Yield Stability of Commercial Hybrid Maize in Thailand

 

Parinya Sebunruang¹ Ratchada Pratchareonwanich² Anon Malipan³

Annop Kasivivat⁴ and Preecha Sangsoda⁵

 

1. E-mail: psebunruang@yahoo.com

………………………………………….

 

Approximately, 95% of maize seed required by Thai maize farmers for maize planting each year are hybrids.  Two public agencies which produce hybrid maize seed can not meet the farmers' requirement.  Therefore: many private seed companies are involved in this business and offer many hybrids to the farmers.  To verify only good hybrids be sold to the farmers Nakhon Sawan Field Crops Research Centre has carried out an experiment to test the performance of hybrids which available in the market.   The experiment was conducted in 2004, rainy season to evaluate yield potential and yield stability of 17 hybrids from private companies and government sectors.  Treatments were arranged in a randomized complete block with four replications at four environments in maize belt of Thailand, namely, Nakhonsawan Field Crops Research Centre (NSW), Nakhon ratchasima Field Crops Research Centre (NM), Praphutthabat Field Crops Experiment Station (LB) and Phetchabun Field Crops Experiment Station (PBN).  Individual plot consisted of four rows of 5 meters long with the row spacing of 75 cm. and 20cm. between plants.  The result revealed that varieties had highly significant differences in yield potential with coefficient of variation among environments ranged from 5.7-15.7%.   The variety by environment interactions was also highly significant differences indicating that each variety performed differently in different environments.  The deviation from regression of the variety mean square was more important for estimating yield stability than the regression of the variety mean on the environmental index.  Thirteen hybrids produced 10-25% (P>0.05) higher yields than check variety, NS72 (7.6 t/ha).  Stability parameters were estimated by the method proposed by Eberhart and Russell (1966).  A desirable variety is the one with a high mean yield, a unit of regression coefficient (b = 1) and the least deviation from regression.  Therefore four out of the thirteen hybrids i.e. KSX 4452, 30A33, BIG 949, and NK 48 produced consistently high yield over the test environments.