Selection procedures for durable resistance in wheat

A wheat breeding programme for durable resistance to all locally important pathogens: leaf rust, stem rust, powdery mildew, Septoria nodorum, Septoria tritici, Cochliobolus sativus, Fusarium graminearum, Common Root Rot, Barley Yellow Dwarf Virus and Soil Borne Mosaic Virus was conducted in Brazil. The objective of the programme was to determine the feasibility of accumulating useful levels of partial resistance, which is quantitative resistance based on minor genes and considered to be more durable than resistance based on major racespecific genes.Four different selection procedures of breeding for partial resistance were applied and evaluated for their efficiency: Line Selection (LS), Single Plant Selection (SPS), Bulk Seed Selection (BSS) and Natural Selection (NS). The SPS procedure was made up of a system in which wheat was changed from a selfpollinated crop to a cross-polinated crop with the aid of a male gametocide and a recurrent selection was carried out in the segregating population. Selected plants were multiplied in the off-season and intercrossed again in the following generation. The LS procedure consisted of a selection in the F4 in ear to row plots. The best plants of the best rows were multiplied in the off-season and in the F6 the best lines were selected for a new cycle of outcrossing and line selection. The BSS was also an early generation selection as the SPS procedure based on grain qualities, while the NS procedure consisted of crossing only without any artificial selection.Crossing was done with the male gametocide ethrel. The optimum concentration for the application of ethrel in combination with the growth regulator gibberellic acid-3 under field conditions has been investigated. From these experiments one application of 2000 ppm ethrel a.i. 2-chloro-ethyl-phosphonic acid in 1000 l. water/ha at growth stage 41 - 43 DC in combination with an application of 150 ppm gibberellic acid-3, 500 l. water/ha three to four days later was most effective. Depending on environmental conditions and genotype about 60% - 80% cross pollination can be achieved. One new product R-111601 was tested but did not give the desired result.The selection procedures were compared for partial resistance to leaf rust, stem rust, powdery mildew, S. nodorum, C. sativus, F. graminearum, Common Root Rot, BYDV and SBMV and yield.In general, selection in later generations (F4 and F6) of the LS procedure showed to be more efficient than selection in early generations (F2) of the SPS procedure. The lower efficiency of SPS to be due to a relatively low heritability for partial resistance and the effect of interplot interference in the field under such conditions. As selection in later generations requires a large amount of time in the breeding procedure, an attempt was made to reduce this delay by a Single Seed Descent method, by which system generations can be advanced in 70 - 75 days. Through BSS seed selection it also was possible to accumulate partial resistance, while progress through natural selection was almost nil.The results convincingly showed that it is possible to simultaneously improve partial resistance to several pathogens and subsequent yield. Substantial increase in partial resistance was obtained for all pathogens with the exception of SBMV.The yield increase of the LS with 18.6% per selection cycle is superior to those of the SPS, BSS and NS procedures with 4.4, 2.8 and 3.7% respectively.From the results it is concluded that this new wheat breeding approach is of special interest for areas where many pathogens occur and the durability of resistance is a main concern. This applies in the first place to developing countries where stable crop yields are of extreme importance