Effects of seeding rate, nitrogen rate and cultivar on barley malt quality

Research Article Received: 28 October 2011 Revised: 29 February 2012 Accepted article published: 5 March 2012 Published online in Wiley Online Library: 23 April 2012 (wileyonlinelibrary.com) DOI 10.1002/jsfa.5687 Effects of seeding rate, nitrogen rate and cultivar on barley malt quality Michael J Edney, a John T O Donovan, b T Kelly Turkington, b George W Clayton, c Ross McKenzie, d Pat Juskiw, e Guy P Lafond, f Stewart Brandt, g Cynthia A Grant, h K Neil Harker, b Eric Johnson g and William May f Abstract BACKGROUND: Crop management tools have been shown to affect barley kernel size and grain protein content, but the direct effect on malt quality is not well understood. The present study investigated the effect of seeding rate, nitrogen fertilisation and cultivar on malt quality. RESULTS: Higher seeding rates produced barley with less grain protein and smaller, more uniformly sized kernels. The small, uniformly sized kernels modified more completely, leading to malt with higher extract and lower wort β-glucan than malt from low-seeding-rate barley. Increasing rates of nitrogen fertilisation caused grain protein levels to increase, which limited endosperm modification and reduced malt extract levels. AC Metcalfe showed better modification and higher malt extract than CDC Copeland, but CDC Copeland had better protein modification at higher fertilisation rates, which resulted in less reduction of malt extract as nitrogen rate increased. CONCLUSION: Higher seeding rates reduced kernel size and grain protein levels without compromising malt extract owing to better endosperm modification of the more uniformly sized kernels. Negative effects of higher nitrogen rates on malt quality can be reduced through development of cultivars with improved ability to modify protein during malting. c 2012 Her Majesty the Queen in Right of Canada Keywords: endosperm modification; friability; germination index; grain protein; kernel size variability 2672 INTRODUCTION Western Canada is an important producer of high-quality malting barley, but on average less than 25% of annual production is selected for malting. The short growing season with limited precipitation and hot temperatures leads to a quality malt with high levels of starch-degrading enzymes and free amino acids, but conditions can restrict barley from achieving strict malt barley specifications. The short growing season limits kernel filling, leading to lower kernel plumpness, while inadequate precipitation and hot temperatures produce high levels of grain protein. 1,2 The strict specifications ensure the barley processes efficiently in the malthouse and brewery, leading to the maximum amount of beer possible. Malt extract, the malt quality parameter traditionally used to predict beer production, is maximised by processing plump barley with minimum levels of grain protein. The high levels of starch associated with plump, low-protein barley increase malt extract, but the barley endosperm must still be well modified to realise that extract potential. Endosperm modification is a physical change characterised by breakdown of the β-glucan in cell walls and the protein matrix entrenched among the starch granules and cell walls. 3 The walls must be degraded to allow starch-degrading enzymes eventual access to starch granules and release the sugars that contribute to malt extract. 4 More importantly, the β-glucan must be broken down to prevent Correspondence to: Michael J Edney, GRL 1050, Grain Research Laboratory, 1404-303 Main Street, Winnipeg, MB, R3C 3G8, Canada. E-mail: michael.edney@grainscanada.gc.ca a Grain Research Laboratory, 1404-303 Main Street, Winnipeg, MB, R3C 3G8, Canada b Agriculture and Agri-Food Canada, Lacombe Research Centre, 6000 C and E Trail, Lacombe, AB, T4L 1 W1, Canada c Agriculture and Agri-Food Canada, Lethbridge Research Centre, 5403 1st Avenue South, Lethbridge, AB, T1J 4B1, Canada d AlbertaAgricultureandRuralDevelopment, 54011stAvenueSouth, Lethbridge, AB, T1J 4V6, Canada e Alberta Agriculture and Rural Development, 5030 50th Street, Lacombe, AB, T4L 1 W8, Canada f AgricultureandAgri-FoodCanada, IndianHeadResearchFarm, Box760, Indian Head, SK, S0G 2K0, Canada g Agriculture and Agri-Food Canada, Scott Research Farm, PO Box 10, Scott, SK, S0K 4A0, Canada h Agriculture and Agri-Food Canada, Brandon Research Centre, PO Box 1000A, Brandon, MB, R7A 5Y3, Canada J Sci Food Agric 2012; 92: 2672 2678 www.soci.org c 2012 Her Majesty the Queen in Right of Canada

Effects of seeding rate, nitrogen rate and cultivar on barley malt quality www.soci.org problems in the brewery such as slow wort and beer filtration times and poor fermentation. 5,6 Protein must also be hydrolysed to allow access to the granules and in addition release the free amino acids essential for fermentation. Endosperm modification can be restricted in plump barley, because large kernels take water up more slowly 3 and contain higher levels of grain β-glucan. 7 Modification can also be restricted in grain with higher levels of protein owing to slow water uptake and uneven processing. 3,8 Kernel plumpness and grain protein levels are affected by crop management practices such as seeding rate, nitrogen fertilisation rate and cultivar selection. Higher seeding rates reduce kernel plumpness and, although grain protein level is also reduced, it is generally accepted that a lower malt extract level will result. 9 Higher seeding rates, however, also reduce variability in kernel size, 10,11 which can result in a more consistent endosperm modification. 12 Increasing the nitrogen fertiliser rate, necessary for maximum barley yields, results in higher grain protein levels, thinner kernels and greater variability in kernel size. 10 Acceptable levels of barley protein have been achieved at high fertilisation rates with breeding lines, but malt extract levels decreased significantly. 2 Commercial barley cultivars also differ in their response to nitrogen fertilization, as seen with CDC Copeland and AC Metcalfe, 10 but effects on malt quality are not well understood. Crop management studies for barley seldom determine malt quality directly owing to the expense and time requirements for malting and malt analysis. Comments on malt quality tend to be extrapolated from barley data or based on previous experiences. The present study, an extension of our previous publication that reported on effects of crop management on barley quality, 10 investigated malt quality directly. The study investigated seeding rates, nitrogen fertilisation rates, differences between two commonly grown Canadian malting cultivars and interactions among these treatments. MATERIALS AND METHODS Plant materials and treatments An experiment was conducted at multiple sites over four years to determine the effects of seeding rate, nitrogen rate and barley variety on malt quality. A factorial combination of seeding rate (200 and 400 seeds m 2 ), nitrogen rate (0, 30, 60, 90 and 120 kg ha 1 actual nitrogen) and barley cultivar (AC Metcalfe 13 and CDC Copeland 14 ) was randomised in a complete block with four replicates. The nitrogen, applied as urea (46-0-0) at seeding time, was banded to the side of and below the seed. Field experiments were conducted under no-tillage management at Fort Vermilion, AB (58 24 N, 116 0 W), Beaverlodge, AB (55 11 N, 119 26 W), Lacombe, AB (52 28 N, 113 44 W), Lethbridge, AB (49 41 N, 112 46 W), Canora, SK (51 63 N, 102 45 W), Scott, SK (52 21 N, 108 51 W), Indian Head, SK (50 32 N, 103 40 W) and Brandon, MB (49 50 N, 99 57 W), Canada between 2005 and 2008. Details of field operations and soil types were described previously. 10 Barley grain analyses Analyses of grain protein, germination energy and water sensitivity were performed according to the standard methods of the American Society of Brewing Chemists. 15 The present study used percentage germination under 8 ml conditions as an indication of water sensitivity rather than the official use of the difference between the 8 and 4 ml tests. Germination index was calculated from germinative energy results according to Riis and Bang- Olsen. 16 A Single Kernel Characterization System (SKCS 1400, Perten Instruments, Springfield, IL, USA) was used to measure kernel weight, diameter and hardness. 17 The system calculated averages for weight, diameter and hardness of 300 barley kernels together with their respective standard deviations, the latter being used as an indicator of kernel uniformity. Micromalting process and malt analyses Barley samples from all growing locations were tested for plumpness, germinative energy and protein content to determine suitability for malting. Constraints on capacity for malting and quality analysis limited the number of locations that could be malted and analysed each year. Selection of locations was based on barley quality (grain protein and germination). Barley from a total of 16 location/years was malted. Barley from two locations (Beaverlodge and Scott) was only malted from one year. Barley from all other locations was selected for multiple years. The number of replicates malted for each selected location/year varied but generally was greater than two. A total of 39 replicates spread across the 16 location/years were malted (780 samples). Only plump barley (>2.38 mm slotted sieve) was malted in a Phoenix Automated Micromalting machine (Adelaide, SA, Australia) using the following malting schedule: steeping (8 h wet,16 h air, 8 h wet, 12 h air at 13 C), germination (96 h at 15 C) andkilning(12 hat55 C,6 hat65 C,2 hat75 C,4 hat85 C 24h total). Steep-out moisture was calculated from the difference in weight between dry matter barley and steeped-out barley. Malt analyses included: (1) malt extract (fine grind), a measurement of the solubility of malt that indicates a malt s beer production potential; (2) Kolbach index, the ratio of soluble to total malt protein that indicates the extent of protein modification;(3) wort β- glucan, an indicator of the extent to which cell walls were degraded during malting; (4) diastatic power and α-amylase, enzymes that produce fermentable sugars from malt starch during mashing, the first phase of brewing. Analyses were performed according to the standard methods of the American Society of Brewing Chemists. 15 Malt modification and homogeneity of modification were assayed with both the friability method 15 and the Calcofluor staining method. 18 Statistical model and analyses Data were analysed using PROC MIXED of SAS. 19 Seeding rate, nitrogen rate and cultivar were considered fixed effects. Location by year combinations (environments) and their associated interactions with fixed effects were considered random effects, as were replicates within environments. Yang 20 suggests that in breeding and agronomic studies it may be more appropriate to consider year and location effects and their interactions with fixed effects as random, since the goal of most crop improvement programmes is to infer future performance at many untested locations. The large number of environments investigated in this study facilitated the use of this approach. Barley cultivar and seeding rate means were compared using Fisher s protected least significant difference (LSD) test. Contrast statements were used to test for linear and quadratic responses to nitrogen rate. The mixed model was used to obtain regression equations to describe relationships between nitrogen rate and dependent variables. All differences were deemed significant at α<0.05. 2673 J Sci Food Agric 2012; 92: 2672 2678 c 2012 Her Majesty the Queen in Right of Canada wileyonlinelibrary.com/jsfa

www.soci.org MJ Edney et al. RESULTS The average quality of barley from the 16 location/years selected for malting was appropriate for end use malting. Average grain protein level (dry matter, DM) for all malted samples was 111 ± 15 g kg 1 (mean ± standard deviation). Average kernel plumpness was 928 ± 50 g kg 1. The average 4 day germination was 98± 2%. Barley showed some water sensitivity, but the average value (92 ± 9%) was well above the commercially acceptable level of 80%. Malts produced from barley grown at the 16 location/years were reasonably well modified, as indicated by low wort β-glucan levels (124 ± 90 mg L 1 ) and high Kolbach indices (42.7 ± 4.0%). Malts had good levels of malt extract (805 ± 11 g kg 1 )and starch-degrading enzymes, the latter indicated by diastatic power (132± 27 Lintner) and α-amylase (58.0± 11.0 dextrinising units). Effect of seeding rate Seeding rate affected several aspects of barley and malt quality (Tables 1 and 2). Barley grown at the high seeding rate had lower levels of grain protein, greater potential for a rapid initiation to germination (germination index) and higher steep-out moistures (Table 3). Malts produced from high-seeding-rate barley had better endosperm modification, as indicated by lower levels of wort β- glucan, 112 vs 143 mg L 1, higher Kolbach indices, 42.9 vs 42.0%, and higher values for both friability and Calcofluor. Malts made from the high-seeding-rate barley showed slightly higher levels of malt extract, 806 vs 805 g kg 1 DM, but lower seeding rates resulted in higher malt yields, 921 vs 920 g kg 1. Levels of starchdegrading enzymes were similar between the two seeding rates, although malt from the high seeding rate had more α-amylase while malt from the low seeding rate had more diastatic power. Effect of nitrogen rate Increasing levels of nitrogen fertilisation affected nearly all aspects of malt processing and malt quality (Tables 1 and 2). Grain protein levels increased with increasing nitrogen rate, while germinative energy showed a parabolic relationship (Fig. 1). Germination index and steep-out moistures decreased with increasing nitrogen and, as a result, endosperm modification decreased, as indicated by increasing levels of wort β-glucan, lower Kolbach indices (Fig. 2) and lower values for both friability and Calcofluor (Fig. 3). Malt extract levels decreased with increasing levels of nitrogen (Fig. 4), but, as expected, levels of diastatic power and α-amylase increased (Fig. 2). Malt yields were unaffected by nitrogen rate. Table 1. P values from analysis of variance for fixed effects of barley cultivar, seeding rate and nitrogen rate on grain and malt processing variables Grain hardness Effect Barley protein Average Standard deviation Germinative energy Water sensitivity Germination index Steep-out moisture Malt yield Cultivar (C) <0.001 a <0.001 0.774 0.022 <0.001 <0.001 <0.001 <0.001 Seeding rate (S) <0.001 0.301 0.124 0.002 0.617 <0.001 <0.001 0.005 C S 0.418 0.031 0.903 0.247 0.806 0.656 0.894 0.575 Nitrogen rate (N) <0.001 0.035 <0.001 0.111 0.850 0.040 <0.001 0.023 N linear <0.001 0.819 <0.001 0.795 0.994 0.011 <0.001 0.031 N quadratic <0.001 0.428 0.040 0.007 0.302 0.934 0.093 0.011 C N 0.335 0.448 0.914 0.826 0.968 0.788 0.360 0.346 S N 0.779 0.903 0.977 0.400 0.781 0.688 0.277 0.908 C S N 0.662 0.976 0.996 0.660 0.510 0.269 0.365 0.330 Environment interaction b 15 24 18 18 25 4 7 1 a Significant effects (P < 0.05) indicated in bold. b Variance associated with the effects of environment by treatment (fixed effects) interaction expressed as a percentage of the sum of the total variance associated with the effect of environment. All environment interactions were significant at P < 0.01. Table 2. Effect P values from analysis of variance for fixed effects of barley cultivar, seeding rate and nitrogen rate on malt quality variables Malt extract Soluble protein Kolbach index Wort β-glucan Diastatic power α-amylase Friability modification Calcofluor modification Calcofluor homogeneity Cultivar (C) 0.010 a <0.001 <0.001 0.678 <0.001 <0.001 <0.001 <0.001 0.222 Seeding rate (S) 0.009 0.668 <0.001 <0.001 <0.001 0.004 <0.001 <0.001 <0.001 C S 0.156 0.787 0.913 0.011 0.929 0.389 0.179 0.537 0.934 Nitrogen rate (N) <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 N linear <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 N quadratic <0.001 0.008 0.006 <0.001 <0.001 0.329 <0.001 0.399 0.549 C N 0.018 0.657 0.958 0.953 0.692 0.425 <0.001 0.926 0.908 S N 0.538 0.977 0.994 0.956 0.531 0.812 0.262 0.951 0.418 C S N 0.335 0.109 0.289 0.806 0.175 0.843 0.843 0.893 0.933 Environment interaction b 14 14 16 9 10 14 18 ND c ND 2674 a,b See Table 1. c Not determined. wileyonlinelibrary.com/jsfa c 2012 Her Majesty the Queen in Right of Canada J Sci Food Agric 2012; 92: 2672 2678

Effects of seeding rate, nitrogen rate and cultivar on barley malt quality www.soci.org Table 3. Parameter Effect of seeding rate on grain and malt quality variables 200 seeds m 2 400 seeds m 2 LSD (P < 0.05) Barley protein (g kg 1 DM) a 114 b 111 1.2 Average grain hardness (SKCS units) 51.5 51.1 0.79 Standard deviation grain hardness 14.14 14.27 0.17 Germinative energy (%) 97.9 98.3 0.30 Water sensitivity (%) c 93.1 92.8 1.05 Germination index 6.9 7.3 0.09 Steep-out moisture (g kg 1 ) 456 459 0.9 Malt yield (g kg 1 ) 921 920 0.84 Fine grind malt extract (g kg 1 DM) 805 806 1.0 Soluble protein (g kg 1 DM) 48.2 48.3 0.4 Kolbach index (%) 42.0 42.9 0.36 Wort β-glucan (mg L 1 ) 143 112 8.07 Diastatic power ( L) d 133 129 2.01 α-amylase (DU) e 57.4 58.2 0.56 Friability modification (%) 78.7 81.6 1.00 Calcofluor modification (%) 92.1 93.9 0.62 Calcofluor homogeneity (%) 83.5 85.4 0.78 a DM, dry matter. b Pair means in bold differ significantly (P < 0.05). c Presented as % germination with 8 ml of water. d L, degrees Lintner. 15 e DU, dextrinising units. 15 Effect of cultivar Several aspects of barley and malt quality were affected by cultivar (Tables 1 and 2). CDC Copeland had lower protein than AC Metcalfe, 109 vs 116 g kg 1 DM, slightly better germinative energy, 98.2 vs 97.9%, and less water sensitivity, 94.2 vs 91.5% (Table 4). AC Metcalfe kernels initiated germination more rapidly, as indicated by higher germination indices, 7.37 vs 6.88, which contributed to higher steep-out moistures, 459 vs 456 g kg 1, even though AC Metcalfe had harder kernels, 56.6 vs 46.1 SKCS units. The endosperm of the two cultivars modified differently. Levels of wort β-glucan were similar on average between cultivars, but AC Metcalfe had better Calcofluor modification values, 93.8 vs 92.2%, a measurement of the pattern of β-glucan breakdown in contrast to absolute levels of this undesirable compound in wort. AC Metcalfe produced more soluble protein, 48.9 vs 47.6 g kg 1 DM, but the percentage of protein breakdown, as indicated by Kolbachindex,washigherinCDCCopeland,42.9vs42.1%,andCDC Copeland also had higher friability, 83.9 vs 76.3%, a parameter also associated with protein modification. AC Metcalfe malt produced a slightly higher level of malt extract, 806 vs 805 g kg 1 DM, but CDC Copeland had higher malt yield, 922 vs 919 g kg 1, or reduced losses during malting. AC Metcalfe had higher levels of diastatic power and α-amylase. An interaction of seeding rate by cultivar was observed for levels of wort β-glucan (Table 2). CDC Copeland had more wort β-glucan than AC Metcalfe at the low seeding rate but less wort β-glucan than AC Metcalfe at the high seeding rate (Table 5). Grain hardness also had a seeding rate by cultivar interaction, although grain protein was not affected. The hardness of AC Metcalfe kernels remained constant between seeding rates, while CDC Copeland had significantly softer kernels and higher malt extract at the higher seeding rate. An interaction between nitrogen rate and cultivar was detected for two aspects of malt quality (Table 2). The negative effect of nitrogen fertilisation on friability was more pronounced with AC Metcalfe than with CDC Copeland (Fig. 3). As a result, AC Metcalfe s advantage of higher levels of malt extract was slowly eroded with increasing rates of nitrogen fertilisation (Fig. 4). DISCUSSION Environment was a significant source of variance for most barley and malt parameters analysed (data not shown). However, the objective of the study was to determine how seeding rate, nitrogen rate and cultivar affected malt quality, and not effects of environment on quality, which have been previously Figure 1. Effect of nitrogen rate on (A) grain protein, (B) germinative energy, (C) germination index and (D) steep-out moisture. Symbols represent data averaged over all environments. Lines were derived from regression coefficients as calculated with mixed model analysis. 2675 J Sci Food Agric 2012; 92: 2672 2678 c 2012 Her Majesty the Queen in Right of Canada wileyonlinelibrary.com/jsfa

www.soci.org MJ Edney et al. Figure 2. Effect of nitrogen rate on (A) wort β-glucan, (B) α-amylase, (C) Kolbach index and (D) diastatic power.symbols represent data averaged over all environments. Lines were derived from regression coefficients as calculated with mixed model analysis. Figure 3. Effect of nitrogen on (A) friability modification, (B) friability homogeneity, (C) Calcofluor modification and (D) Calcofluor homogeneity. Symbols represent data averaged over all environments. Lines were derived from regression coefficients as calculated with mixed model analysis. 2676 documented. 1,2 Consistency of treatment effects across environments (locations/years) was a concern and was indicated by the percentage of environment variance associated with the sum of environment/treatment interactions (Tables 1 and 2). The percentage was low (<10%) for the malt processing parameters, germination index, steep-out moisture and malt yield but higher (15 20%) for barley parameters such as grain protein, grain hardness, germinative energy and water sensitivity (Table 1). The percentage for malt quality parameters averaged nearly 16%. All percentages of environment/treatment interaction variance were still relatively low compared with the overall variances associated with the environment, indicating that treatment effects were relatively consistent across environments. Seeding rates and nitrogen fertilisation rates are known to affect barley quality, but effects on malt quality have often only been surmised from the barley data. However, the endosperm modification so necessary for good malt quality is a complex process dependent on all aspects of barley quality, including their interactions. Malt quality, therefore, is best determined directly if wileyonlinelibrary.com/jsfa c 2012 Her Majesty the Queen in Right of Canada J Sci Food Agric 2012; 92: 2672 2678

Effects of seeding rate, nitrogen rate and cultivar on barley malt quality www.soci.org Figure 4. Effect of nitrogen rate on malt extract. Symbols represent data averaged over all environments. Lines were derived from regression coefficients as calculated with mixed model analysis. Table 4. Parameter Effect of barley cultivar on grain and malt quality variables AC Metcalfe CDC Copeland LSD (P < 0.05) Barley protein (g kg 1 DM) a 116 b 109 1.2 Average grain hardness 56.6 46.1 0.79 (SKCS units) Standard deviation grain 14.19 14.22 0.17 hardness Germinative energy (%) 97.9 98.2 0.30 Water sensitivity (%) c 91.5 94.2 1.05 Germination index 7.37 6.88 0.08 Steep-out moisture (g kg 1 ) 459 456 0.9 Malt yield (g kg 1 ) 919 922 0.8 Fine grind malt extract 806 805 1.0 (g kg 1 DM) Soluble protein (g kg 1 DM) 48.9 47.6 0.4 Kolbach index (%) 42.1 42.9 0.36 Wort β-glucan (mg L 1 ) 127 128 8.1 Diastatic power ( L) d 146 116 2.00 α-amylase (DU) e 66.2 49.4 0.56 Friability modification (%) 76.3 83.9 1.03 Calcofluor modification (%) 93.8 92.2 0.62 Calcofluor homogeneity (%) 84.2 84.7 0.78 a e See Table 3. effects of crop management on end use quality are to be clearly demonstrated. Higher seeding rates for malting barley have not been recommended owing to poor barley plumpness and a perceived negative effect on malt extract. 9 The slight decrease in grain protein was not considered sufficient to overcome the negative effects of small kernels. However, recent studies showed a reduction in variability of kernel size as seeding rates increased, which improved malt quality. 12 In the present study, higher seeding rates produced barley with smaller but more uniformly sized kernels 10 that processed differently, with a quicker start to germination and higher steep-out moistures. This was expected, as smaller kernels are known to take water up at a greater rate than larger kernels. 3 The uniform kernel size also allowed barley grown at higher seeding rates to modify more uniformly, as indicated by better Calcofluor homogeneity. The result was higher malt extract levels, despite the smaller kernels, and lower levels of wort β-glucan. Therefore relatively high seeding rates can be used to grow malting barley with excellent malt potential. The necessity to reduce the rate of nitrogen fertilisation to achieve required grain protein levels, which compromises on yield, can be overcome with specific barley cultivars. Cultivars have been developed that produce lower grain protein, 2 but the ability to maintain acceptable malt quality under high fertilisation rates is not well understood. In the present study, higher rates of nitrogen led to increased grain protein and a consequent reduction in malt extract and endosperm modification. The two cultivars tested did behave differently, although differences were often subtle, as both cultivars had been developed to produce the high but narrow malt quality demanded by industry. CDC Copeland showed greater ability to resist the negative effects of increasing rates of nitrogen, as indicated by smaller reductions in friability and malt extract compared with AC Metcalfe. The two cultivars differed in their manner of endosperm modification. CDC Copeland showed better protein modification, as indicated by the higher levels of friability at all nitrogen levels. The enhanced protein modification of CDC Copeland was expected owing to specific breeding for high friability with this cultivar (BL Harvey, breeder of CDC Copeland, personal communication). AC Metcalfe showed better Calcofluor modification, indicating more complete modification of cell walls. In contrast to protein modification, Calcofluor showed no cultivar/nitrogen interaction, suggesting different constraints to cell wall versus protein breakdown. Kernel size appeared to be a greater restriction to cell wall breakdown, as indicated by the cultivar/seeding rate interaction on wort β-glucan. Breeding barley cultivars with better ability to break protein down during malting, and not necessarily β-glucan, could lead to cultivars with good malt quality potential, even at higher rates of nitrogen fertilisation. In conclusion, higher seeding rates resulted in barley that produced malt with good quality owing to smaller, more uniformly sized kernels and more complete endosperm modification. Better modification led to less wort β-glucan and good levels of malt extract despite the smaller kernels. Varieties responded differently to changing crop management, which affected malt quality owing to differences in protein modification. Negative effects on malt Table 5. Effects of interaction between barley cultivar and seeding rate (low, 200 seeds m 2 ; high, 400 seeds m 2 ) on kernel diameter, kernel hardness, wort β-glucan and malt extract Average kernel diameter (mm) Average grain hardness (SKCS units) Wort β-glucan (mg L 1 ) Maltextract(gkg 1 DM) Cultivar Low rate High rate Low rate High rate Low rate High rate Low rate High rate AC Metcalfe 2.56 a 2.53 56.3 56.8 136 117 806 806 CDC Copeland 2.56 2.50 46.8 45.5 149 108 804 805 a Seeding rate pairs in bold differ significantly (P < 0.05). 2677 J Sci Food Agric 2012; 92: 2672 2678 c 2012 Her Majesty the Queen in Right of Canada wileyonlinelibrary.com/jsfa

www.soci.org MJ Edney et al. quality of higher nitrogen rates can be reduced by breeding and growing cultivars that achieve better protein modification during malting. ACKNOWLEDGMENTS This research was funded by grants from the Alberta Barley Commission, the Canadian Wheat Board, RAHR Malting and the Matching Investment Initiative of Agriculture and Agri-Food Canada. The excellent technical assistance of the following is gratefully appreciated: Len Dushnicky, Jason Herman, Justin Highmoor, April Johnson, Anders Leung, Shirley Lowe, Marnie MacLean, Aaron MacLeod, Ashley Orr, Shawn Parsons, Mary Richardson and Mike Svistovski. REFERENCES 1 Therrien MC, Carmichael CA, Noll JS and Grant CA, Effect of fertilizer management, genotype, and environmental factors on some malting quality characteristics in barley. Can J Plant Sci 74:545 547 (1994). 2 Weston DT, Horsley RD, Schwarz PB and Goos RJ, Nitrogen and planting date effects on low-protein spring barley. Agron J 85:1170 1174 (1993). 3 Briggs DE, Hough JS, Stevens R and Young TW, Malting and Brewing Science, Vol. 1, Malt and Sweet Wort. Chapman and Hall, London, pp. 42, 80, 121 (1981). 4 Palmer GH, Cereals in malting and brewing, in Cereal Science and Technology,ed.byPalmer GH.AberdeenUniversityPress,Aberdeen, pp. 147 165 (1989). 5 Crabb D and Bathgate GN, The influence of β-glucanase on efficiency ofwortseparation. JInstBrew79:519 524 (1973). 6 Edney MJ, Eglinton JK, Collins HM, Barr AR, Legge WG and Rossnagel BG, Importance of endosperm modification for malt wortfermentability. JInstBrew113:228 238 (2007). 7 Duijnhouwer IDC, Grashoff C and Angelino SAGF, Kernel filling and malting barley quality. Proc. European Brewing Convention, Oslo, pp. 121 128 (1993). 8 SheehyM,MarafiotiA,KrawecC,Löfqvist B and Stewart D, Active stack management actively undoing malt quality? Proc. 14th Australian Barley Technical Symposium, Brisbane, (2009). 9 McKenzie RH, Middleton AB and Bremer E, Fertilization, seeding date, and seeding rate for malting barley yield and quality in southern Alberta. Can J Plant Sci 85:603 614 (2005). 10 O Donovan JT, Turkington TK, Edney MJ, Clayton GW, McKenzie RH, Juskiw PE, et al, Seeding rate, nitrogen rate and cultivar effects on malting barley production. Agron J 103:709 716 (2011). 11 Wade A and Froment MA, Barley Quality and Grain Size Homogeneity for Malting, Vol. I, Agronomic Effects on Cultivars. Project Report #320. Home Grown Cereals Authority, Kenilworth (2003). 12 Muller R, Barley Quality and Grain Size Homogeneity for Malting, Vol. II, Assessment and Control. Project Report #320. Home Grown Cereals Authority, Kenilworth (2003). 13 Legge WG, Metcalfe DR, Haber S, Harder DE, Noll JS, Tekauz A, et al, ACMetcalfebarley. Can J Plant Sci 83:381 384 (2003). 14 CDC Copeland. [Online]. Available: http://www.inspection.gc.ca/ english/plaveg/pbrpov/cropreport/bar/app00002297e.shtml [28 February 2012]. 15 ASBC, Methods of Analysis (9th edn). American Society of Brewing Chemists, St Paul, MN (2004). 16 Riis P and Bang-Olsen KB, Germination profile a new term in barley analyses. Proc. European Brewing Convention, Lisbon, pp. 101 108 (1991). 17 Osborne BG and Anderssen RS, Single-kernel characterization principles and applications. Cereal Chem 80:613 622 (2003). 18 Aastrup S, Gibbons GC and Munk L, A rapid method for estimating the degree of modification in barley malt by measurement of cell wall breakdown. Carlsberg Res Commun 46:77 86 (1981). 19 SAS, SAS User Manual 9.2. SAS Institute, Cary, NC (2008). 20 Yang RC, Towards understanding and use of mixed model analysis of agriculturalexperiments. Can J Plant Sci 90:605 627 (2010). 2678 wileyonlinelibrary.com/jsfa c 2012 Her Majesty the Queen in Right of Canada J Sci Food Agric 2012; 92: 2672 2678