Spring Wheat Breeding: Evaluation of Selected Adapted Spring Wheat Germplasm in Eastern Canada

Twenty-five hard red spring wheat (Triticum aestivum) lines, including three known cultivars used as checks, were grown in seven locations across Eastern Canada. The objective of this multi-location experiment was to evaluate selected Eastern Cereal and Oilseed Research Centre advanced lines (ECAD lines) from the Spring Wheat Breeding Program in order to identify the best lines for performance and grower trials. The lines from this trial performed very well compared to the check varieties, especially at the Ontario locations. Overall, the ECAD lines were on a par with or superior to the checks in terms of several attributes, including yield, protein content, and Fusarium head blight resistance.


Introduction
Wheat (Triticum aestivum) breeding programs across Canada aim to produce high-yielding varieties (McCaig and DePauw 1995) with improved resistance to diseases, especially Fusarium head blight (FHB) (Gilbert & Tekauz 2000;McCaig & DePauw, 1995), early maturity (Thomas & Graf, 2014), and high protein content (Wang et al., 2002).Through these programs, crop losses have been greatly reduced and grain quality has been increased (Thomas and Graf 2014).Of particular interest to Canadian wheat growers are cultivars that have a short growing season, produce high yields, and are resistant to FHB (Preston et al. 1991;McCaig & DePauw 1995).Common attributes of interest to commercial mills are grain protein, falling number, test weight, and deoxynivalenol (DON) values (Darby, 2015).
Fusarium head blight resistance is currently of great importance for wheat production in Canada.Epidemics of FHB in 1980 in Eastern Canada andin 1993 in Manitoba sparked interest in developing FHB-resistant cultivars and FHB management strategies (Gilbert & Tekauz, 2000).There is currently no single control method for FHB; the best approach involves the use of a cultivar of intermediate resistance in combination with fungicide applications, and suitable farming practices (Gilbert & Tekauz, 2000).More than 0.25% Fusarium-damaged kernels by weight can downgrade wheat to the point of significant economic loss (Fernandez et al., 2005).In a spring wheat analysis conducted by Darby (2015), 13 out of 19 tested varieties had FHB symptoms, indicating that the disease is a severe problem.The risk of FHB is increased by warm and humid weather, short plant height, conservation tillage, and wheat-maize rotations (Gilbert & Tekauz, 2000).Practices that reduce FHB risk consist of shredding maize residues to decrease DON values, performing conventional tillage, and using fungicides such as tebuconazole (Gilbert & Tekauz, 2000).Despite such practices, FHB is difficult to manage, because shorter plants are more susceptible, and more than half of the world's wheat cultivars contain two major dwarfing genes that are gibberellic-acid-insensitive (Gilbert & Tekauz, 2000).Furthermore, FHB thrives in warm, humid climates, and climate change may create environmental conditions that are favorable for the disease (Gilbert & Tekauz, 2000).
Grain yield is very important to wheat growers, and thus many studies on how to increase yield have been published.The optimal test weight is considered to be 56 to 60 lb/bushel (Darby, 2015), and newly registered wheat cultivars often have increased yields relative to check varieties with which they are compared.In previous studies, two examples of this are 'Snowbird', which gave higher yields than all check varieties except 'McKenzie' (Humphreys et al., 2007) and 'Carberry', which yielded 9.6% more than all checks (DePauw et al., 2011).Wheat breeders attempting to produce high-yielding cultivars can adopt an approach that involves using hybrids of spring wheat and winter wheat: the mean yields of spring/winter hybrids were found to surpass those of all controls (McKenzie & Grant, 1974).A study conducted by Hucl and Baker (1987) found that crop grain yield was correlated with biological yield and that cultivars with a later heading date produced more spikelets and more kernels per spike and also had increased yield.The same study found that some cultivars produced significantly fewer spikes per square meter, but the kernels were heavier, resulting in a net 25% increase in crop yield (Hucl & Baker 1987).Another study on yield reported that grain yield was not associated with tillering capacity or mortality but was instead associated with kernel weight (Hucl & Baker, 1988).That study concluded that grain yield showed no association with tiller density or tiller mortality (Hucl & Baker, 1988).Biomass production can be modeled by the new DNDC-CSW crop model, which was found to be acceptable for describing crop growth processes (Kröbel et al., 2011).
In addition to FHB resistance and high yield, grain protein is an important attribute of newly registered spring wheat cultivars.Industry standards for protein content are 12% to 14% (Darby 2015).In a study that used mouse bioassays to evaluate the protein quality and digestibility of wheat, it was found that barley protein is superior to wheat protein and that lysine is the first-limiting amino acid in wheat (Bell & Anderson 1984).That study also found certain wheat cultivars with superior performance: 'Neepawa' had the highest protein digestibility and 'Twin' had a superior protein rating (Bell & Anderson, 1984).Nevertheless, in another study, grain yield and grain protein concentration were found to be negatively correlated (Löffler & Busch, 1982).
The objective of this multi-location experiment was to evaluate selected Eastern Cereal and Oilseed Research Centre advanced lines (ECAD lines) from the Spring Wheat Breeding Program at seven different locations across Eastern Canada in order to identify the best lines for performance and grower trials.

Statistical Analysis
Analyses of data obtained from the 25 wheat lines as well as all data obtained from the seven locations in Eastern Canada were analyzed using the MIXED procedure of SAS 9.3 (SAS Institute, Cary, NC, USA) after the homogeneity of the experimental error was examined, and the means were compared using least significant differences (0.05) when the differences were significant.

Results
No interaction between line and location, indicating that the relative performance of the lines was similar.
When the yield parameter was compared among the seven locations (Table 3), the CEFNT location was found to have the highest mean value (4222 kg/ha), which was significantly higher than the mean values recorded for the Kincardine, Princeville, St. Isidore, and Harrington locations (3833, 3427, 1892, and 1020 kg/ha, respectively) but did not differ significantly from the mean values for the Palmerston and CEF9C locations (4206 and 4128 kg/ha, respectively).Comparing the days to head parameter, the CEF9C location had the lowest mean value (47), which was significantly lower than the mean values for the Harrington locations (60) but did not differ significantly from the mean values for the Kincardine, CEFNT, St. Isidore, and Palmerston locations (48, 49, 49, and 49, respectively).In terms of the test weight parameter, the Harrington location had the highest mean value (94.0 kg/hl), which was significantly higher than the mean values at the Princeville, CEFNT, CEF9C, St. Isidore,and Palmerston locations (78.9,77.9,77.5,77.5,and 76.9 kg/hl, respectively).When the thousand-kernel weight parameter was compared among the seven locations, the Palmerston location had the highest mean value (40.0 g), which was significantly higher than the mean values for the CEF9C, CEFNT,Harrington,St. Isidore,and Princeville locations (37.6,36.9,34.5,34.0,and 33.0 g,respectively).With regard to the plant height parameter, the Palmerston location had the highest mean value (117.9 cm), which was significantly higher than the mean values recorded for the Princeville, CEF9C, CEFNT,Kincardine,St. Isidore,and Harrington locations (105.4,101.1,99.2,89.2,84.8,and 73.3 cm,respectively).Comparing the protein content parameter among the seven locations, the Princeville location had the highest mean value (14.9%), which was significantly higher than the mean values for the CEF9C, St. Isidore, and CEFNT locations (14.6%, 14.1%, and 12.3%, respectively) but did not differ significantly from the mean value at the Palmerston location (14.7%).LSD -least significant difference at the 0.05 level.

Discussion
A In the current study, the ECAD lines generally performed best among all the tested lines based on the data analysis results for the different attributes.This is probably due to the pedigree of the lines, considering, for example, that 'Norwell' has performed well in terms of most of the attributes described (Canadian Food Inspection Agency, 2015).The lines at the research station in Quebec also have a resistant pedigree thanks to varieties such as 'Cadillac', which has been shown to be resistant to mildew disease and which performs better in drier and cooler environments.As well, FL62R1, which is in the pedigree of some of the varieties, has been shown to be resistant to FHB, which prefers wet and humid environments; the best performance for FL62R1 is obtained in cooler and drier environments.
The results obtained for the average attributes per location for the 25 spring wheat lines (Table 3) clearly demonstrate that some sites contain extreme outliers for certain attributes.These extreme highs and lows can alter the values given in Table 2, which represent the average values for each ECAD line across the seven locations tested.For example, the very low yield and protein content values for the Harrington location could explain the lower per-line yield and protein content estimates shown in Table 2.If these outliers are taken into account, the yield and protein content values for the ECAD lines would likely be much higher than the respective estimates in Table 2, whereas the values for days to head, test weight, and height are likely lower than their respective estimates in Table 2 are.Considering this, the ECAD varieties performed very well when compared to the check varieties.
Given the range of conditions to which the ECAD lines were exposed at the seven different locations, they performed quite well, especially at the CEF9C, CEFNT, Palmerston, and Kincardine locations.

Table 1 .
Pedigree of selected lines

Table 2 .
Average attributes of 25 spring wheat lines tested at seven locations in Eastern Canada in 2014

Table 3 .
Average attributes per location of 25 spring wheat lines tested at seven locations in Eastern Canada in 2014 Legend: TSTWT -test weight; TKW -thousand-kernel weight; PROT -protein content; LODG -lodging; FHB -Fusarium head blight; DON -deoxynivalenol; FDK -Fusarium-damaged kernels; CEF9C -Central Experimental Farm field, CEFNT -Central Experimental Farm field with no-till, both located at the Government of Canada research center in Ottawa, Ontario; St. Isidore -Government of Canada Rental field in St. Isidore, Ontario; Harrington -field located at the Government of Canada research center in Harrington, Prince Edward Island; Palmerston -field located at the C&M Seeds company's research facility in Palmerston, Ontario; Kincardine -field located at the Dow AgroSciences research facility in Kincardine, Ontario; Princeville -field located at the Semican company's research facility in Princeville, Quebec; " -" -attribute not measured at that location.