HOME PAGE CONTACT DETAILS OBJECTIVES DISCLAIMER MEETING DATES 2019 INTERNET LINKS NEWS
CROPS
RESEARCH PROJECTS
RESEARCH DATABASE
BURSARIES & AWARDS
STATISTICS & ESTIMATES
INCOME & COST BUDGETS
WORKING GROUP MINUTES
NEWSLETTERS
OILSEEDS FOCUS MAGAZINE
Reports
RESEARCH REPORTS
2016/2017
OIL AND PROTEIN SEEDS DEVELOPMENT TRUST
2015/2016
OIL AND PROTEIN SEEDS DEVELOPMENT TRUST
OILSEEDS ADVISORY COMMITTEE
RESEARCH PROJECTS
CONCLUSION
GALLERY
VIDEO GALLERY
ARCHIVES
SEARCH
Cover of Oilseeds Focus Vol 5 No 2 March 2019
REPORTS RESEARCH REPORTS 2015/2016 RESEARCH PROJECTS New Research Projects
PLEASE CLICK ON THE LINKS BELOW TO VIEW INDIVIDUAL FILES
 NEW RESEARCH PROJECTS CONTINUATION PROJECTS JOINT PROJECTS TRANSFORMATION PROJECTS PROVISIONS

RESEARCH REPORT 2015/2016

New Research Projects


  1. Evaluation of commercially available sunflower cultivars
  2. Soybean cultivar selection for improved yield and yield stability
  3. The dynamics of research uptake by South African oilseeds producers


Evaluation of commercially available sunflower cultivars

Dr AA Nel
ARC-GCI

Cultivar trials from previous years showed that the mean yield of the five best cultivars is usually about 0.2t ha-1 higher than the overall mean yield of all the tested cultivars. Considering that the national mean yieldthat farmers obtain is normally between 1.0 and 1.4 t ha-1, it is clear that cultivar selection has a significantaffect on the profitability of sunflower production. This project is the only independent source of informationon sunflower cultivar performance, available to farmers. The aim of this project is to evaluate commercially available sunflower cultivars at different localities in collaboration with seed companies. During the2015/2016 season, 21 cultivars were evaluated in 11 planting date-locality field trials. The highest trialmean yield of 3.12 t ha-1 was obtained at Boskop and the lowest of 0.79 t ha-1, at Reitz. The six bestperforming cultivars, in terms of average yield calculated over localities, were PAN 7080, P 65LL14, P65LC54, PAN 7160 CLP, P 65LL02 and PAN 7095 CL. Six Clearfield cultivars namely: NK ADAGIO CL, P65LC54, PAN 7095 CL, PAN 7102 CLP, PAN 7160 CLP and SY 3970 CL were entered. Four of thesecultivars namely, P 65LC54, PAN 7095 CL, PAN 7102 CLP and PAN 7160 CLP had above average yields.The probability to obtain an above average yield were calculated for all cultivars across a range of yieldpotentials. Cultivars can be selected by comparing their yield probabilities.

Soybean cultivar selection for improved yield and yield stability

Dr R van der Merwe and C Basson
University of the Free State

Introduction

Yield instability across locations and seasons makes it difficult to identify one high yielding soybean cultivar that shows good yield potential and stable yields at one specific location or adaptation across various locations. Grain yield is a complex trait and in order to estimate yield, various yield components need to be considered. Since yield 7components are of quantitative nature it is necessary to acquire information about the nature and magnitude of genetic variability present in the available cultivars and also to know the interrelationships among yield components and their direct effects on yield. Since yield components are expected to be more reliable indicators for the expression of yield than grain yield per se (Burton 1987), the aim of this study is to identify the most stable yield component(s) that has a significant and direct effect on grain yield, and that can be successfully applied to select for stable high yielding genotypes in cultivar trials.

Materials and methods

Plant material

The experimental material consists of 18 genotypes of soybean (Glycine max L. Merrill). Genotypes include registered soybean cultivars that are commercially available as well as large-seeded cultivars that show potential for registration on the variety list. Genotypes were selected upon the following criteria: maturity type, growth habit, seed yield or yield potential, yield reliability and seed shattering.

Field trials

Trials will be planted for three consecutive seasons (2015/2016, 2016/207 and 2017/2018) with the first planting com­mencing in mid-November 2015. The three locations include Petrusburg, Potchefstroom and Bethlehem; this represents the warm-, moderate- and coolproduction areas, respectively. The trial design is a randomised complete block with three replications. Depending on the amount of seed available, each genotype will be raised in six rows of 5 m in length with a between row spacing of 0.75 m. Each plot, sown by hand, will be over-planted to compensate for germination rate and will be thinned after seedling emergence to 20 plants rrr². Seeds will be inoculated with Bradyrhizobium japonicum (strain WB74) at planting by use of liquid inoculation, applied in the plant furrow.

Standard agronomic practises will be followed for growing soybean. Fertilizer and pesticides will be applied at rates recommended to ensure optimal yield. Supplementary irrigation will be supplied to ensure optimal seedling emergence and crop production.

Data collection and statistical analysis

Parameters collected during crop growth will include average days to flowering, average days to R6 growth stage and average days to maturity (R8). Twenty randomly sampled plants will be taken, at harvest, from the four middle rows of each plot. Average plant height (cm), average pod height (cm), average number of branches per plant, average number of reproductive nodes per plant, average number of pods per reproductive node, average number of pods per plant, average seeds per pod, average seeds per plant, average seed weight (g) per plant, average seed size per plant, harvest index per plant as well as pod shattering percentage per plot at harvest and 3 weeks after harvest will be collected. On a plot area basis, number of nodes, number of pods and number of seed will be recorded. The weight of 100 seed will be recorded as the average of three 100-seed samples. Seed mass per plot will be transformed to grain yield ha'¹.

Data will be subjected to analysis of variance (ANOVA) and the means will be tested for significance. Coefficient of variation (%) and broad sense heritability (H²%) will be determined for each measured trait from the results of the ANOVA. Genotypic and phenotypic correlation coefficients will be determined from the mean values of all traits. Stepwise regression will be applied to estimate the direct effect of each yield component (independent variable) on grain yield (the dependent variable). Statistical analyses will be conducted using GenStat and Agrobase software.

Anticipated results

Analysis of variance

The level of genetic diversity among the 18 cultivars for grain yield and yield components will be revealed with the analyses of variance. The presence and magnitude of genetic variability in a gene pool is a pre-requisite of a soybean breeding programme. In addition, the knowledge of certain genetic parameters is essential for proper understanding and their manipulation in any crop improvement programme (Aditya et al. 2011).

Heritability, genetic variance and phenotypic variance estimates will be revealed for all traits analysed and these will be useful for further determination of genetic variability. Determining the magnitude of these parameters allows for the identification of traits that are effective for selection. A high heritability estimate for a specific trait indicate that this trait is less influenced by environmental conditions and would be a favourable selection criterion.

Correlation co-efficients

This analysis will indicate the strength of the relationship between the yield components as well as the magnitude and direction of changes expected during selection. When a specific yield component shows a negative correlation with other yield components and also with grain yield, this might indicate that this yield component contributes no value for selection for grain yield in soybean. Literature has indicated that yield components such as pods per plant and seeds per plant have shown strong positive correlations with grain yield (Arshad et al. 2014; Ghodrati et al. 2013). However, correlation coefficients for these traits need to be evaluated for South African genotypes and under South African production conditions. Results from this analysis will indicate which yield component shows strong positive correlations with other yield components and grain yield and that shows value for selection for yield.

Stepwise regression

This analysis will indicate which yield component (from all components tested) has the highest direct effect on grain yield (t ha'¹). Literature has indicated that, among other traits, pods per plant, 100-seed weight and seeds per pod have shown strong direct effects on grain yield (Arshad et al. 2014). However, these influences have not been tested on South African genotypes and neither under South African production conditions. In addition, the heritability of the yield component should also be considered. For example, if a trait shows that it has a strong positive effect on grain yield, but it has a low heritability, then this trait would not be an effective selection measure for grain yield. Combined results from all analyses will indicate which yield component can serve as an effective selection criterion for high and stable grain yield under South African production conditions.

The dynamics of research uptake by South African oilseeds producers

Dr N Boshoff
Stellenbosch University

This project was terminated due to a lack of progress.

© 2019   OIL & PROTEIN SEEDS DEVELOPMENT TRUST (OPDT) / OILSEEDS ADVISORY COMMITTEE (OAC)  TIGME.COM – PLEASE REPORT ANY ERRORS TO support@tigme.com