ITEMS FROM YUGOSLAVIA

INSTITUTE OF AGRICULTURAL RESEARCH 'SERBIA'

Center for Small Grains, Save Kovacevica 31, Kragujevac 34000, Yugoslavia.

Harvest index and yield components in 12 Yugoslav winter wheat cultivars. [p. 312-315]

Desimir Knezevic, Veselinka Zecevic, Nevena Djukic, Dusan Urosevic, Milanko Pavlovic, Danica Micanovic, Vesna Urosevic, Miroslav Kuburovic, and Srdjan Jordacijevic.

Abstract. Variability of harvest index and some yield components (plant height, number of spikelets/spike, number of kernels/spike, and 1,000-kernel weight) were studied in 12 cultivars from different Yugoslav wheat breeding centers. Wheat cultivars were grown in the experimental field in five replications. Samples of 100 wheat plants (20 plants in five replications) were analyzed at maturity. Harvest index was computed as the ratio of grain yield to the total above-ground part of the plant. We determined the average difference for all the studied parameters among the investigated cultivars. A high variability in the investigated traits was established. The highest variability was for number of kernels/spike and the lowest were for harvest index, number of spikelets/spike, and stem height. The strongest correlation was found between grain harvest index and 1,000-kernel weight.

Introduction. Increasing the yield in wheat has been an important task of breeders. In the last five decades, breeders have been investigated trends for any sign of a decrease in progress and to ascertain to what extent increases in yield have been due to improved genetic yield potential as opposed to those caused by improved agronomy. Water, fertilizer, pesticide, and herbicide levels application influence wheat yields. In a successful breeding program, wheat breeders has produce a series of cultivars with improved resistance to pests and diseases, wider adaptation to adverse soil and climatic conditions, and better quality. Genetic yield potential may be defined as the yield of a cultivar grown in environments to which it is adapted, where nutrients and water are nonlimiting, and with pests, diseases, weeds, lodging, and other stresses controlled effectively. The genetic yield potential of a wheat cultivar may depend on favorable growing conditions and good agronomy for its expression. The ideal high yielding cultivar or for any other desirable trait, needs to express its genetic potential in different environments with a low variance in the different environmental factors favorable for growth. The grain yield of wheat is a variable trait, which depends on numerous yield components and environmental factors (Kraljevic-Balalic et al. 1995). The harvest index represents the ratio between economical yield (wheat grain yield) and biological yield (yield of grain and straw) (Kraljevic-Balalic and Borojevic 1985). The proportion of total mass to grain yield in most modern cultivars is more than 2:1, which indicates excessive consumption of nutrients for straw formation instead of grain and is one reason for decreasing the vegetative part of a wheat plant to optimal proportions in order to increase the harvest index to greater than 50 % (Zecevic and Knezevic 1997). Plant height, length of spike, and number of spikelets/spike are positively correlated with grain yield (Zecevic 1996). Yield variability is less studied than yield itself. We have studied the variability of harvest index and yield components (height of plants, number of spikelets/spike, number of kernels/spike, and 1,000-kernel weight) in genetically divergent wheat cultivars.

Materials and methods. The variability of harvest index and some yield components (plant height, number of spikelets/spike, number of kernels/spike, 1,000-kernel weight) were studied in 12 cultivars from different Yugoslav wheat breeding centers. For this investigation, we used the cultivars KG-56S, KG-100, Levcanka, Studenica, Tara (Kragujevac), Zemunka, Zadruga (Zemun), Dejana, Rodna, Partizanka, Pobeda (Novi Sad), and Kraljevica (Zajecar). Experiments were performed in a randomized-block design in five replication on the experimental field of the Center for Small Grains, Kragujevac, Yugoslavia. Seed was sown in basic 5-m2 plots. For analysis of plant height, spike length, and number of spikelets/spike, we used 100 plants at maturity (20 plants/replication) and computed the average (x), variance (s2), and coefficient of variation (CV) as an index of relative variability of these traits. Significant differences between the average value were estimated by F-test values. An analysis of variance was made according to a random-block system with one factor, which allowed the calculation of the components of variance (s2g-genetic). The quality of wheat cultivars was estimated on the basis of harvest index, plant height, number of spikelets/spike, number of kernels/spike, and 1,000-kernel weight.

Results. Harvest index (HI). This trait was chosen because it represents the plant's efficiency at nutrient translocation from vegetative to reproductive tissue. Harvest index is the ratio between grain weight/plant and plant weight. Depending on two distinctly quantitative traits, this index is highly variable. The HI value ranged from 45.7 % for KG-100 to 48.5 % for Rodna (Table 1). The variability expressed by coefficient of variability was 2.43 %. Partitioning the total sum of squares by ANOVA showed that all the sources of variation statistically had highly significant F values, except for the replication. Environmental differences contributed more to the total sum of squares for the trials than to genotypic diversity. The sensitivity of HI under environmental variation was noted by Dimitrijevic et al. (2000) as well. Environmental conditions particularly were noticable for the examined genotypes (both season and the genotypes had the same PCA sign). Possibly, weather conditions caused stem shortening, which was favorable for translocation. Varieties were different in their reduction in plant height, which consequently had an impact on HI as well.

Table 1. Average values of analyzed harvest index and yield components in 12 Yugoslav wheats.

Cultivar	Plant height	No. of spikelets/spike	No. of kernels/spike	1,000-kernel weight	Harvest index
KG-56 S	70.82	17.44	33.47	41.08	0.465
KG-100	58.82	18.74	44.09	38.30	0.457
Zemunka	70.18	17.55	40.07	41.70	0.475
Levcanka	79.15	17.96	36.57	39.10	0.470
Kraljevica	76.35	17.36	36.11	38.10	0.478
Zadruga	84.19	16.55	37.18	40.70	0.477
Dejana	60.13	17.25	38.24	38.64	0.465
Studenica	76.31	17.96	39.06	42.08	0.470
Rodna	64.18	17.55	38.72	45.14	0.485
Partizanka	74.98	19.17	35.80	44.66	0.482
Tara	69.97	16.76	32.63	44.00	0.479
Pobeda	74.40	17.78	38.05	45.58	0.486
CV %	1.95	2.15	7.08	1.31	2.43

The increase in the genetic potential of wheat has not come from an increase in the rate of photosynthesis or growth, but rather from a progressive rise in the proportion of biomass that is partitioned into the harvest organs. Increasing the genetic yield potential of wheat can be realized by increasing of harvest index to 50 % or more Borojevic (1986). Increasing the harvest index with increase the yield potential in wheat (Austin et al. 1980, 1982; Petrovic et al. 1984). For wheat and its wild progenitors grown in the field, the harvest index ranged from 5­17 % in the wild diploids to 51 % in modern cultivars.

Plant height. We chose this trait because of its direct effect on 'genotype x environment' interaction and influence on harvest index and yield. Semidwarf wheat genotypes enhanced the interaction between these genotypes and environmental conditions (Dimitrijevic et al. 2000). All the varieties in the study were semidwarf with heights of 58.82 (KG-100) to 84.19 cm (Zadruga) according to mean values over the years of the study. Semidwarf cultivars have advantages over dwarf and tall cultivars because they are aa source of desirable genes in the future breeding programs (Petrovic and Worland 1992). According to the ANOVA, partitioning the total sum of squares for the trial revealed that all the main effects (environmental and genotypic main effects) statistically were highly significant and agronomically important. Similar results were reported by Dimitrijevic et al. (1996) and Zecevic (1996). On average for all cultivars and years, the coefficient of variability was very low (V = 1.95 %), which indicated that genotype had a higher influence on plant height than environment (Table 1). Decreasing stem height gave a higher possibility of efficient utilization mineral elements for the spike development. Translocation of minerals also is more efficient in semidwarf cultivars.

Number of spikelets/spike. Partizanka had the highest average value of number of spikelets/spike (x = 19.17). The lowest average number of spikelets/spike was in Zadruga (x = 16.55). For the expression of high yield in wheat, more than 18 spikelets/spike and an average of three kernel/spikelets are desirable (Borojevic 1972). In our study, few cultivars had more than 18 spikelets because of the high density of planting. The number of productive spikelets/spike has an influence on grain yield. Variability in the number of spikelets/spike, on average, was expressed, and the coefficient of variability was low (CV = 2.15). Similar results were obtained for the variability in the number of spikelets/spike by other investigators (Petrovic et al. 1993, Ivezic 1994, Dimitrijevic et al. 1996, Knezevic et al. 2000).

Number of grain/spike. The number of grain and their spike mass are yield components that are independent from other componenets in the early stages of onthogenesis (Kraljevic-Balalic et al. 1982). Kernel number results from the number of spikelets and number of florets/ spike. The success of pollination and fertilization also have a big influence on expression of this trait. The average value for the number grain/spike ranged from 32.63 (Tara) to 44.09 (KG-100) (Table 1). ANOVA indicated that all sources of variation were statistically highly significant. On average, the analyzed cultivars expressed variability in the number of spikelets/spike and coefficient of variability was low (CV = 7.08 %).

1,000-kernel weight. During the experiment, the highest average value for 1,000-kernel weight was in Pobeda (x = 45.58) and the lowest value was in Kraljevica (x = 38.10). The number of grain/spike and 1,000-kernel weight have a strong influence on spike productivity. On average, variability in 1,000-kernel weight was expressed and the coefficient of variability was low (CV = 1.31). Similar results of in the variability of 1,000-kernel weight have been reported (Bozinovic 1996, Ivanovski et al.1996).

These correlations represent important factor for the plant breeding. Genetic factors and environment influence the correltion between traits. We established different levels of correlation. A highly significant correlation between 1,000-kernel weight and harvest index was established (r = 0.720**) (Table 2). The correlation between harvest index and each of the analyzed yield components was very low or negative. Correlation between stem height and harvest index was negative, what is in agreement with results of Zecevic (1996).

Conclusion. A rise in the harvest index may involve reduction in stems, leaves, and reserves. The genetic control of stem height and other yield components are very important, but environment has influence on expression of traits. Generally speaking, all genotypes reacted similarly to environmental variation for plant height, number of spiklets/spike, number of kernels/spike, and 1,000-kernel weight. However, the reaction of the genotype for harvest index differed in main effect and the interaction made the phenotypic expression for this trait more unpredictable. The remobilization of reserves into spikes and seed varied among the cultivars.

References.

• Austin RB, Bingham J, Blackwell RD, Evans LT, Ford MA, Morgan CL, and Taylor M. 1980. Genetic improvements in winter wheat yields since 1900 associated physiological changes. J Agric Sci Chamb 94:675.
• Austin RB, Morgan CL, Ford MA, and Bhagwat SG. 1982. Flag leaf photosinthesis of Triticum aestivum and related diploid and tetraploid species. Ann Bot 49:177.
• Borojevic S. 1986. Harvest index in function of yield increasing. Contemp Agric 38:25-44 (in Serbian).
• Bozinovic I. 1996. Effects of hormonal herbicides on quantitative traits in different wheat genotypes (Triticum aestivum L). MS Thesis, Agric Faculty of Novi Sad.
• Dimitrijevic M, Knezevic D, Petrovic S, and Zecevic V. 2000. Stability of yield components in wheat (Triticum aestivum L.). In: Proc XIth EUCARPIA, Paris, France. pp. 106-107.
• Dimitrijevic M, Ivezic J, Kraljevic-Balalic M, and Petrovic S. 1996. Phenotypic variability of yield components and phenotypic distance of wheat cultivars (Triticum aestivum L.). Contemp Agric 44:47-60 (in Serbian).
• Ivanoski M, Knezevic D, and Menkovska M. 1996. Inheritance of grain quality traitsin hybrids F4 generation in wheat. Contemp Agric 44:19-23 (in Serbian).
• Ivezic J. 1994. Genetic divergence of wheat yield components. MS Thesis, Agric Faculty of Novi Sad.
• Knezevic D, Zecevic V, Dimitrijevic M, and Petrovic S. 2000. Variability of yield components in wheat (Triticum aestivum L.). In: Proc XIth EUCARPIA, Paris, France. pp. 104-105.
• Kraljevic-Balalic M, Mihaljev I, and Leghari MA. 1982. Ecological and genetic variability of number and weight of grain per spike. J Agric Sci 43:495-504.
• Kraljevic-Balalic M and Borojevic S. 1985. Inheritance of stem height and harvest index in wheat. J Agric Sci 46:253-266.
• Kraljevic-Balalic M, Dimitrijevic M, and Petrovic S. 1995. Interaction genotype/environment for yield components. First Symp of Breeding Organisms. V. Banja. p. 58.
• Petrovic S, Dimitrijevic M, and Kraljevic-Balalic M. 1994. Influence of stem height on harvest index in different wheat genotypes (Triticum aestivum L.). Contemp Agric 42:95-98 (in Serbian).
• Petrovic S, Kraljevic-Balalic M, and Dimitrijevic M. 1993. Variability and heritability of wheat yield components. Contemp Agric 40:73-76 (in Serbian).
• Petrovic S and Worland AJ. 1992. Genes reductors of stem height. 1. Determination of its presence in Yugoslav wheat cultivars. Contemp Agric 40:81-85 (in Serbian).
• Zecevic V. 1996. Genetic identification of different winter wheat cultivars (Triticum aestivum L.). Ph.D. Thesis. Univ of Novi Sad.
• Zecevic V and Knezevic D. 1997. Variability and components of variance for harvest index in wheat (Triticum aestivum L.). Genetika 29:195-202.