GrainGenes

Heritability Estimates for Forage Quality in Barley

Mario C. Therrien

AAFC Brandon Research Centre

Box 1000A, RR#3, Brandon, MB. Canada, R7A 5Y7

MTherrien@agr.gc.ca

Introduction

One of the principle agricultural industries in Canada is beef production, with an estimated net worth of $10 billion Cdn. In most cases, the primary source of feed is barley with approximately 300,000 ha devoted to forage barley production, on an annual basis, across Canada (Therrien, unpublished). Except for dairy rations, most producers did not assess the quality of forage barley until recently, as barley forage or silage was regarded as a cost-effective way of “filling the rumen” and maintaining the animal. With recent challenges in the beef industry, the producer has taken a holistic approach to forage production to minimize costs while optimizing the quality of beef. While the quality characteristics of forage are reasonably well understood (Guines et al., 2002), there has been limited work on determining the inheritance of forage quality in barley (Draghici and Bude, 1974). In our efforts to improve forage quality in our breeding program, we utilized multi-site, multi-year forage quality data from co-operative trials to estimate broad sense heritability for three key forage quality traits under western Canadian conditions.

Materials and Methods

Four barley varieties, used as checks, were grown in the western Canadian Forage Barley Co-operative Trial (FBCoop), in the years 1997, 1999, 2001, 2002, 2003, in a four-replicate Randomized Complete Block (RCB) design at five or six locations across western Canada. The varieties Virden, Westford, AC Lacombe and AC Rosser were grown across all site-year combinations and represent a fairly broad range of forage quality. Each variety was grown in 5 or 6 square meter plots and the entire plot harvested at the mid-dough stage of kernel development. A 500 gram sample (dry weight basis) was taken from each plot and ground to a uniform 1 mm particle size. A 10 g sub-sample was taken and used to determine Crude Protein (CP; % by weight), Acid Detergent Fibre (ADF; % by weight) and Neutral Detergent Fibre (NDF; % by weight). CP was determined using Near Infrared Scanning (NIRS) at Nowest Labs, Lethbridge, AB., and ADF and NDF were determined at our lab using the Ankom® Fibre Digester and according to manufacturer specifications. Analysis of Variance and heritability estimates were performed using the method of Allard (1960).

Results and Discussion

Table 1. summarizes the results of Analysis of Variance (ANOVA) for the three traits studied, as well as their broad-sense heritability estimates (H). For CP, genotypic variance was far greater than environmental variance, with H=63.9, indicating that CP is fairly highly heritable and amenable to improvement via breeding. This is not unexpected as protein in barley is considered a highly heritable trait. For both ADF and NDF, environment was the main influence and, thus, heritability was moderate to low (H=42.9 and 35.9, resp.). This indicates that breeding for these traits would present a greater challenge (than CP) and that genetic gains would be longer term and incremental. This also points to the need to select over multiple environments for these two important forage quality traits.

Table 1. Analysis of Variance and Heritability (H) estimates for Crude Protein (CP), Acid Detergent Fibre (ADF) and Neutral Detergent Fibre (NDF) in four barley cultivars over multiple (16 to 19) environments:

A. Crude Protein:

B. Acid Detergent Fibre:

C. Neutral Detergent Fibre:

References

Allard, R.W. 1960. Principles of Plant Breeding. John Wiley & Sons, New York. pp. 75-108.

Draghici, L., Bude, A. 1974. The quality of  barley for malting and for fodder from the aspects of genetics and methods of breeding. Probl. Genet. Theor. Appli. 6(5): 371-409.

Guines, F., Julier, B., Ecalle, C., Huyghe, C. 2002. Genetic control of quality traits of Lucerne (Medicago sativa L). Austral. J. Agric. Res. 53(4): 401-407.