(go to KOMUGI Home) (go to WIS List) (go to NO.87 Contents)

Wheat Information Service
Number 87: 39-41 (1998)
Research information

Combining ability analysis in bread wheat adapted to the East African highlands

William W. Wagoire, Olav Stolen, and Rodomiro Ortiz*

Department of Agricultural Sciences, The Royal Veterinary and Agricultural University, 40 Thorvaldsenvej, DK- 1871 Frederiksberg C, Copenhagen, Denmark

The Uganda Wheat Development Project (UWDP) aims to develop widely adapted, high yielding, cultivars with good resistance to diseases. The objective of this study was to determine the combining ability for days to heading, plant height and grain yield of cultivars widely tested in Uganda but whose yellow rust resistance varied from very resistant to very susceptible.

Sixty-four genotypes, derived from a full diallel cross among eight cultivars were assessed for agronomic characteristics. The parental genotypes were selected for their good agronomic characteristics but with varying responses to yellow rust (Wagoire et al. 1998a). The experiments were planted in two cropping seasons (A and B) at two locations (Kalengyere and Buginyanya), both having bimodal rainfall. Kalengyere (South latitude 1-degree 15 minutes, East latitude 29-degree 45 minutes ) is at 2400 m.a.s.l, has an Andosol with pH 5.7, and an average temperature of 16C throughout the year. The high rainfall (750 mm) season (B) lasts from September to March and the relatively low rainfall (480 mm) season (A) from March to August. Buginyanya (North latitude 1-degree 1 minute, East latitude 34-degree 2 minutes) is at 2100 m.a.s.l., has an Andosol with pH 5.5, and an average temperature of 18C. The high rainfall (560 mm) season (B) occurs from September to March and the relatively low rainfall (470 mm) season (A) occurs from March to August (470 mm). The study was carried out in three growing seasons from August 1994 to March 1996. An additional environment was obtained by applying fungicide to control yellow rust at Kalengyere in 1995 (B) season. All plots were fertilized at a rate of 50 kg N ha-1 prior to planting. The combining ability effects were calculated using Griffing's (1956) method. The phenotypic stability of all the characters was investigated using Eberhart and Russell's procedures (1966). In addition, correlations between combining ability values of days to heading, plant height and grain yield and their stability parameters were calculated.

The diallel analysis revealed the predominant role of additive gene action for days to heading, plant height and grain yield (Table 1). Parental genotypes showed significant general combining ability (GCA) for days to heading, plant height, and grain yield (Table 2). Similarly, there was significant specific combining ability for the same characters. A highly significant genotype-by-environment interaction affected all the characters, which suggests that selection for specific environments could maximize the use of available germplasm. The significance of environmental effects in our investigation was attributed to the biotic and biotic differences. For example, the amount of precipitation received in each season as well as the annual total precipitation per site were different. Also, Kalengyere had a lower mean temperature as compared to Buginyanya. In addition, one of the experiments in 1995 (B) season at Kalengyere was sprayed with a fungicide in order to control yellow rust. The yellow rust significantly affected the phenotypic performance for grain yield and plant height (Wagoire et al. 1998b).

Significant genotypic effects were expected since the parental lines had been selected to provide a full range of reaction types to yellow rust (Wagoire et al. 1998a). This also explained to some extent the observed genotype-by-environment effects for all the characters. The experiments at Buginyanya were yellow rust-free during the test period while those at Kalengyere always had yellow rust infestation which varied in severity between seasons. The yellow rust severity (coefficient of infection) at Kalengyere averaged 7.89 % in 1994 (B) season, 30.17 % in 1995 (A) season and 24.05 % in 1995 (B) season.

GCA was correlated with the phenotypes of the parents across locations (r < 0.94, P < 0.001 suggesting that phenotypic selection may be adequate for choosing parents. However, GCA was not correlated with phenotypic stability, thereby multilocational testing would be required to select stable genotypes. Specific combining ability (SCA) was only correlated with yield stability across environments (r = 0.53, P < 0.001 ). We suggest that a stable wheat cultivar for Uganda should be defined as one with a high mean yield (above the national average of 1. 8 t ha-1), regression coefficient for performance across environments (b) equal to 1, and almost nil deviations from the .regression slope. These criteria were used because Uganda relies on wheat germplasm introductions that were bred for high yielding environments (Braun et al. 1996).

References

Braun H-J, Rajaram S and Van Ginkel M (1996) CIMMYT's approach to breeding for wide adaptation. Euphytica 92: 175-183.

Eberhart SA and Russell WA (1966) Stability parameters for comparing varieties. Crop Sci 6: 36-40.

Griffing B (1956) Concept of general and specific combining ability in relation to diallel crossing systems. Austral J Biol Sci 9: 463-493.

Wagoire WW, Stolen 0, Hill J and Ortiz R (1998a) Is there a cost for wheat cultivars with genes for resistance to yellow rust caused by Puccinia striiformis? Crop Protection 17: 337-340.

Wagoire WW, Stolen 0, Hill J and Ortiz R (1998b) Inheritance of adult field resistance to yellow rust disease among broad-based hexaploid spring wheat germplasm. Theor Appl Genet, in press.

(go to KOMUGI Home) (go to WIS List) (go to NO.87 Contents)