Previous PageTable Of ContentsNext Page

Grain size distribution is related to small grain screenings in wheat

Darshan L. Sharma1 and Walter K. Anderson2

1 Centre for Cropping Systems, Department of Agriculture, Northam, Western Australia 6401, www.agric.wa.gov.au
Email dsharma@agric.wa.gov.au
2
Department of Agriculture, 444 Albany Highway, Albany, WA, 6330, , www.agric.wa.gov.au
Email wanderson@agric.wa.gov.au

Abstract

Wheat cultivars are known to differ for propensity of small grain screenings which is one of the most important causes of price dockages of wheat in Australia. An efficient way of assessing germplasm nurseries and crossbreds is required for situations when test material can not be tested under conditions that produce small grains, given that cultivars can differ for screenings despite similarity of grain weight, the best known correlated trait. Harvest samples from two field trials, one cultivar x time of sowing and the other cultivar x plant population, were separated into five fractions using four slotted sieves of hole widths 2.5, 2.8, 3.1 and 3.4mm. Comparison of grain size distribution based on percentages of grain passing each sieve revealed that factors that increased small grain screenings (cultivar and delayed seeding) also tended to skew grain size distribution towards lower grain size. Both the low screenings cultivars, Wyalkatchem and Westonia, had peaks at fraction F3 (2.8-3.1mm) while Harrismith, the cultivar with highest screenings, had most grain at F4 (2.5-3.1mm). Carnamah, with medium screenings propensity, had most of its grain in fraction F4 (like Harrismith) but its F3 proportion was significantly greater than Harrismith (like Westonia, Wyalkatchem). Fraction F3 was highly and negatively correlated with small grain screenings (r > -0.85; P < 0.001) in both experiments. The consistency of cultivars for ranking of the F3 fraction suggested that grain size distribution can possibly be used as a selection criterion without specifically subjecting breeding material to stressful conditions.

Media summary

Grain size distribution can be used as an indicator trait for screening wheat germplasm and crossbreds having higher propensity for small grain screenings and possibly flour extraction.

Key Words

Wheat, small grain screenings, grain size distribution, cultivar, price dockages

Introduction

Small grains that pass through a 2mm, slotted screen (sievings or screenings) are one of the most important causes of price dockages of wheat in Australia since grain size variation greatly affects flour yield and commercial value. Screenings are known to be influenced by season, crop management and/or cultivar in wheat (Anderson et al. 1995; Anderson and Sawkins 1997; Sharma and Anderson, 2001, Sadras et al. 2002), barley (Cranstoun and Garstang 1993; Eagles et al. 1995; Conry 1997) and oats (Barr et al. 1994) but central to all these factors are hot, dry conditions during grain filling. Many breeding experiments do not experience conditions sufficiently stressful to distinguish between lines for the propensity to produce small grains. There is thus a need for a method that can discriminate susceptible lines in the absence of stressful conditions in the field. Such situations are unavoidable in breeding programs during early stages of bulking when breeders are keen to rapidly multiply seed under well watered and well fertilised conditions. Grain weight is the known best correlated trait with screenings but a poor linearity in the relation of small grain screenings to grain weight has also been reported (our unpublished results, Lee et al. 1989, Blakely Paynter, personal communication). Finding another closely related trait that can be readily measured from field-grown samples is desirable.

Methods

Experimental site and management variables

Two field experiments were conducted in 2000 at Mullewa (28o 33.30’S, 115o 29.00E) in the low rainfall zone of the Northern Agricultural Region of Western Australia. The soil type was a sandy loam and the previous crop was lupins.

Dates of seeding were 25th May, 26th June and 5th July in the cultivar x time of sowing experiment (CV.TS) while the cultivar x plant population experiment (CV.PP) was seeded on 13th June. Targeted plant populations for the latter experiment were 50, 100, 150, 200 and 250 plants/m2 while targeted plant population for the former was 150 plants/m2.

Data recording and statistical analysis

Two-kg grain samples were collected from each plot. The samples were cleaned in a Carter Day seed cleaner using a 1.5 mm wide sieve. Small grain screenings were determined by shaking half a litre of grain for 40 shakes over 2.0 mm elongated slots (Agtator).

About 250g sub-samples were drawn from each plot and passed through a Sortimat machine fitted with four sieves of the screen widths of 2.5, 2.8, 3.1 and 3.4 mm thus resulting in five fractions which were designated F1 to F5 in decreasing order of seed size. All fractions were weighed, expressed as a percentage of the total, and referred to as grain size distribution.

Data were analysed using analysis of variance procedures in Genstat.

Results

All the main effects and interactions for small grain screenings were significant (P value ?) in both experiments. Cultivars Harrismith and Westonia were the cultivars with highest and the lowest screenings propensities in both the experiments.

Inability of grain weight to differentiate small grain screenings was apparent in these experiments as anticipated. For example, in the CV.TS experiment, the rate of screenings increase with reduction in grain weight (seeding delay) was very very similar for both Harrismith and Arrino but the former invariably produced more screenings than the latter at the same grain weight (Fig. 1).

Figure 1. Grain weight relative to small grain screenings for two wheat cultivars in the cultivar x time of sowing experiments.

Both variety and management strongly influenced grain size distribution. Interaction effects were relatively small compared to main effects. In the CV.PP experiment the Mean Square (MS) due to cultivar differences was more prominent than MS for plant population, suggesting that distribution differences due to cultivars were much greater than those due to plant populations. This is in line with the actual screenings trend in this experiment where differences due to plant population were small compared to cultivar differences. However, differences due to time of sowing were more important than those due to cultivar in the CV.TS experiment (again in line with screenings trend).

A sequential change in the direction from bigger to smaller grain width was found for fraction proportions as seeding was delayed. Figure 2 shows conclusively that the increase in screenings with delayed sowing in every cultivar was associated with an increase in the proportions of fractions F4 and/or F5 at the cost of fractions F2 and F1

.

Figure 2. Grain size distribution of six wheat cultivars at three times of sowing (25th May, 26th June and 05th July). Lsd (P=0.05) was 3.4. F1 and F5 are the fractions with largest and the smallest grain size, respectively.

In order to identify which fraction proportion best identifies high screenings cultivars, correlation coefficients with individual fractions were calculated. Fractions F5 and F3 were respectively the best positively and negatively correlated fraction to screenings; correlation coefficients values being more than 0.93 and -0.85, respectively in the both these experiments. This implies that cultivars with greater proportions of F5 under most situations will have high screenings while those with greater proportion of F3 under most situations will indicate its tolerance. High correlation with fraction F5 (grain width 2.5mm to 1.5mm) is not surprising given that this size is nearly the same as for screenings. A high F3 (2.8mm to 3.1mm) represents extreme skewness against smaller grain.

Conclusion

Grain sizes distribution offers promise in distinguishing wheat cultivars for small grain screenings. The trait could potentially be superior to grain weight for ranking cultivars for screenings propensities irrespective of the production conditions.

References

Anderson WK, Crosbie GB, Lemsom K (1995). Production practices for high protein, hard wheat in Western Australia. Australian Journal of Experimental Agriculture 35, 589-595.

Anderson WK, Sawkins D (1997). Production Practices for Improved Grain Yield and Quality of Soft Wheats in Western Australia. Australian Journal of Experimental Agriculture 37, 173-80.

Barr AR, Jefferies SP, Ward P, Tasker SD, Hoppo TM, Dube A, Lewis J (1994). Registration of Australian winter cereal cultivars. Avena sativa (oats) cv. Potoroo. Australian Journal of Experimental Agriculture 34, 706-707.

Conry MJ (1997). Effect of fertiliser N on the grain yield and quality of spring malting barley grown on five contrasting soils in Ireland. Biology and Environment, 97, 185-196.

Cranstoun DAS, Garstang JR (1993). The effect of husbandry inputs on the quality of spring malting barley with special reference to screenings. Aspects of Applied Biology, 36, 211-220.

Eagles HA, Bedggood AG, Panozzo JF, Martin PJ (1995). Cultivar and environmental effects on malting quality in barley. Australian Journal of Agricultural Research, 46, 831-844.

Lee SH, Scott WR, Love BG (1989). Sources of screenings in malting barley in relation to the pattern of tillering. In ‘Proceedings Annual Conference of the Agronomy Society of New Zealand’ 19, 43-54.

Sharma DL, Anderson WK (2001). Agronomic responses of new wheat cultivars in the northern wheat belt. In ‘Western Australia Crop Updates 2001: Cereal Updates’ (Eds. R Jettner, J Johns) pp 41-42.

Sadras V, Roget D, O'Leary G (2002). On-farm assessment of environmental and management factors influencing wheat grain quality in the Mallee. Australian Journal of Agricultural Research, 53, 811-820.

Previous PageTop Of PageNext Page