeding and Genetics, Sindh Agriculture University Tandojam, during the year 2012. The experiment was laid-out in a Randomized Complete Block Design with four replications and eight varieties viz. NIAB-78, CIM-496, BH-160, CRIS-134, IR-1524, FH-113, MG-6, and CIM-534. The data were recorded on plant height (cm), sympodial branches plant-1, bolls plant-1, boll weight (g), seed cotton yield plant-1 (g), lint%, fiber length (mm), seed index (g) and micronaire value (µg/inch).The analysis of variance revealed significant differences among the genotypes for all the traits studied. Based on mean performance, variety FH-113 produced tallest plants; NIAB-78 produced highest number of sympodial branches plant-1; IR-1526 recorded maximum bolls plant-1; CIM-534 weighed bigger bolls and measured longer fibre length; CRIS-134 produced maximum seed cotton yield plant-1 and also gave desirable micronaire. While maximum lint% was ginned by MG-6 and more seed index was recorded by BH-160. The phenotypic correlation and regression analysis revealed significantly positive associations of sympodial branches plant-1 with bolls plant-1; between bolls plant-1 and seed cotton yield plant-1 and the coefficient of determination (r2) indicated that about 3.96% of total variation in bolls plant-1 was due to its association with sympodial branches plant-1, while 78.49 % of total variation in seed cotton yield was due to its correlation with seed cotton yield. Bolls plant-1 demonstrated a highly significant but negative correlation with boll weight and r2 revealed that 40.57 % of total variation in boll weight was due to its relationship with bolls plant-1. These results indicated that increase in bolls plant-1 caused decrease in boll weight. The significant negative association between lint% and seed index suggetted that increase in seed index caused corresponding decline in lint% while r2 determined about 50.41 % of the total variation in lint%. Results generally indicated that number of bolls plant-1 and lint% may be the means to obtain higher seed cotton yield.

Keywords

• Correlation
• regression analysis
• upland cotton
• yield and earliness traits