Printer Friendly

Variation in grain functional quality for soft winter wheat.

Wheat grain quality is determined by the chemical and physical constituents of the kernels themselves. Those chemical constituents are assembled following anthesis in a manner dependent on the genetic make up of the cultivar and the growing environment. Past research has indicated that genotype, environment and genotype x environment interaction all influence the milling and baking quality of classes of wheat studied to date (Baenziger et al., 1985; Basset et al, 1989; Bhatt and Derera, 1975; Lukow and McVetty, 1989; Peterson et al., 1992).

Cultivars are classified in the US on the basis of kernel hardness (soft or hard), bran color (red or white), and growth habit (spring or winter). Each grain class is expected to function as an ingredient in a broad range of class-specific end-use products. The products made from soft wheat are numerous and include cakes, cookies, crackers, noodles, pastries, and pie crusts (Faridi et al., 1994). Quality criteria evaluated during cultivar development include milling yield, protein content, kernel hardness, rheological properties, and volume and texture of actual end-use products. The mixograph is one of the common analytical tools used to assess rheological properties of hard wheat grain. (AACC, 1983; Hoseney, 1994). The mixograph measures the time to reach dough development and overmixing, and the dough's ability to resist overmixing. Mixograph results largely dependent on the quality and quantity of the gluten protein present in a flour sample (Finney and Shogren, 1972).

There is interest in the milling and baking community in soft wheat flours which exhibit dough mixing properties more commonly associated with hard wheat. The traditional solution to the need for soft wheat flours with hard wheat rheological properties has been the blending of hard and soft wheat grain. In the eastern United States, this approach forces millers to import hard wheat from the western regions of the country. An alternative approach would be to identify soft wheat cultivars with more appropriate rheological properties.
Table 1. Environment means values of quality traits for five
cultivars.

Year-county   FY(*)         PC       MPT     MXH      CH      CD

                      %              min      cm          mm

93-Ingham     63.3b(**)    8.32cd   3.38cd   4.98e   0.96   7.70cd
93-Lenawee1   65.9a        8.24cd   3.37cd   4.97e   0.96   7.79bc
93-Lenawee2   63.9b        9.12b    3.58cd   5.22d   1.00   7.68cd
93-Saginaw    65.5a        7.79e    3.23d    4.70f   0.93   7.92a
93-Sanilac    65.7a        8.15d    3.31d    4.93e   0.94   7.88ab
94-Huron      63.9b        9.46b    3.69bc   5.48c   0.97   7.64d
94-Ingham     63.2b        9.88a    3.96ab   5.79b   0.96   7.69cd
94-Lenawee    61.6c        8.51c    3.43cd   5.18d   0.96   7.63d
94-Sanilac    64.5ab      10.02a    4.05a    6.19a   0.97   7.50e
Range          4.31        2.23     0.82     1.49    0.04   0.42
Mean          64.2         8.76     3.56     5.27    0.96   7.73

* Values within a column followed by the same letter are not
significantly different at P [less than] 0.05 based on Duncan's
Multiple Range test.

** FY, % flour yield; PC, % protein content; CD, cookie diameter;
CH, cookie height; MPT, mixograph peak time; MXH, mixograph peak
height.




The purpose of this research was to analyze variation in yield test-derived soft wheat grain samples for traditional (flour yield and protein content) and non-traditional (mixograph and wire-cut formulation cookie test) quality characteristics.

MATERIAL AND METHODS

Grain samples were collected from multi-location yield trials conducted in Michigan by Michigan State University in 1993 and 1994. Individual locations were arranged as three replication lattices. Over 60 cultivars and lines were included in the trials. Samples were taken from all replications at each location for five cultivars. Two soft white winter wheat cultivars: Augusta (CItr 17831, released by Michigan State University, 1979) and Chelsea (P1506408, released by Michigan State University, 1993); and three soft red winter wheat cultivars: Freedom (PI 562382, released by Ohio State University, 1991), Mendon (PI 584472, released by Michigan State University, 1994), and Twain (PI 583811, released by AgriPro Seed, inc., Indiana, 1987), were tested. Test locations changed with years and are identified by county and year (Table 1). Fall nitrogen regime varied with local farming practices. Spring nitrogen rates of 90 or 100 kg of actual N/ha were applied as urea in late winter or early spring. Seeding rates were 4.94 to 5.19 million seeds/ha. Plots were 3.0 to 3.7 m long with seven rows spaced 0.21 m apart. Each location-year combination was considered a distinct environment.

Combine harvested grain samples from each plot were lightly aspirated to remove chaff and other debris and sieved [TABULAR DATA FOR TABLE 2 OMITTED] through a 1.79- by 12.7-mm slot dockage tester. Flour yield was determined by milling 300 g samples at 140 [+ or -] 10 g [kg.sup.-1] grain moisture content on a Brabender Quadromat Jr. experimental mill (C.W. Brabender Instruments, Inc., South Hackensack, NJ; AACC approved method 26-50, 1983). Protein was determined according to the AACC approved Kjeldahl method 46-13 (AACC, 1983) and expressed on a 140 g [kg.sup.-1] moisture basis. Flour mixograph properties were measured using a mixograph with a 35 g bowl (National Manufacturing Co., Lincoln, NE; AACC approved method 54-40, 1983). Mixographs were evaluated for both the time to reach the peak of the curve and the height of the peak. Cookie baking quality was determined using the wire-cut formulation of the AACC approved method 10-54 (AACC, 1992). Milling, mixograph, and baking properties were determined in a constant environment baking lab at 19.5 [+ or -] 0.4 [degrees] C and 42.0 [+ or -] 1.0% relative humidity.

Mean squares were generated and tests of significance were conducted using the Statistical Analysis System's PROC GLM under the assumptions that environment effects were random and cultivar effects were fixed (SAS, 1988). That same set of data was used to estimate the percentages of the total variance originating from individual sources. To that end, variance components were calculated using PROC VARCOMP under the assumptions that both environment and cultivar effects were random (SAS, 1988). Principal component analysis and Huhn's nonparametric stability statistics were carried out as described in Hazen and Ward (1997).

RESULTS

Analysis of variance revealed that cultivar, environment, and CEI effects were significant (P [less than] 0.01) for flour yield, protein content, cookie diameter, and mixograph peak height (Table 2). Only cultivar significantly (P [less than] 0.05) affected cookie height. Mixograph peak time was significantly (P [less than] 0.01) affected x cultivar and environment, but not CEI. Variation among replications at a location was significant for protein content, mixograph peak height, and cookie diameter. 'Mendon' did not form dough, as evidenced by a flat mixograph, and was therefore not included in the analysis of mixograph (data not shown).
Table 3. Means values for quality traits of cultivars grown in nine
environments.

Cultivar     FY(*)         PC       MPT       MXH      CH       CD

                      %             min                 cm

Augusta      61.5b(**)    8.41c     3.75a    5.15c    0.93b    7.83a
Chelsea      64.7a        8.14d     3.65a    5.30b    0.93b    7.89a
Freedom      65.4a        8.74b     3.82a    5.01d    0.96ab   7.64b
Mendon       64.3a        8.22c       -        -      0.99a    7.61b
Twain        64.7a       10.18a     3.08b    5.68a    0.98a    7.67b
Range         3.89        2.04      0.74     0.67     0.06     0.28
Mean         64.2         8.76      3.56     5.27     0.96     7.73

* Values within a column followed by the same letter are not
significantly different at P [less than] 0.05 based on Duncan's
Multiple Range test.

** FY, % flour yield; PC, % protein content; CD, cookie diameter,
CH, cookie height; MPT, mixograph peak time; MXH, mixograph peak
height.




Percentage of the total variance contributed by each source of variation for each trait was estimated through the calculation of variance components (Table 2). Variance components attributed to cultivar were greater than those of environment and CEI for all traits except mixograph peak height. Environment and CEI had nearly equal variation for flour yield. Error variance was the largest term for flour yield, cookie diameter, cookie height, and mixograph peak time. Error contributed least to the variance of mixograph peak height. Cultivar x environment interaction variance was much less than cultivar variance and less than error variance for protein content.

Cultivar means for protein content ranged from 81 to 84 [[micro]gram] [g.sup.-1] with the exception of the cultivar 'Twain' which had a mean protein content of 102 [[micro]gram] [g.sup.-1] (Table 3). Twain also had unique rheological properties compared to the other cultivars. It's mean mixograph peak time and mixograph peak height were the shortest and highest, respectively. 'Augusta', 'Chelsea', and 'Freedom' had similar mixograph peak times, but differed in mixograph peak height. The range for mixograph peak height for environments was greater than twice that of cultivars (Tables 1 and 3). The two white seeded cultivars, Augusta and Chelsea, had the largest cookie diameter and the smallest cookie height.

The range of average protein content for environments was nearly equal to that of cultivars (Table 1). [TABULAR DATA FOR TABLE 4 OMITTED] Environment means for protein content ranged from 78 [[micro]gram] [g.sup.-1] in Saginaw county in 1993 to 100 [[micro]gram] [g.sup.-1] in Sanilac county in 1994. There was a broader range of cookie diameter values among environments than among cultivars, but there were no significant differences in cookie height among environments (Tables 1, 2, and 4).

Correlation coefficients (n = 9) for pairs of dependent variables were calculated for each cultivar using environment means. Mixograph peak height versus protein content was the only trait combination with a significant (P [less than] 0.05) correlation for all cultivars [r = 0.88 (Freedom), 0.95 (Twain), 0.95 (Chelsea), and 0.98 (Augusta)]. Some significant (P [less than] 0.05) within-cultivar r values were observed for flour yield versus protein content [r = -0.53 (Freedom) and -0.73 (Chelsea)], flour yield versus cookie diameter [r = 0.86 (Chelsea)], flour yield versus mixograph peak time [r = -0.52 (Freedom) and -0.72 (Chelsea)], flour yield versus mixograph peak height [r = -0.71 (Freedom) and -0.78 (Chelsea)], cookie diameter versus cookie height [r = -0.65 (Mendon), -0.68 (Twain), -0.71 (Chelsea), and -0.72 (Freedom)], and cookie diameter versus mixograph peak time [r = -0.57 (Freedom) and -0.65 (Chelsea)].

Correlation analysis of the combined cultivar results (n = 45) showed significant (P [less than] 0.05) r values for protein content versus cookie diameter (r = -0.37), mixograph peak time (r = -0.33), and mixograph peak height (r = 0.90). Significant combined results r values were also found for flour yield versus mixograph peak time (r = -0.37), cookie diameter versus cookie height (r = - 0.65), and cookie diameter versus mixograph peak height (r = -0.39).

Principal component analysis of the CEI effects are presented in Fig. 1 through 4. Cultivar and environment principal components are expressed as vectors on a biplot and are to be considered separately in each graph. Vectors have two properties: direction and length. Points of the same type (cultivars or environments) that are close to each other have similar CEI values across the other factors (cultivars or environments). Points with similar direction behaved relatively alike whereas points with opposite direction behaved conversely. Points whose vector directions are perpendicular behaved independently of one another, i.e., there is no negative or positive relationship between the two. For cultivars, five parameters (cultivars) are measured for nine variables (environments). The converse is true when discussing environments, i.e., nine environments are analyzed for five variables (cultivars). For instance, Twain and Freedom are close to each other in protein content and mixograph height [ILLUSTRATION FOR FIGURE 2 AND 4 OMITTED]. This means that the CEI effects across environments or the differential response to environment for these two cultivars are similar for protein content and mixograph height. It is also true that Chelsea is either independent of or opposite to all other cultivars for CEI pattern for all traits [ILLUSTRATION FOR FIGURE 1-4 OMITTED]. Freedom and Meudon are largely independent of each other for each trait.

The cultivar arrangement in the biplots for protein content and mixograph peak height are very similar in that the relationship among cultivars is alike [ILLUSTRATION FOR FIGURE 3 AND 4 OMITTED]. These two traits have a significant linear correlation (r = 0.90). The relationship between protein content and mixograph peak height biplots does not exist when considering cultivars as the variables and environment as the parameter [ILLUSTRATION FOR FIGURE 3 AND 4 OMITTED].

There is little or no pattern for counties or years for any traits [ILLUSTRATION FOR FIGURE 1-4 OMITTED]. It appears as though the differences in environments that contribute to the interaction are unrelated to spatial or temporal characteristics of each environment. Ingham and Huron counties in 1994 cluster on each biplot. Interaction effects in these two environments were therefore similar. The interaction effects in Lenawee 1993 were contrary to those that occurred in Ingham and Huron 1994 in all cases. Other Lenawee and Ingham fields did not have a consistent relationship across traits.

Relative rank stability was estimated using Huhn's nonparametric stability statistic (Tables 4 and 5). The sum of the test statistic, [Z.sub.i], has a [[Chi].sup.2] distribution with degrees of freedom equal to the number of cultivars tested, 5. That sum tests the null hypothesis that there is no difference in overall cultivar rank stability. The null hypothesis was rejected for flour yield ([Z.sub.i] = 16.27) alone (Table 4). Upon observation of the [Z.sub.i] values for each cultivar, Freedom differed from the other cultivars. Freedom's relatively low [S.sub.i] value for flour yield reveals that it exhibited more stability than Augusta, Chelsea, Meudon, and Twain. Neither the sum of the stability test statistic nor the individual cultivar test statistic exceeded the critical values for protein content, mixograph peak time, mixograph peak height, cookie height or cookie diameter. Cultivars had overall rank stability differences for flour yield with Freedom exhibiting greater rank stability than other cultivars. For other quality traits, cultivars did not exhibit differences in rank stability overall, or when considering cultivars individually.

DISCUSSION

Mixograph properties are not traditional targets for selection in soft wheat breeding programs. The results reported here show a large range in mixograph peak height and peak time. The area under the mixograph peak curve is related to dough viscosity and cookie diameter in soft wheat (Morris et al., 1943). Sugar-snap cookie diameter had a strong positive correlation with mixograph peak time but was negatively correlated with mixograph peak height (Yamamoto et al., 1996). Mixograph peak height was strongly and positively related to protein content. Hard wheat has relatively high protein content and gives a well defined mixograph peak that is generally higher and has a more dramatic 'breakdown' than that of soft wheat. Mixograph peak time ranges from 2.9 to 6.1 min and 2.3 to 5.0 min for hard red winter and hard red spring wheat, respectively (Peterson et al., 1992; Lukow and McVetty, 1991). Mendon's mixograph curve was flat whereas Twain had a high peak that took on average 3.08 min to reach and exhibited a more dramatic breakdown than the other cultivars studied. It appears that theological properties of some soft winter wheat samples measured by the mixograph are similar in some instances in peak time and height to those of hard wheat (Finney et al., 1987). Soft wheat such as Twain may therefore possibly be used as a substitute ingredient for hard wheat or to expand the range of soft wheat products. Protein content was negatively correlated to cookie diameter, which is in agreement with previous studies (Gaines, 1985; Basset et al., 1989, Kaldy et al., 1993; Yamamoto et al., 1996). Mixograph peak height was also significantly related to cookie diameter, however the correlation coefficients of these two traits are not large enough to warrant substitution of one test for another.

The micro wire-cut formulation cookie method is a relatively new test compared to the sugar-snap formulation cookie test. The wire-cut formulation has a higher shortening and lower sugar content than the sugar-snap cookie formulation. The hardness and texture of the wire-cut cookie is more closely related to that of commercial products than the sugar-snap test (AACC, 1992). Cookie diameter was highly variable with all sources of variation being significant. The range of cookie height was small and significantly affected by cultivar alone. This measure of baking quality appears to be useful as a method to evaluate and improve soft winter wheat quality. Genetic improvement of both cookie width and height may be made using this method of evaluation.

Interactions between genotype and environment can diminish a breeders ability to recognize differences among breeding materials. Changes in cultivar rankings for a particular trait across environments is indicative of a severe CEI. A mild CEI often results in a change in the magnitude of the differences among cultivars [TABULAR DATA FOR TABLE 5 OMITTED] rather than a change in rank order. In instances of crossover interaction, rank order of the mean performance of cultivars is not a true representation of each cultivar's phenotypic behavior in each environment tested. Understanding the nature of the interactions and error variance leads to a more informed strategy of sampling and possibly more defined regions of cultivar production.

Cultivars may differ for stability of quality characteristics. Stability measured using regression analysis outlined by both Eberhart and Russell (1966) and Moll et al. (1978) have shown few cultivars have either a high or low response to environments for flour yield, protein content, mixograph properties, and cookie diameter (Busch et al., 1965; and McGuire and McNeal, 1974; Peterson et al., 1992; Lukow and McVetty, 1991; Basset et al., 1989; Baenziger et al., 1985). The regression coefficient from this approach is perceived by many as a measure of genotypic response to favorable conditions rather than a stability measurement (Becker and Leon, 1988). A cultivar may be adaptive to a wide range of environments while others may only perform well under certain environmental conditions. Stability was estimated here using nonparametric methods by measuring the absolute rank differences of cultivars across environments. Cultivars exhibited significant differences in rank stability for flour yield alone. Further analysis of the individual test statistics showed that the cultivar Freedom in this instance was significantly more stable than the other cultivars. Freedom's rank across environments was relatively constant. Freedom has the 1BL/1RS wheat rye translocation (Berzonsky et al., 1991). Studies have shown that this rye gene complex is associated with increased yields due to increased kernel weight in lower yielding environments (Rajaram et al., 1983; Moreno-Sevilla et al., 1996). In every other instance, no cultivar significantly exceeded or fell below the expected amount of rank change based on [[Chi].sup.2] distribution. Thus, stability issues are not important to consider for the cultivars tested.

It is possible that cultivars change rank in response to particular environmental factors associated with regions or management practices (Allard and Bradshaw, 1964). It may be prudent to divide a growing region into smaller regions where different cultivar recommendations are made (Horner and Frey, 1957). Here, principal component analysis showed environmental factors that elicit a differential response across genotypes are not associated with year or geographical proximity. This lack of association shows that the magnitude of interaction effects for the traits evaluated is unpredictable. It would therefore not be useful to divide Michigan into micro growing regions.

Analysis of multiple environments within a year is the most efficient manner to accurately estimate the relative differences between cultivars in non-crossover interaction situations. Quality analysis is both labor and technically intensive as well as expensive. A check cultivar which has been analyzed over many seasons should be used as a benchmark in quality evaluation. Sampling strategies based on these data should account for the significant effects of CEI and the magnitude of the error variance. In order to obtain optimum accuracy, it is necessary to sample multiple replications from each location. Random selection of locations within a year is sufficient. However, the analysis of multiple replications from each site is often not an option because of resource constraints. A reasonable degree of accuracy can be obtained by evaluating bulked replications from a few locations per year. When purchasing grain, knowledge of the cultivar is useful in predicting quality relative to other cultivars. Knowledge of local environments is of little use when attempting to predict grain quality.

Abbreviations: CEI, cultivar x environment interaction.

ACKNOWLEDGMENTS

We thank the Michigan State Millers Association for their support and appreciate the assistance of Vince Rinaldi, Brian Diers, David Glenn, and Erica Jenkins.

REFERENCES

American Association of Cereal Chemists. 1983. Approved methods of the AACC. 8th ed. AACC, St. Paul, MN.

Allard, R.W., and A.D. Bradshaw. 1964. Implications of genotype-environmental interactions in applied plant breeding. Crop Sci. 4:503-508.

Baenziger, S.P., R.L. Clements, M.S. McIntosh, W.Y. Yamazaki, T.M. Starling, D.J. Sammons, and J.W. Johnson. 1985. Effects of cultivar, environment, and their interaction and stability on milling and baking quality of soft red winter wheat. Crop Sci. 25:5-8.

Basset, L.M., R.E. Allan, and G.L. Rubenthaler. 1989. Genotype x environment interactions on soft white winter wheat quality. Agron. J. 81:955-960.

Becker, H.C., and J. Leon. 1988. Stability analysis in plant breeding. Plant Breeding. 101:1-23.

Bersonsky, W.A., R.L. Clements, and H.N. Lafever. 1991. Identification of 'Amigo' and 'Kavkaz' translocation in Ohio soft red winter wheats (Triticum aestivum L.). Theor Appl Genet 81:629634.

Bhatt, G.M., and N.F. Derera. 1975. Genotype by environment interactions for, heritabilities of, and correlations among quality traits in wheat. Euphytica 24:597-604.

Busch, R.H., W.C. Shuey, and R.C. Frohberg. 1969. Response of hard red spring wheat (Triticum aestivum L.) to environments in relation to six quality characteristics. Crop Sci. 9:813-817.

Eberhart, S.A., and W.A. Russell. 1966. Stability parameters for comparing varieties. Crop Sci. 6: 36-40.

Faridi, H., C. Gaines, and P. Finney. 1994. Soft wheat quality on production of cookies and crackers. p. 1-11. In W. Bushuk and V.F. Rasper (ed.) Wheat production, properties, and quality. Blackie Academic and Professional, Glasgow.

Finney, K.F., and M.D. Shogren. 1972. A ten-gram mixograph for determining and predicting functional properties of wheat flours. Baker's Dig. 46:32-35, 38-42, 77.

Finney, K.F., W.T. Yamazaki, V.L. Youngs, and G.L. Rubenthaler. 1987. Quality of hard, soft, and durum wheats. p. 677-741. In E.G. Heyne (ed.) Wheat and wheat improvement. CSSA and ASA, Madison, WI.

Gaines, C.S. 1985. Associations among soft wheat flour particle size, protein content, chlorine response, kernel hardness, milling quality, white layer cake volume, and sugar-snap cookie spread. Cereal Chem 62:290-292.

Hazen, S.P., and R.W. Ward. Variation in soft winter wheat characteristics measured by the single kernel characterization system. Crop Sci. 37:1079-1086 (this issue).

Horner, T.W., and K. Frey. 1957. Methods for determining natural areas for oat varietal recommendations. Agron. J. 49:313-315.

Hoseney, R.C., 1994. Principles of cereal science and technology, 2nd ed.. AACC, St. Paul, MN.

Kaldy, M.S., G.R. Kereliuk, and G.C. Kozub. 1993. Influence of gluten components and flour lipids on soft white wheat quality. Cereal Chem 70:77-88.

Lukow, O.M. and P.B.E. McVetty. 1991. Effect of cultivar and environment on quality characteristics of spring wheat. Cereal Chem. 68:597-601.

McGuire, C.F., and F.H. McNeal. 1974. Quality response of 10 hard red spring wheat cultivars to 25 environments. Crop Sci. 14:175-178.

Moreno-Sevilla, B., P.S. Baenziger, C.J. Peterson, R.A. Graybosch, and D.V. McVey. 1995. The 1BL/1RS translocation: agronomic performance of [F.sub.3]-derived lines from a winter wheat cross. Crop Sci. 35:1051-1055.

Morris, V.H., C.E. Bode, and H.K. Heizer. 1943. The use of the mixograph in evaluating quality in soft wheat varieties. Cereal Chem. 21:49-57.

Moll, R.H., C.C. Cockerham, C.W. Stuber, and W.P. Williams. 1978. Selection responses, genetic-environmental interactions, and heterosis with recurrent selection for yield in maize. Crop Sci. 18: 641-645.

Peterson, C.J., R.A. Graybosch, P.S. Baenziger, and A.W. Grombacher. 1992. Genotype and environment effects on quality characteristics of hard red winter wheat. Crop Sci. 32:98-103.

Rajaram, S., Ch.E. Mann, G. Ortiz-Ferrera, and A. Mujeeb-Kazi. 1983. Adaptation, stability and high yield potential of certain 1B/1R CIMMYT wheats. p. 613-621. In S. Sakamoto (ed.) Proc. Int. Wheat Genetics. Symp., 6th, Kyoto Univ., Kyoto, Japan. 28 Nov.-3 Dec. 1983. Plant Germplasm Inst., Kyoto Univ., Kyoto, Japan.

SAS Institute. 1988. SAS/STAT user's guide. Statistics. SAS Institute Inc., Cary,NC.

Yamamoto, H, S.T. Worthington, G. Hou, and P.K.W. Ng. 1996. Rheological properties and baking qualities of selected soft wheats grown in the United States. Cereal Chem. 73:215221.
COPYRIGHT 1997 Crop Science Society of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1997 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Hazen, S.P.; Ng, P.K.W.; Ward, R.W.
Publication:Crop Science
Date:Jul 1, 1997
Words:4182
Previous Article:Variation in soft winter wheat characteristics measured by the single kernel characterization system.
Next Article:Characterization and inheritance of adult plant stem rust resistance in durum wheat.
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters