Relative efficiency of experimental designs in evaluation of plant genetic resources

Abstract

An experiment was conducted in a Reinforced Alpha Design (RAD) with 50 accessions and 5 control treatments of tomato during the Rabi season of 2006. The data on 9 distinct descriptors namely, plant height, primary branches, flower clusters per plant, flowers per cluster, days to first fruit set, fruits per plant, fruit weight per plant, locules per fruit and pericarp thickness were recorded on 12 randomly selected plants per accession. The designs, namely, Completely Randomized Design, Randomized Block Design and Augmented Block Design were derived from RAD. A comparison of the relative efficiency of these four designs revealed that in plant genetic resources evaluation for identification of promising material in case of insufficient seed quantity, one should adopt a two phase strategy using reinforced alpha design. During phase I (first year), when sufficient seed quantity is not available, one replication of reinforced alpha design may be used. During phase II (second year), the second replication of the design may be deployed.