Printer Friendly

Factors Affecting Groat Percentage in Oat.

THE OAT GROAT OR CARYOPSIS is typically covered with a lemma and palea, or hull after harvest. This hull must be removed before the groat can be processed for food purposes. The proportion of the groat (the oat without the hull) to the whole oat is known as the groat percentage. The groat percentage is important because it represents the economic yield an oat miller can derive from a given lot of oat grain. It also provides an estimate of the digestible portion of the grain, if being fed to animals.

Groat percentage, also referred to as groat proportion, caryopsis percentage, or reported as hull percentage, has long been recognized as an important indicator of oat quality (Stoa et al., 1936; Love et al., 1925; Atkins, 1943; Bartley and Weiss, 1951). Peek and Poehlman (1949) considered test weight to be a more valuable oat quality evaluation tool because hand dehulling of oats was considered too tedious. Stoa et al. (1936) suggested that early maturing oats were superior in groat percentage, and rust susceptible lines were generally high in percent hull. These conclusions were also supported by the findings of Bunch and Forsberg (1989). The studies of Bartley and Weiss (1951) indicated strong environmental effects on groat percentage and demonstrated correlations between groat percentage and yield, test weight, and kernel weight. Youngs and Shands (1974) demonstrated that tertiary kernels had a higher groat percentage than primary and secondary kernels. However, Palagyi (1983) found that genotypes with higher levels of tertiary kernels had lower groat percentage, and suggested that tertiary kernels compete with primary and secondary kernels for assimilate, preventing them from filling properly. Groat percentage is a quantitatively inherited trait with a broad sense heritability of 36 to 92% (Wesenberg and Shands, 1973; Ronald et al., 1999). Stuthman and Granger (1977) found a narrow sense heritability of 34 to 72%.

Any of three methods are commonly used to evaluate groat percentage on an experimental basis. The most basic of these is hand dehulling, where oat grains are stripped of their hulls by hand. Two mechanical methods of oat dehulling include the impact dehuller (Ganssmann and Vorwerck, 1995), and the compressed air dehuller (Kittlitz and Vetterer, 1972). The impact dehuller feeds grain into a spinning rotor. The impact of the grain with wall as it is expelled from the rotor knocks the hull from the groat. The hulls are then removed by aspiration. The compressed air dehuller uses a stream of pressurized air to apply mechanical shock to a sample of oat grain, which knocks the hull off the groat. Hulls are removed by aspiration. Compressed air dehullers have only recently been marketed commercially and are just starting to become widely used among oat research laboratories. Hand dehulling is thought to provide a ceiling value for groat percentage, because no groats are lost in mechanical treatments. However, mechanical dehullers allow for the dehulling of larger samples, which is impractical by hand dehulling. Impact dehullers are commonly used by commercial millers to dehull oat, although an additional method, known as stone hulling, may also be used by some commercial milling operations (Ganssman and Vorwerck, 1995).

Factors influencing groat percentage are not well understood. This study was initiated to understand better the mechanical and biological factors affecting groat percentage. Various mechanical manipulations were made with a compressed air dehuller to determine how these physical treatments affected groat percentage and related dehulling characteristics. Ten oat cultivars grown at three locations were dehulled by the three methods of laboratory dehulling. Digital image analysis was used to characterize the architecture of the oat kernels to determine relationships between the physical structure of the oat grain and groat percentage.

MATERIALS AND METHODS

Plant Material

The oat cultivar Dumont was used to characterize the compressed air dehulling system. The grain was grown at Fargo, ND, in the 1996 growing season. Ten oat cultivars (AC Assiniboia, Bay, Belle, Brawn, Gem, Hytest, Jerry, Jud, AC Medallion, and Youngs) were used to compare different dehulling methods. These were grown at three locations in southeastern North Dakota (Edgeley, Fargo, and Fullerton). The soil type at Fargo is Fargo silty clay (fine montmorillonitic Typic Endoquert). Soils at Edgeley are Barnes (fine-loamy, mixed Udic Haploboroll) and Cresbard (fine, Montmorillonitic Glossic Udic Natriboroll) loam complex. Soils at Fullerton are Svea (fine-loamy, mixed Pachic Udic Haplobroll) and Cresbard loam. The previous crop at Fargo was soybean [Glycine max (L.) Merr.], at Edgeley, summer fallow, and at Fullerton, corn (Zea mays L.). The experiments were planted at Edgeley and Fullerton on 13 May 1996 and were harvested at Fullerton on 14 Aug. 1996 and at Edgeley on 21 Aug. 1996. The Fargo experiment was planted 29 April 1996 and harvested 7 Aug. 1996. A seeding rate of 2.47 x [10.sup.6] [ha.sup.-1] was used for all experiments. Herbicide treatments consisted of pre-emergence application of 3.93 kg [ha.sup.-1] propchlor (2-chloro-N-isopropylacetanilide) and post-emergence application at the 3-leaf stage with a tank mix of 0.14 kg [ha.sup.-1] thifensulfuron {methyl 3 [[[[(4-methoxy-6-methyl-1,3,5 -triazin-2yl)amino]carbonyl]amino]sulfonyl]-2-thiophenecarboxylate, 0.07 kg [ha.sup.-1] tribenuron {methyl 2-[[[[N-(methoxy-6-methyl-1,3,5-triazin-2yl) methylamino] carbonyl] amino] sulfonyl] benzoate} and 0.14 kg [ha.sup.-1] clopyralid (3,6-dichloro-2-pyridinecarboxylic acid, monoethonolamine salt). Experimental units consisted of four rows spaced 0.3 m apart and 2.4 m long. The two center rows were harvested with a two-row binder and threshed with a plot thresher. Seed was cleaned with an air screen cleaner to remove chaff. Test weight of grain was measured from the mass of a standard volume, multiplied by a constant.

Dehulling

Two-gram samples were dehulled by hand. The mass of samples were measured before and after dehulling. The ratio of the groat mass to the oat mass times 100 was the groat percentage.

Prior to dehulling by mechanical means, 50-g samples were incubated in dessicators over saturated potassium carbonate solutions (Wolf et al., 1984) to equilibrate grain moisture to a uniform level of about 95 g [kg.sup.-1] water. Previous results (Kittlitz and Vetterer, 1972) indicated that grain moisture strongly affected oat-dehulling characteristics.

For the determination of the effects of mechanical conditions during dehulling on groat percentage and dehulling characteristics, 40-g samples of Dumont oat were dehulled with the compressed-air dehuller. All experiments on dehulling conditions were repeated using the cultivar Whitestone, but because results were similar, only Dumont results were presented. The compressed-air dehuller was the Codema(1) Laboratory Oat Huller (Eden Prairie, MN). Compressed air pressure was 0.414, 0.483, 0.522, 0.621, and 0.69 MPa. The dehulling times were 30, 45, 60, 75, and 90 s. The vent opening (which controlled aspiration strength) was 0, 2, 4, 6, 8, and 12 mm. Larger vent openings resulted in weaker aspiration.

For the comparison of cultivars, the compressed air dehuller was operated at 0.522 MPa for 60 s, with a vent opening of 7 mm. Crude groat percentage was defined as the ratio of the mass of material recovered in the groat recovery container to the mass of whole oats fed into the dehuller multiplied by 100. The crude groat preparation was sorted by hand into hulls remaining after dehulling, broken groats, and whole groats. The hulls remaining fraction contained both free hulls and undehulled oats. Any groat with more than 10% of its mass broken off was considered broken for sorting purposes. Corrected groat percentage was the ratio of the mass of the crude groat preparation minus the mass of the hulls remaining, to the mass of the whole oats fed into the dehuller, multiplied by 100. The milling yield was defined as the mass of whole oats necessary to obtain 100 kg of clean whole groats. Milling yield was calculated by subtracting the mass of hulls remaining and broken groats from the mass of the crude groat, dividing this by the mass of the oat starting material, and multiplying the inverse of this value by 100. Percentage hulls remaining is the ratio of the mass of hulls or undehulled oats recovered from the crude groat preparation to the mass of the crude groat preparation, multiplied by 100. Percentage broken groats is defined as the mass ratio of broken groats recovered after dehulling to the mass of the crude groat preparation multiplied by 100.

The impact dehuller was manufactured by local contract according to plans provided by the Quaker Oat Company (Barrington, IL). Its design is similar to those shown by Ganssmann and Vorwerck (1995) or by Deane and Commers (1986). Fifty-gram oat samples were passed through the dehuller, then hulls were removed by aspiration on a "clipper" dockage tester with constant fan opening settings. After aspiration, samples dehulled by the impact dehuller were sorted into whole groats, broken groats, and hulls remaining, as were the compressed air dehuller samples, and data obtained were treated as described for the compressed air dehuller.

Digital Image Analysis

Oat and groat architecture was described by digital image analysis. Samples of about 10-g were spread on a lightbox so that a photograph could be taken including every kernel in the sample. Kernels were arranged so that none were touching each other. A photograph was taken of the kernels, including a transparent ruler, for distance calibration, and a sample number label, with a 35-mm SLR optical camera. Prints of these photographs (10 x 15 cm) were scanned into a personal computer, at 236 dots per cm resolution, to create digital images. The computer package, SigmaScan (Jandel Corporation, San Rafael, CA), was used to calculate the length, width, and image area of each kernel in each photograph. Data were downloaded into a spreadsheet program, where the mean kernel lengths, widths, and image areas were calculated, along with the coefficients of variation (CV) for length, width, and image areas. The number of kernels in each image (also provided by the computer image analysis package) was divided by the mass of the sample photographed to get the mean kernel mass. The mean kernel mass was divided by the mean image area to obtain the oat mass: area ratio. Photographs of both whole oat kernels and cleaned groats were analyzed. Additional values derived from the oat and groat data included the oat length: groat length difference, oat mass: groat mass difference, and the groat mass: oat mass ratio.

Experimental Design

Effects of compressed air dehuller conditions on dehulling characteristics of Dumont oat were determined by one-way ANOVA by the Statistix computer package (Analytical Software, Tallahassee, FL) with three replicates, where dehuller conditions were considered fixed. Differences among means were compared by Tukey's Honestly Significant Difference (HSD), as described by Steel et al. (1997, p. 191-193). For the evaluation of cultivars, oats were grown in the field at three locations in a randomized complete-block design with three replicates. Effects of cultivar, location and dehulling method on groat percentage, hulls remaining after dehulling, broken groats, and milling yield were determined by a three-way analysis (ANOVA) by the Statistix computer package, where cultivar and dehulling method were considered fixed and location was considered random. Physical characteristics of oat kernels determined by digital image analysis were analyzed by two-way ANOVA, where cultivar was considered fixed and location was considered random. Differences among means were compared by HSD also calculated by Statistix computer package. Phenotypic correlations of dehulling characteristics among cultivars were initially calculated for each location by the Statistix computer package, then r values from the separate locations were pooled according to Steel et al. (1997, p. 295-296). A chi-square test (Steel et al., 1997, p. 296-297) indicated the homogeneity of all pooled r values.

RESULTS AND DISCUSSION

Influence of Dehuller Conditions on Oat Dehulling

Increased air pressure during dehulling resulted in a decrease in the crude groat percentage, a decrease in the hulls remaining after dehulling, an increase in broken groats, and an increase in the corrected groat percentage, without affecting the mill yield (Table 1). The decrease in the crude groat percentage appeared largely to be due to the decrease in the hulls remaining after dehulling. The increased air pressure appeared to allow for a greater dehulling efficiency, and improved the removal of hulls from the groats. However, the increased mechanical stress associated with the increased air pressure also caused more groat breakage. The improved dehulling efficiency at the higher air pressures appeared to be responsible for the increased corrected groat percentages. But the improved dehulling efficiency at higher air pressures was offset by the increased groat breakage, resulting in no net change in milling yield (Table 1).

Table 1. Effects of air pressure in compressed air oat huller on Dumont oat dehulling characteristics, when samples were dehulled for 60 s with a vent opening of about 7 mm (n = 3).
Air            Crude groat          Hulls     Broken
pressure   percentage([dagger])   remaining   groats

MPa                                   %

0.414              74.0             12.2        6.3
0.483              73.4              9.1        8.2
0.522              72.8              7.6        8.6
0.621              71.2              6.2       13.4
0.690              71.6              5.0       13.6
HSD                 1.1              2.0        2.3

Air          Corrected groat               Mill
pressure   percentage([dagger])   yield([double dagger])

MPa                 %                      kg

0.414              65.3                    160
0.483              68.8                    157
0.522              67.4                    158
0.621              66.9                    159
0.690              68.1                    158
HSD                 1.6                     NS


([dagger]) Crude groat percentage is calculated from the groat preparation recovered immediately after dehulling and aspiration. Corrected groat percentage is calculated from the mass of groats after hulls remaining have been removed by hand sorting.

([double dagger]) Mill yield is defined as the mass of whole oats required to yield 100 kg of whole groats.

Increased dehulling duration resulted in decreased crude groat percentage, decreased hulls remaining, and increased broken groats (Table 2). Corrected groat percentage increased with increased dehulling duration up to 60 s, but dehulling durations over 60 s resulted in decreased corrected groat percentage. Mill yield was not affected by increasing dehulling duration up to 60 s, but at dehulling durations over 60 s, milling yield increased (Table 2). The decreased crude groat percentage with increasing dehulling duration was partially due to the improved removal of hulls and partially due to the improved dehulling efficiency at longer dehulling durations. The increased mechanical stress on the groats with increasing dehulling duration accounted for the increased groat breakage. The improved dehulling efficiency with increased dehulling duration was responsible for the increase in corrected groat percentage from 30 to 60 s dehulling duration. The decrease in corrected groat percentage at dehulling durations over 60 s could have been due to increased breakage, if the pieces were then removed by aspiration. This would suggest that the estimation of groat breakage, based on broken groats recovered in the groat fraction, underestimated total breakage. This loss of broken groats by aspiration would also have been responsible for the increased milling yield at dehulling durations over 60 s.

Table 2. Effects of dehulling duration on Dumont oat dehulling characteristics using air pressure of 0.522 MPa, and a 7-mm vent opening (n = 3).
Dehulling       Crude groat          Hulls     Broken
duration    percentage([dagger])   remaining   groats

s                                %

30                  75.8             17.3        6.4
45                  72.7             11.0        9.1
60                  70.9              6.5       11.5
75                  67.7              5.4       14.3
90                  64.7              2.9       16.8
HSD                  1.5              2.0        1.8

Dehulling     Corrected groat               Mill
duration    percentage([dagger])   yield([double dagger])

s                                            kg

30                  61.2                    177
45                  63.0                    177
60                  64.6                    176
75                  62.4                    189
90                  61.2                    198
HSD                  1.1                      9


([dagger]) Crude groat percentage is calculated from the groat preparation recovered immediately after dehulling and aspiration. Corrected groat percentage is calculated from the mass of groats after hulls remaining have been removed by hand sorting.

([double dagger]) Mill yield is defined as the mass of whole oats required to yield 100 kg of whole groats.

Increasing the vent opening decreased the strength of aspiration and resulted in increased crude groat percentage, increased hulls remaining, increased broken groats, increased corrected groat percentage, and decreased mill yield (Table 3). At the smallest vent opening, aspiration was the strongest. Thus, lesser amounts of hulls were present at the stronger levels of aspiration. Assuming that changes in aspiration did not affect the amount of groat breakage, the decreased amounts of broken groats at the highest levels of aspiration were probably due to the increased removal of broken groats by aspiration. Increased crude groat percentage at larger vent openings was largely due to increased broken groats and hulls remaining in the groat preparation. Increased corrected groat percentage at larger vent openings were partially due to less broken groats being removed by aspiration, but this was not sufficient to account for all of the increase observed. Assuming that changes in aspiration alone did not affect the dehulling efficiency, then it would appear likely that whole groats, as well as hulls were being removed at the highest level of aspiration. Lower levels of aspiration would result in less groat removal and corresponded with increased corrected groat percentage. Decreased milling yields with increased vent opening also reflected the decreased loss of groats at the largest vent openings.

Table 3. Effects of vent opening on oat dehulling characteristics on the compressed air dehuller. The air pressure was 0.522 MPa, and the dehulling duration was 60 s (n = 3).
           Crude groat          Hulls     Broken
Vent   percentage([dagger])   remaining   groats

mm                            %

0              60.3              1.3       2.9
2              62.2              1.8       2.5
4              65.6              1.3       3.2
6              69.3              1.0       2.7
8              70.6              1.4       4.2
12             75.1              4.7       6.7
HSD             1.5              2.5        NS

         Corrected groat               Mill
Vent   percentage([dagger])   yield([double dagger])

mm                                     kg

0              58.4                    176
2              60.1                    171
4              63.5                    163
6              67.3                    153
8              68.2                    153
12             70.2                    153
HSD             2.3                      9


([dagger]) Crude groat percentage is calculated from the groat preparation recovered immediately after dehulling and aspiration. Corrected groat percentage is calculated from the mass of groats after hulls remaining have been removed by hand sorting.

([double dagger]) Mill yield is defined as the mass of whole oats required to yield 100 kg of whole groats.

Optimal dehulling conditions obviously represents a compromise between improved dehulling efficiency, improved hull removal, minimization of broken groats, and a minimization of whole groats removed by aspiration. The evaluations of mechanical stress and aspiration presented here should to be applicable to the impact dehuller as well, where rotor speed and number of passes through the dehuller control mechanical stress.

Cultivar Effects on Oat Dehulling Characteristics

ANOVA (not shown) indicated significant location x genotype interaction, and significant location and genotype main effects (P [is less than] 0.01) for groat percentage as determined by three methods of oat dehulling. Significant interaction is attributed to differences in ranking among genotypes at different locations and not to differences in magnitude. Method of dehulling had a significant effect (P [is less than] 0.01) and the method x genotype and method x location interactions were also significant (P [is less than] 0.01). Genotypic means of hand dehulled groat percentages, crude and corrected groat percentage derived from the impact dehuller, and the compressed air dehuller are presented in Table 4. Large differences between crude and corrected groat percentages are indicative of errors which can be introduced into groat percentage estimations if hulls are not removed by hand sorting from crude groat preparations. Larger differences between crude and corrected groat percentage with the impact dehuller than the compressed air dehuller indicated that aspiration was stronger with the compressed air dehuller. Higher values of groat percentage were derived from hand dehulling for all cultivars. Groat percentages derived from the impact dehuller were higher than those derived from the air pressure dehuller. Larger corrected groat percentage values from the impact dehuller suggest that the impact dehuller applied more physical stress on the oat kernels and achieved a greater dehulling efficiency over the air pressure dehuller. Larger ranges in groat percentage were found with mechanical dehulling than with hand dehulling (Table 4).
Table 4. Groat percentages of 10 oat cultivars as determined by
different methods (n = 9).

                           Groat percentage

                                 Impact dehuller
                       Hand
Cultivar             dehulling   Cruder([dagger])

AC Assiniboia
 ([double dagger])     77.3            82.5
Bay                    73.9            79.3
Belle                  78.5            83.0
Brawn                  73.6            80.4
Gem                    73.8            79.2
Hytest                 77.4            82.0
Jerry                  74.3            80.5
Jud                    74.1            81.2
AC Medallion           77.2            82.8
Youngs                 73.8            82.8
HSD (0.05)              2.4             2.4

                                            Compressed air
                                              dehuller

Cultivar             Corrected([dagger])   Crude([dagger])

AC Assiniboia
 ([double dagger])          74.9                 72.0
Bay                         70.4                 64.5
Belle                       76.6                 71.8
Brawn                       71.1                 68.8
Gem                         72.1                 68.0
Hytest                      76.3                 72.1
Jerry                       72.0                 68.6
Jud                         70.4                 69.9
AC Medallion                74.3                 71.6
Youngs                      72.4                 70.4
HSD (0.05)                   2.4                  2.4

Cultivar             Corrected([dagger])

AC Assiniboia
 ([double dagger])          68.3
Bay                         63.5
Belle                       71.3
Brawn                       63.6
Gem                         64.8
Hytest                      71.6
Jerry                       67.7
Jud                         65.3
AC Medallion                70.6
Youngs                      66.3
HSD (0.05)                   2.4


([dagger]) Crude groat percentage is calculated from the groat preparation recovered immediately after dehulling and aspiration. Corrected groat percentage is calculated from the mass of groats after hulls remaining have been removed by hand sorting.

([double dagger]) Critical value for comparing values in the same row is 0.5.

The cultivars Jud and Bay ranked higher in groat percentage by hand dehulling than by mechanical dehulling. These small-groated cultivars may have lost significant amounts of groats to aspiration. Conversely, Youngs oat ranked higher in groat percentage by mechanical dehulling than by hand dehulling, suggesting this large-groated cultivar had fewer groats lost to aspiration.

Certain cultivars ranked higher in the crude groat percentage than in the corrected groat percentage, as determined by the impact dehuller. These included Jud, AC Medallion, and Youngs. The hulls of these cultivars appeared to be more difficult to separate from the groats. Because aspiration removes material according to mass, the hulls of these cultivars may have been heavier than the others. Rankings of crude and corrected groat percentage were more similar by the compressed air dehuller than by the impact dehuller. Cultivars that ranked lower with the correction for groat percentage by the air pressure dehuller included AC Assiniboia, Youngs, Brawn, and Jud. Cultivars that ranked higher included Belle, AC Medallion, Jerry, and Gem. These changes in ranking were due to the relative amounts of hulls remaining in the crude groat preparation. Those cultivars with small amounts of hulls remaining advanced in the rankings and those with large amounts of hulls remaining declined in ranking.

Analysis of Hulls Remaining and Groat Breakage after Mechanical Dehulling

ANOVA (not shown) indicated significant genotype main effect and genotype x location interactions for hulls remaining after dehulling (P [is less than] 0.01). The location effect was not significant (P [is greater than] 0.05). The significant cultivar x location interaction appeared to be due to differences in ranking of cultivars at different locations, as well as differences in magnitude. Four cultivars: Gem, Assiniboia, Brawn, and Youngs, had higher hulls remaining at the Fullerton location than at the other locations. These were also the cultivars with the largest kernels. Failure of the larger groats to fill completely at this location may have contributed to the retention of hulls. ANOVA (not shown) also indicated that method of dehulling had a significant effect (P [is less than] 0.01) on hulls remaining after dehulling. More hulls remained with the groats after impact dehulling than after compressed air dehulling (Table 5). This indicated that the aspiration used after the impact dehulling was not as strong as that used by the compressed air dehuller. More hulls remained with the cultivars Brawn, Jud, and Youngs by either method. Belle and Hytest had the lowest amount of hulls remaining by either method. Because mass of hulls is likely to have influenced the efficiency of their removal by aspiration, it appears likely that those cultivars with greater hulls remaining also had heavier hulls.

Table 5. A comparison of the impact dehuller and the compressed air dehuller for hulls remaining in the groats after dehulling, broken groats, and milling yield.
                   Hulls remaining            Broken groats

Cultivar        Impact   Compressed air   Impact   Compressed air

                               %                         %

AC Assiniboia     9.4         5.1           9.7         1.6
Bay              11.4         1.5           5.8         1.5
Belle             7.8         0.6          10.2         3.4
Brawn            11.9         7.5           7.5         1.7
Gem               9.1         4.7           7.9         3.0
Hytest            7.2         0.7          11.1         2.3
Jerry            10.6         1.4           8.5         1.8
Jud              13.3         6.6          11.5         2.2
AC Medallion     10.3         1.3          11.9         3.7
Youngs           12.9         5.8           7.3         1.9
HSD (0.05)        3.5         3.5           2.5         2.5

                Milling yield([dagger])

Cultivar        Impact   Compressed air

                        kg

AC Assiniboia    149          149
Bay              152          160
Belle            147          145
Brawn            154          160
Gem              152          159
Hytest           149          143
Jerry            154          151
Jud              164          157
AC Medallion     155          147
Youngs           151          156
HSD (0.05)         3            3


([dagger]) Mill yield is defined as the mass of whole oats required to yield 100 kg of whole groats.

ANOVA (not shown) indicated significant location and genotype main effects, as well as a significant genotype x location interaction on groats broken during dehulling (P [is less than] 0.01). The genotype x location interaction appeared to be due to differences in magnitude and ranking of genotypes among the different environments. ANOVA (not shown) also indicated that method of dehulling had a significant effect (P [is less than] 0.01) on groat breakage (Table 5). There was much less groat breakage by the compressed air dehuller than with the impact dehuller. This indicated that the impact dehuller exerted more mechanical stress on the oat grain during dehulling than the air pressure dehuller, as concluded earlier when comparing corrected groat percentages. Bay and Brawn typically had low breakage, whereas AC Medallion, Belle, and Hytest had high levels of breakage.

Milling Yield

By definition, values of milling yield are inverse to groat percentage. A high groat percentage will lead to a low mill yield. ANOVA (not shown) indicated a significant genotype effect, a significant genotype x location interaction (P [is less than] 0.01), but not a significant location effect (P [is greater than] 0.05) on mill yield. The genotype x location interaction appeared to be due primarily to a difference in ranking of cultivars among the locations. Although the dehulling method main effect was not significant (P [is greater than] 0.05), the method x genotype interaction was significant (P [is less than] 0.01). Mill yield of AC Medallion was much higher by the impact dehuller than by the compressed air dehuller, because of increased groat breakage with the impact dehuller. Hytest, Belle, and AC Assiniboia were low ranking by both methods, which reflects their high groat percentage and low breakage. Brawn was high ranking in mill yield by both methods. Jerry, Gem, and Youngs were in the midrange for both methods. Jud was high in milling yield by impact dehulling, but in the midrange by the compressed air dehuller. Bay was high in milling yield by the compressed air dehuller, but in the midrange by the impact dehuller (Table 5). The higher corrected groat percentage of Bay by the impact dehuller, attributed to the lower aspiration of the compressed air dehuller, is likely to be responsible for the difference.

Image Analysis

ANOVA (not shown) for whole oat size characteristics as determined by image analysis indicated significant location and genotype main effects (P [is less than] 0.01) and significant genotype x location interaction (P [is less than] 0.01) for all characteristics tested, including kernel mass, kernel length, kernel width, image area, mass: area ratio, and area CV. Genotype x location interactions appeared to be due to differences in ranking as well as magnitude of genotypes among locations. Genotypic means of whole oat and groat size characteristics are shown in Tables 6 and 7.
Table 6. Whole oat kernel physical characteristics determined by
digital image analysis.

               Whole oat physical characteristics

Cultivar             Mass         Length   Width

                      mg
                [kernel.sup.-1]         mm

AC Assiniboia        36.7          10.8    2.79
Bay                  28.6           9.8    2.88
Belle                29.6          10.2    2.62
Brawn                34.9          11.4    2.79
Gem                  32.4          10.6    2.84
Hytest               32.4           9.4    2.81
Jerry                30.2           9.4    2.74
Jud                  28.2          10.7    2.44
AC Medallion         33.7          10.1    2.78
Youngs               38.0          11.3    2.82
HSD (0.05)            2.5           0.7    0.12

                  Image       Mass:area      Uniformity
Cultivar           area         ratio       of oat area

                [mm.sup.2]   g [m.sup.-2]   CV([dagger])

AC Assiniboia      24.0          1.53           24.5
Bay                22.1          1.29           25.3
Belle              21.3          1.39           23.9
Brawn              25.0          1.40           28.1
Gem                23.9          1.35           26.3
Hytest             21.4          1.52           23.8
Jerry              20.6          1.46           23.8
Jud                20.9          1.35           25.1
AC Medallion       22.4          1.50           24.9
Youngs             25.7          1.47           27.9
HSD (0.05)          1.8          0.15            4.0


([dagger]) Coefficient of variation of the image area of whole oat kernels.
Table 7. Physical characteristics of oat groats derived from digital
image analysis.

                     Oat groat physical characteristics

                                                    Image
Cultivar             Mass        Length   Width      area

                      mg
                [groat.sup.-1]   --mm--           [mm.sup.2]

AC Assiniboia        28.7          6.97    2.24      13.4
Bay                  20.8          5.49    2.36      11.0
Belle                24.2          6.57    2.16      12.3
Brawn                27.6          6.75    2.27      13.4
Gem                  23.9          5.95    2.35      11.8
Hytest               26.2          6.53    2.31      12.8
Jerry                22.7          6.04    2.23      11.6
Jud                  22.0          6.68    1.99      11.6
AC Medallion         26.4          6.59    2.25      12.7
Youngs               29.1          7.10    2.26      13.9
HSD (0.05)            2.3          0.35    0.13       1.0

                Oat groat physical characteristics

                  Mass:area      Uniformity
Cultivar            ratio       of groat area

                g [mm.sup.-2]   CV([dagger])

AC Assiniboia       6.97            20.3
Bay                 5.49            21.9
Belle               6.57            19.4
Brawn               6.75            20.1
Gem                 5.95            23.8
Hytest              6.53            19.9
Jerry               6.04            22.0
Jud                 6.68            24.9
AC Medallion        6.59            22.4
Youngs              7.10            22.7
HSD (0.05)          0.10             3.7


([dagger]) Coefficient of variation of the cross-sectional area of oat groats.

Correlation Analysis

Values of groat percentage as estimated by different methods were significantly correlated with each other (Table 8). Groat percentage determined by hand dehulling was more highly correlated with the corrected groat percentages than with the crude groat percentages as determined by either of the mechanical dehulling methods. This indicated that hand sorting crude groat preparations to remove hulls improved the estimation of groat percentage.

Table 8. Pooled correlation coefficients for groat percentage as determined by hand dehulling, impact dehulling, or compressed air dehulling. Phenotypic correlation coefficients (r) were pooled from three locations (n = 10).
                            Correlation coefficients
                     Hand
                     groat        Impact crude
                     percentage   groat percentage

Impact crude
  groat percentage   0.652(**)
Impact corrected
  groat percentage   0.887(**)    0.671(**)
Impact mill yield    0.430(*)     0.319
Compressed air
  crude groat
  percentage         0.714(**)    0.837(**)
Compressed air
  corrected groat
  percentage         0.867(**)    0.745(**)
Compressed air
  mill yield         0.837(**)    0.740(**)

                     Impact
                     corrected          Impact milling
                     groat percentage   yield

Impact crude
  groat percentage
Impact corrected
  groat percentage
Impact mill yield    0.654(**)
Compressed air
  crude groat
  percentage         0.739(**)          0.186
Compressed air
  corrected groat
  percentage         0.903(**)          0.440(*)
Compressed air
  mill yield         0.893(**)          0.484(*)

                     Compressed air     Compressed air
                     crude              corrected
                     groat percentage   groat percentage

Impact crude
  groat percentage
Impact corrected
  groat percentage
Impact mill yield
Compressed air
  crude groat
  percentage
Compressed air
  corrected groat
  percentage         0.793(**)
Compressed air
  mill yield         0.779(**)          0.983(**)


(*), (**) Significant at the 0.5 and 0.01 level respectively.

Oat and groat physical characteristics that were significantly correlated with groat percentage, as determined by the three methods, included test weight, whole oat image area CV (uniformity), oat weight:area ratio, groat weight: oat weight ratio, groat area: oat area ratio, oat length: groat length difference, and the oat weight: groat weight difference (Table 9).

Table 9. Phenotypic correlations of groat percentage determined by different methods with oat and groat physical characteristics (n = 10).
                                      Correlation coefficients

                                     Hand groat   Impact groat
Factor                               percentage    percentage

Test weight                           0.542(**)     0.603(**)
Whole oat weight                      0.007         0.152
Whole oat length                     -0.354        -0.354
Whole oat width                      -0.089        -0.046
Whole oat image area                 -0.324        -0.208
Whole oat area CV                    -0.604(**)    -0.543(**)
Oat weight:area ratio                 0.444(*)      0.531(**)
Groat weight                          0.222         0.348
Groat length                          0.257         0.295
Groat width                          -0.065         0.082
Groat image area                     -0.525(**)    -0.527(**)
Groat weight:groat area ratio         0.280         0.397
Groat weight:oat weight ratio         0.685(**)     0.584(**)
Groat area:oat area ratio             0.643(**)     0.656(**)
Oat length:groat length difference   -0.646(**)    -0.651(**)
Oat weight:groat weight difference   -0.548(**)    -0.447(*)

                                      Compressed air
Factor                               groat percentage

Test weight                               0.741(**)
Whole oat weight                          0.080
Whole oat length                         -0.483(*)
Whole oat width                          -0.059
Whole oat image area                     -0.381
Whole oat area CV                        -0.704(**)
Oat weight:area ratio                     0.649(**)
Groat weight                              0.266
Groat length                              0.288
Groat width                              -0.055
Groat image area                         -0.390
Groat weight:groat area ratio             0.330
Groat weight:oat weight ratio             0.614(**)
Groat area:oat area ratio                 0.783(**)
Oat length:groat length difference       -0.837
Oat weight:groat weight difference       -0.459(*)


(*), (**) Significant at the 0.05 and 0.01 level respectively.

Test weight is a measure of whole oat density although it is also affected by the packing efficiency of kernels. Because the groat is denser than the oat hull, a relationship between test weight and groat percentage might be expected. Relationships between test weight and groat percentage have been observed previously (Atkins, 1943; Bartley and Weiss, 1951; Pomeranz et al., 1979; Souza and Sorrells, 1988; Stoa et al., 1936). However, as emphasized by Pomeranz et al. (1979), test weight alone was not a reliable predictor of groat percentage. The oat weight: area ratio (Table 6) is also an estimate of oat density, derived from image analysis. Because this factor is derived from single kernels, it would not be affected by the packing efficiency of kernels, as is test weight. The oat weight: area ratio was highly correlated with test weight (r = 0.883; P [is less than] 0.01), but was not as highly correlated with groat percentage as test weight (Table 9).

Whole oat size uniformity, as estimated by the whole oat image area CV was also highly correlated with groat percentage (Table 9). The relationship between oat size uniformity and groat percentage is well known to many oat millers, who will sort oats according to size and dehull the more uniformly sized oat preparations separately to maximize milling yield (Ganssmann and Vorwerck 1995). Different size oats dehull optimally at different rotor speeds in an impact dehuller. Although the size distributions of oat kernels is typically multimodal, because of size differences among primary, secondary and tertiary kernels (Symons and Fulcher, 1988), very little work has been done to characterize differences is size distributions among cultivars and environments. In light of the importance of this characteristic to groat percentage and oat milling yield, this subject may deserve further attention.

The groat weight: oat weight ratio, calculated from oats and groats dehulled with the compressed air dehuller was significantly correlated with groat

percentage (Table 9). The observation that it was not more highly correlated with experimentally obtained groat percentage values (which are also groat weight: oat weight ratios) indicates that significant amounts of groats were lost in the air pressure dehulling process, either by breakage or by aspiration. The groat area: oat area ratio is also an estimate of groat percentage, based on image areas, and was highly correlated with groat percentage (Table 9).

Additional correlation analyses (Table 10) indicated a strong negative correlation between the hulls remaining after dehulling with groat percentage (by either of the mechanical dehulling methods used). If oats with heavier hulls have lower groat percentage, then these heavier hulls may be more difficult to remove by aspiration. Thus, oats with lower groat percentages might be expected to have more hulls remaining after dehulling. This observation emphasizes the importance of hand-sorting crude groat preparations from mechanical oat dehullers, because hulls remaining with the groats will increase the crude groat percentage. If the crude groat percentage were the only selection factor for groat percentage, one could inadvertently select for a low groat percentage cultivar because its groat preparation contained more hulls, when one intended to select for a high groat percentage cultivar. Oat length was also correlated with hulls remaining, suggesting that longer hulls are heavier hulls (Table 10).

Table 10. Phenotypic correlations of hulls remaining and groats broken in groat preparations derived from the impact and compressed air dehullers with other dehulling characteristics and physical characteristics of the oat kernel.
                                     Hulls remaining
Factor                            after impact dehulling

Hand groat percentage                    -0.701(**)
Impact groat percentage                  -0.812(**)
Compressed air groat percentage          -0.653(**)
Oat length                                0.544(**)
Oat width                                -0.203
Oat image area                            0.342
Oat area CV                               0.544(**)
Oat weight                                0.128
Oat weight:oat area ratio                -0.264
Test weight                              -0.460(*)
Impact hulls remaining                      -
Impact broken groats                        -
Compressed air hulls remaining              -

                                      Broken groats
Factor                            after impact dehulling

Hand groat percentage                    0.644(**)
Impact groat percentage                  0.541(**)
Compressed air groat percentage          0.673(**)
Oat length                              -0.249
Oat width                               -0.525(*)
Oat image area                          -0.454(*)
Oat area CV                             -0.494(*)
Oat weight                              -0.142
Oat weight:oat area ratio                0.386
Test weight                              0.567(**)
Impact hulls remaining                  -0.384
Impact broken groats                        -
Compressed air hulls remaining              -

                                     Hulls remaining after
Factor                              compressed air dehulling

Hand groat percentage                       -0.540(**)
Impact groat percentage                     -0.577(**)
Compressed air groat percentage             -0.695(**)
Oat length                                   0.820(**)
Oat width                                   -0.208
Oat image area                               0.552(**)
Oat area CV                                  0.600(**)
Oat weight                                   0.327
Oat weight:oat area ratio                   -0.215
Test weight                                 -0.337
Impact hulls remaining                       0.622(**)
Impact broken groats                        -0.212
Compressed air hulls remaining                 -

                                      Broken groats after
Factor                              compressed air dehulling

Hand groat percentage                       -0.532(**)
Impact groat percentage                      0.397
Compressed air groat percentage              0.464(*)
Oat length                                  -0.058
Oat width                                   -0.180
Oat image area                              -0.125
Oat area CV                                 -0.186
Oat weight                                  -0.006
Oat weight:oat area ratio                    0.143
Test weight                                  0.246
Impact hulls remaining                      -0.322
Impact broken groats                         0.588(**)
Compressed air hulls remaining              -0.237


(*), (**) Significant at the 0.05 and 0.01 level respectively.

The positive correlation between broken groats and groat percentage (Table 10) was also very interesting. It appears that thicker hulls, associated with lower groat percentages, may provide groats with more protection from the mechanical stress involved in the dehulling process, resulting in less groat breakage. Thin-hulled oat may provide the highest groat percentages, but also provide for increased groat breakage during dehulling.

CONCLUSIONS

Mechanical dehulling affected values for groat percentage, as did cultivar and environmental factors. The results suggest that excess mechanical stress from either higher air pressure or increased dehulling duration resulted in increased dehulling efficiency, but also resulted in increased groat breakage. Increased aspiration removed greater amounts of hull, but also removed small groats and broken groats, resulting in decreased groat percentages.

Groat percentage determined by hand dehulling is considered to be a ceiling value. Although crude groat percentages were often higher than the hand dehulled value, this was because samples were often heavily contaminated with hulls. Corrected groat percentages were highly correlated with hand dehulled groat percentage values, indicating the value of hand sorting mechanically dehulled samples to obtain reliable groat percentage values. Cultivars with lower groat percentage were more likely to provide erroneously high crude groat percentage values because more hulls remained with the groats after dehulling by mechanical means. Interestingly, these cultivars were also less likely to have broken groats after dehulling. Thicker hulls may provide more protection to groats during dehulling, resulting in less breakage.

Test weight and size uniformity were whole oat characteristics that were most highly correlated with groat percentage. More detailed analyses of oat size distributions may provide improved strategies for identifying oat size uniformity.

(1) Mention of firm names or trade products does not imply that they are endorsed or recommended by the U.S. Department of Agriculture over other firms or similar products not mentioned.

REFERENCES

Atkins, R.E. 1943. Factors affecting milling quality in oats. J. Am. Soc. Agron. 35:532-539.

Bartley, B.G., and M.G. Weiss. 1951. Evaluation of physical factors affecting quality of oat varieties from Bond parentage. Agron. J. 43:22-25.

Bunch, R.A., and R.A. Forsberg. 1989. Relationship between groat percentage and productivity in an oat head-row series. Crop Sci. 29:1409-1411.

Deane, D., and E. Commers. 1986. Oat cleaning and Processing. p. 371-412. In F.H. Webster (ed.) Oats: chemistry and technology. American Association of Cereal Chemists, St. Paul, MN.

Ganssmann, W., and K. Vorwerck. 1995. Oat milling, processing and storage, p. 369-408. In R.W. Welch (ed.) The oat crop: Production and utilization. Chapman & Hall, London.

Kittlitz, E.V., and H. Vetterer. 1972. Untersuchungen zur maschinellen Entspelzung von Haler mit einem Druckluftspelzer. (In German). Dtsch. Mullerzeitung 70:245-246.

Love, H.H., T.R. Stanton, and W.T. Craig. 1925. Improved oat varieties for New York and adjacent states. USDA Department Circular 353.

Palagyi, A. 1983. Tertiary seed proportions in the grain yield of several oat varieties. Cereal Res. Comm. 11:269-274.

Peek, J.M., and L.M. Poehlman. 1949. Grain size and hull percentage as factors in the milling quality of oats. Agron. J. 41:462-466.

Pomeranz, Y., G.D. Davis, J.L. Stoops, and F.S. Lai. 1979. Test weight and groat-to-hill ratio in oats. Cereal Foods World 24:600-602.

Ronald, P.S., P.D. Brown, G.A. Penner, A. Brule-Babel, and S. Kibite. 1999. Heritability of hull percentage in oat. Crop Science 39:52-57.

Souza, E.J., and M.E. Sorrells. 1988. Mechanical mass selection methods for improvement of oat groat percentage. Crop Sci. 28:618-623.

Steel, R.G.D., J.H. Torrie, and D.A. Dickey. 1997. Principles and procedures of statistics. 3rd Edition, McGraw-Hill Book Company, New York.

Stoa, T.E., R.W. Smith, and C.M. Swallers. 1936. Oats in North Dakota. ND Agr. Exp. Sta. Bull. 287.

Stuthman, D.D., and R.M. Granger. 1977. Selection for caryopsis percentage in oats. Crop Sci. 17:411-414.

Symons, S.J., and R.G. Fulcher. 1988. Determination of variation in oat kernel morphology by digital image analysis. J. Cereal Sci. 7:219-228.

Wesenberg, D.M., and H.L. Shands. 1973. Heritability of oat caryopsis percentage and other grain quality components. Crop Sci. 13: 481-484.

Wolf, W., W.E.L. Spiess, G. Jung, H. Weisser, H. Bizot, and R.B. Duckworth. 1984. Standardization of isotherm measurements. J. Food Engin. 3:51-73.

Youngs, V.L., and H.L. Shands. 1974. Variation in oat kernel characteristics within the panicle. Crop Sci. 14:578-580.

Douglas C. Doehlert,(*) Michael S. McMullen, Robert R. Baumann

D.C. Doehlert, USDA/ARS Wheat Quality Lab., Harris Hall, North Dakota State Univ., Fargo, ND 58105; M.S. McMullen and R.R. Baumann, Dep. of Plant Sciences, North Dakota State Univ., Fargo, ND 58105. Received 6 Nov. 1998. (*)Corresponding author (doehlert@ plains.nodak.edu).
COPYRIGHT 1999 Crop Science Society of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1999 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Doehlert, Douglas C.; McMullen, Michael S.; Baumann, Robert R.
Publication:Crop Science
Geographic Code:1USA
Date:Nov 1, 1999
Words:7391
Previous Article:Nitrogen Fertilization and Supplementation Effects on Performance of Beef Heifers Grazing Limpograss.
Next Article:Selection and Evaluation of Smooth Bromegrass Clones with Divergent Lignin or Etherified Ferulic Acid Concentration.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters