Heat stress is one of the major constraints of wheat (Triticum aestivum L.) production in many areas around the world. Heat stress at late growth stages is a problem in 40% of wheat areas in the temperate environments (Reynolds et al. 2001). Moreover, wheat encounters heat stress during the early growth stages in the tropical and subtropical regions (Elahmadi 1996). In the Sudan, a number of environmental stresses affect wheat productivity, but heat stress remains the main constraint and affects the crop at all stages of its development (Ageeb 1994; Elahmadi 1996). Late heat stress is quite common because of the delay in sowing often practiced by farmers due to many reasons (Ibrahim 1996). However, early heat stress is also important because of the great yearly weather fluctuations (Elahmadi 1996).
Wheat grain yield is the product of the two major yield components; the number of grains/m2 and the individual grain weight. When the crop is exposed to high temperatures before anthesis, reduction in grains number occurs via reduction in spike/m2 and grains/spike (Shpiler and Blum 1991). The reduction in grain weight results from reductions in the grain filling duration and rate (Shpiler and Blum 1991; He and Rajaram 1994). Some studies were able to identify traits that could be used as selection criteria under heat stress conditions. Nonetheless, the uniqueness of the wheat growing environment in the Sudan necessitates the search for relevant selection criteria that might be associated with yield under such environment and accelerate developing heat-tolerant wheat cultivars. In such environment, multiple selection criteria must be applied to predict the crop response to yearly-fluctuating weather. Thus, this experiment was carried out to fulfill the above objectives.
Materials and methods
Eight bread wheat cultivars and two advanced lines were selected to represent different origin and eras of wheat improvement in the Sudan. The cultivars were Beladi, Falchetto, Giza 155, Mexicani, Condor, Debeira, El Nielain and Sasaraib. The two lines, VYT #3 97/98 and VYT #8 97/98, were selected from the national yield trial of the Sudan in 1997/98. The experiment was conducted for two seasons, 1997/98-1998/99 at the Gezira Research Farm (GRF), Agricultural Research Corporation (ARC), Wad Medani, located on the clay plain of the central Sudan (14°24'N, 33°29'E, 407 m asl). The soils of GRF are cracking heavy clay vertisols, with very low water permeability, pH of 8.5, poor in organic matter (0.5%), deficient in nitrogen (300-400ppm), and low in available phosphorus (4-5 ppm extractable P).
Three sowing dates; early, optimum and late (1st, 3rd week of November and 2nd week of Dece tmber, respectively) were used during the two seasons. Early sowing date subjects wheat to heat stress during the early stages of the crop, which affect the establishment of the crop and the development of tillers and spikelets. This permits differentiation of genotypes according to their reaction to the early heat stress. The late sowing allows the crop to grow and mature under high temperatures and the response of different genotypes to late heat stress can be assessed.
Seeding at the rate of 120 kg/ha was done manually in plots consisted of six rows 5 m long and 20 cm apart. Fertilization was done using triple superphosphate (43 kg/ha of P205) prior to planting, while urea was broadcasted before the second irrigation (86 kg/ha of N). Irrigation was carried out at 7-10 days intervals to avoid any water stress. The experiment was arranged in a randomized complete block design with three replications.
Data were collected on date of anthesis, date of maturity, grain filling duration, plant height and number of spikes/m2. A net area of 3.2 m2 was hand harvested from the ground level. The harvested material was sun-dried, weighed, threshed and the grains were weighed again to give biomass and grain yield. Grain number/m2 was calculated as grain yield (g/m2) divided by individual grain weight. Grains/ spike, thousand kernel weight and harvest index were also calculated. Grain growth rate was calculated as grain yield divided by grain fill duration. Biomass growth rate was calculated as biomass divided by days from seedling emergence to maturity, while vegetative growth rate was calculated as (biomass less grain weight) divided by the days from emergence to anthesis.
The heat stress intensity (HSI) was defined by the formula adopted by Shpiler and Blum (1991): HSI = 1- Xh/Xo: where Xh is the grain yield or other tra tits under early and late sowing conditions and Xo is the grain yield or other traits under optimum sowing condition. Standard statistical analysis of variance was done. Means were compared using Tukey's test. Simple correlations of grain yield with other traits were computed as well as correlations between the HSI of grain yield and other traits.