Analysis Of Accuracy And Stability In Regional Trial Of Potato Varieties

In order to analyze potato regional trial precision, suitable stability analysis model was chosen,which research different level, physiological age and sliced size of potatoes,using CEV and RLSD index in test and variety comparision precision,analysing influence on three facters,found yield and stability right estimate. Meanwhile, using mid-late northwest group(be short of 2006), mid-late northwest group(be short of 2004) and mid-late southwest group(be short of 2004) country regional trial data of potatoes in 2002-2009, which estimated test precision and pilot precision of each point trial, using CEV and RLSD index in test and variety comparision precision and GCV index in pilot precision.Analysis result influence on random and fixed variance analysis model in pilot effect using mid-late northwest group country regional trial six group data in 2005-2010. Stability analysis of varieties was variance using LR model and AMMI model comparision. The main results that showed, Effect of test precision in different level potatoes1.With using potatoes level was reduced,both of CEV and RLSD were increased.Original potatoes of frist generation were CEV<5% and RLSD<12%, original potatoes of second generation were CEV<10% and RLSD>12%, which showed there were good effect of controling trial error and test precision using original potatoes of first and second generation.However,using first level potatoes and productive potatoes were CEV>10% and CEV>15% respectively.which showed first level potatoes and productive potatoes were unfit for potato seed in regional trial.2.Effect of test precision in different physiological age potatoesUsing young age potatoes had minimum numerical value of CEV and RLSD, this is CEV<10%, which showed there were good effect of controling trial error and test precision.But using tender potatoes and small age potatoes had 10%<CEV<15%,older age potatoes had CEV>15%, which showed the older age potatoes weren't fit for potato seed in regional trial.Using the young age potatoes had minimum numerical value of RLSD that showed there were good precision in varieties comparision oftest, using the tender age and older age potatoes had maximum numerical value of RLSD that showed there were bad precision in varieties comparision oftest.3.Effect of test precision in different sliced size potatoesUsing potato seed weight 45g had CEV<5% and RLSD<12%, which showed there were good effect of controling trial error and test precision.second were potato seed weight 60g and 15g, that were CEV<10%, RLSD<20%, the potato seed weight 30g was CEV>10% and RLSD>20%, which showed potato seed weight 60g had good effect of controling trial error and test precision than potato seed weight 30g and 15g.To sum up, Potato seed weight was controled about 45g in potato regional trial,this was best effect of controling trial error and test precision.4.Precision of regional trial Only shortage 20% pilots were CEV<5% in mid-late northeast group; Only shortage 8% test could identify 5% yield variance of varieties, and approximate 14% test only idetifly 8.1% to 12% yield variance of varieties,in addition,69.5% test only idetifly greater than 12% yield variance of varieties in one-year-one-location test from RLSD index.At present,one index of CEV was one-sided which analysed precision in potatoes regional trial,we should use two indexes of CEV and RLSD which were analysis and comprehensive evaluation in one-year-more-location test results.5.Pilot resolutionAll pilots mean value GCV were greater than 12% from mid-late northeast group, mid-late northwest group and mid-late southwest group in one-year-one-location test,i.e there were outstanding variance and obvious resolution.For example,Zhalantun pilot and Keshan pilot of mid-late northeast group weren't obvious resolution,the others pilots were obvious resolution, Weining pilot, Bijie pilot and Xichang pilot of mid-late southwest group weren't obvious resolution,the others pilots were obvious resolution.It was worth that not only Guyuan pilot and Weiyuan pilot have low resolution, and test precision and compare precision of varieties were low.6.Effect of test results to variance analysis modelThere were variance analysis model of pilot effect random and pilot effect fixed in potatoes regional trial in one-year-one-location test. Testing potatoes varieties had whether significant variation or very significant variation to evaluation have direct influence in regional trial,therefore,the variance analysis model of pilot effect random was chosen in potatoes regional trial,for enhancing yield potential of cultivar assessment and reducing promotion risk of new varieties.7.Comparision of stability analysis modelFrom LR model of stability results seen, the others testing varieties had the higher dynamic stability.AMMI model indicated that there was discrepant of different varieties stability, Therefore,the AMMI model had the higher reliability in stability evaluation of variety, the model should be adopted in analysing potatoes regional trial.