Simultaneous selection for yield and stability in cassava (Manihot esculenta Crantz) under optimum condition in Nigeria

High yield and stability of performance across environments are required for success in agricultural systems. Selection of high yielding and stable genotypes is always hindered by Genotype by environment interaction (GEI). Hence plant breeders have proposed diverse methods to help increase their efficiency of selection in multi environment yield trials. The aim of this experiment was to identify and select high yielding and stable cassava genotypes using to GGE biplot and Yield selection index (YSi) techniques and to determine the efficiency of the techniques. 24 hybrids and 2 checks were tested across five locations in Nigeria using the randomized complete block design with 4 replications. ANOVA provided clear evidence of GEI (P<0.001) for dry root yield. 23 of the hybrids showed 10% to 83% economic improvement over the best check variety. GGE biplot identified genotypes TMS-IBA090574, TMS-IBA090521, TMS-IBA090590 and TMS-IBA090576 as high yielding and stable, while YSi selected 14 genotypes TMS-IBA090454, TMS-IBA090504, TMS-IBA090506, TMS-IBA090509, TMS-IBA090521, TMS-IBA090523, TMS-IBA090536, TMS-IBA090564, TMS-IBA090574, TMS-IBA090576, TMS-IBA090581, TMS-IBA090590, TMS-IBA090597 and TMS-IBA090609. Six of these genotypes (TMS-IBA090523, TMS-IBA090536, TMS-IBA090574, TMS-IBA090581, TMS-IBA090506, TMS-IBA090521) were marked as unstable by the Shuklas stability variance which indicates a weakness of YSi. The two techniques identified TMS-IBA090590 and TMS-IBA090576 as high yielding and stable genotypes. These two clones performed above average in at least four of the five locations used for evaluation. The genotypes jointly selected by both methods may therefore be released to increase the productivity of the crop in Nigeria and comparable agro ecology.