Crop Science Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 27 October 2005
Published in Crop Sci 45:2629-2635 (2005)
© 2005 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Teklewold, A.
Right arrow Articles by Becker, H. C.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Teklewold, A.
Right arrow Articles by Becker, H. C.
Agricola
Right arrow Articles by Teklewold, A.
Right arrow Articles by Becker, H. C.
Related Collections
Right arrow Other Oil Crops
Right arrow Crop Genetics

CROP BREEDING, GENETICS & CYTOLOGY

Heterosis and Combining Ability in a Diallel Cross of Ethiopian Mustard Inbred Lines

Adefris Teklewold* and Heiko C. Becker

Georg-August Univ., Institute of Agronomy and Plant Breeding, Von-Siebold Str-8, 37075, Göttingen, Germany

* Corresponding author (adecher{at}yahoo.co.uk)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Heterosis is commercially exploited in rapeseed (Brassica napus L.) and its potential use has been demonstrated in turnip rape (B. rapa L.) and Indian mustard (B. juncea L.). In Ethiopian mustard (B. carinata A. Braun), however, information regarding heterosis has not been previously reported. This study, therefore, was conducted to generate information on heterosis and combining ability in B. carinata. Nine inbred parents and their 36 F1s, obtained by half-diallel cross, were evaluated for 12 traits at three locations in Ethiopia. Analysis of variance showed the presence of significant heterosis for all the traits. Seed yield showed the highest relative mid-parent heterosis that varied from 25 to 145% with a mean of 67%. Relative high-parent heterosis for seed yield varied from 16 to 124% with a mean of 53%. General combining ability (GCA) effects were predominant in all traits except secondary branches and pods per plant. Specific combining ability (SCA) was significant for days to flowering, secondary branches, pods per plant, pod length, seeds per pod, 1000-seed weight and oil content. Interaction effects of GCA x location were significant for all traits except days to flowering, days to maturity, and oil content. All traits had significant SCA x location interaction effects. GCA effect for seed yield was positively correlated with F1 performance (r = 0.77) and absolute mid-parent heterosis (r = 0.67). The presence of high levels of mid- and high-parent heterosis indicates a considerable potential to embark on breeding of hybrid or synthetic cultivars in Ethiopian mustard.

Abbreviations: AHPH, absolute high-parent heterosis • AMPH, absolute mid-parent heterosis • CV, coefficient of variation • GCA, general combining ability • H2, broad sense heritability • HPV, high-parent value • LSD, least significant differences • MPV, mid-parental value • RHPH, relative high-parent heterosis • RMPH, relative mid-parent heterosis • SCA, specific combining ability


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
ETHIOPIAN MUSTARD is one of the oldest oil crops cultivated in Ethiopia (Simmonds, 1979) but is practically not cultivated in any other part of the world. The seed of B. carinata contains high erucic acid and glucosinolate (Becker et al., 1999) that limited the spread of its cultivation to other parts of the world. Because of its drought and heat tolerance, the crop is now considered as an alternative to B. napus and B. juncea in dryer areas of Canada (Rakow, 1995), Spain (Velasco et al., 1995), Australia (Fletche, 1997), India (Singh, 2003), the USA, and Italy (Cardone et al., 2003). In recent years, substantial efforts were made to improve both the quality and quantity of oil and seed and/or transfer its useful traits to related Brassica oil crops (Rakow, 1995; Velasco et al., 1995; Meng et al., 1998; Singh, 2003).

So far, line breeding and, to some extent, mass selection are the dominant breeding methods used to improve Ethiopian mustard. However, development of synthetic or hybrid cultivars have been successful in other oilseed Brassica ssp. (Melchinger and Gumber, 1998; Becker et al., 1999; Miller, 1999). Studies in B. napus, B. rapa, and B. juncea have indicated high levels of heterosis. Pradhan et al. (1993) reported 29 to 92% heterosis over the best-yielding parents in B. juncea by crossing parents of Indian and exotic origin. Banga and Labana (1984) reported up to 200% heterosis in B. juncea. Brandle and McVetty (1989) reported high-parent heterosis reaching 120% for seed yield in B. napus. In summer turnip (B. rapa), high-parent heterosis for seed yield reaching 60% was reported by Schuler et al. (1992) and Falk et al. (1994). Consequently, in several countries hybrid cultivars played an important role in the expansion of B. napus cultivation (Becker et al., 1999; Miller, 1999). Heterosis can be partially utilized also by developing synthetic cultivars (Becker et al., 1998).

Although combining ability studies in oilseed Brassica spp. are scanty, most of these studies emphasized the preponderance effect of GCA on yield and most of the yield components indicating the importance of additive gene action (Brandle and McVetty, 1989; McGee and Brown, 1995; Wos et al., 1999). On the other hand, Pandey et al. (1999) reviewed evidences for the presence of significant SCA effects for yield and yield components. Ramsay et al. (1994) reported that variation for both GCA and SCA were responsible for dry matter yield and other quantitative traits in B. napus.

The commercial use of synthetic and hybrid cultivars in B. napus is a reality (McVetty, 1995; Becker et al., 1999) and their potential use have been experimentally demonstrated in B. rapa and B. juncea (McVetty, 1995; Pandey et al., 1999). Ethiopian mustard (BBCC) sharing one of its genome with B. juncea (AABB) and the other with B. napus (AACC) (U.N., 1935) could be amenable for heterosis breeding as to its close relatives. But, information regarding heterosis is not available. Therefore, this study has the following objectives: (i) to determine the level of mid- and high-parent heterosis for yield and its components; and (ii) to estimate the relative importance of GCA and SCA variances.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Plant Materials
The nine parental inbred lines (all S5) used in this study were selected from progenies derived from 36 accessions collected from different geographic areas distributed over most parts of Ethiopia with large variation in altitude of collection sites (Table 1). The parental lines were chosen in a systematic random way to represent both the geographic and phenotypic diversity.


View this table:
[in this window]
[in a new window]
  		Table 1. Parental inbred lines of Ethiopian mustard used in the study with their accession code, area of collection, and altitude.

 
Crossing between the parental inbred lines was made in a lath house (Ethiopia) during 2003 off-season in half-diallel (without reciprocals). Crossing was done by hand emasculation and bud pollination on 10 to 15 plants per line. During the same period, seeds of the parental inbred lines were increased by growing 20 plants of each line under a separate steel cage covered with light transparent white mesh cloth in the lath house. At flowering, plants in each cage were gently shaken using a separate wooden stick to facilitate pollen transfer and seed set.

Field Experiments
During the main season of 2003 (June–December), the 36 F1 along with their nine parents were grown at three locations (Holetta, Kulumssa, and Debrezeit) in Ethiopia. Holetta (38° 30' E and 9°00' N) is located at 2400 m a.s.l. (above sea level) with long-term annual precipitation of 1095 mm and soil type of Eutric Nitosol. Kulumssa (39°9' E and 8°1' N) is located at 2190 m a.s.l. with long-term annual precipitation of about 830 mm and soil type that is intergraded between a Eutric Nitosol and Luvic Phaeozem. Debrezeit (38°59' E and 8°46' N) is located at 1900 m a.s.l. with mean annual precipitation of about 800 mm (5-yr mean) and soil type of Chromic Vertisol. The experiments at all three locations were laid out in randomized complete block design with three replications. Plots consisted three rows of 5-m length and 30-cm interrow and 10-cm intrarow spacing. The two outer rows of each plot were planted by the genotype PGRC/E 2085, characterized by shorter plant height (112 cm) and intermediate maturity (144 d to maturity) to provide uniform competition. Crop management factors like land preparation, crop rotation, fertilizer, and weed control were followed as recommended for each location.

Data Collection and Analysis
In all the locations, except days to flowering and maturity that were recorded on plot bases, the other phenotypic traits were recorded from the same 10 randomly selected plants of the central row as follows: days to flowering (days from sowing until about 50% of the plants flower); days to maturity (days from sowing until about 90% of the pods mature); number of primary branches per plant (productive branches originating from the main stem); number of secondary branches per plant (productive branches developed from the primary branches); number of pods per plant (pods borne on both primary and secondary branches); pod length (length of six individual pods per plant, two each from bottom-, middle-, and top-borne branches); plant height (main stem length); number of seeds per pod (obtained from the same pods used to estimate pod length); seed yield (adjusted to 70 g kg–1 moisture); 1000-seed weight (g); percentage of oil content [proportion of oil content from 22 g of oven-dried seeds determined by nuclear magnetic resonance spectrometry (Madson, 1976)]; and oil yield (seed yield per plant x percentage of oil content).

A separate variance analysis for each location was run, and after testing homogeneity of error variance, a combined analysis was done by PLABSTAT software (Utz, 2001) based on the model: Yijk = µ + li + gj + glij + bik + {epsilon}ijk, where Yijk = observation of genotype j in location i, and replication k; µ = the general mean; li = effect of location i; gj = effect of genotype j; glij = the interaction effect between location i and genotype j; bik = effect of block k in location i; and {epsilon}ijk = error of observation ijk. The sum of square for genotypes was partitioned into parent, F1s, and parents vs. F1s.

After detecting significant F values for the F1s, a combining ability analysis was performed by AGROBASE software (Agronomix Software Inc., Winnipeg, Manitoba, Canada, 1999) following Method IV Model II of Griffing (1956) for each location separately. Later, the results were combined over the three locations following the model: Yijk = µ + gi + gj + sij + lk + (gl)ik + (gl)jk + (sl)ijk + {epsilon}ijk, where Yijk = observation in location k of parents i and j; µ = the general mean; gi or gj = GCA effect of parents i or j; sij = SCA effect of the cross between parents i and j; lk = effect of location k; (gl)ik or (gl)jk interaction effect between GCA of parent j or i with location k; (sl)ijk = interaction effect between SCA of cross ij and location k; and {epsilon}ijk = error of observation ijk (pooled estimate from analyses of individual locations). F values for testing combining abilities were calculated as follows: {sigma}2scal = MSscal/MS{epsilon}; {sigma}2gcal = MSgcal/MSscal; {sigma}2sca = MSsca/MSscal; {sigma}2gca = /; where {sigma}2gca, {sigma}2sca, {sigma}2gcal, and {sigma}2scal are variances due to GCA, SCA, GCA x location and SCA x location, respectively, and MSgca, MSsca, MSgcal, MSscal, and MS{epsilon} are mean squares due to GCA, SCA, GCA x location, SCA x location and error, respectively. Broad sense heritability (H2) was estimated as H2 = . Heterosis is estimated as follows: absolute mid-parent heterosis = F1 – MPV; relative mid-parent heterosis = x 100; absolute high-parent heterosis = F1 – HPV; relative high-parent heterosis = x 100.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Mean Performance and Variance Analysis
In all traits, except days to flowering, the F1s had greater mean values than the parents (Table 2). The superiority of the F1s over the parents was high for seed yield, secondary branches, and pods per plant, but low for primary branches, seeds per pod, pod length, and oil content. In both parents and F1s, CV was relatively higher for secondary branches (19.3%), pods per plant (18.1%), seed yield (22.0%), and oil yield (22.6%), but low for days to flowering, days to maturity, and pod length. Parental CV for oil content was higher than the F1s.


View this table:
[in this window]
[in a new window]
  		Table 2. Minimum, maximum, mean, CV (%) and LSD (5%) for 12 traits of nine parental lines of Ethiopian mustard and their 36 F1s grown at three locations in Ethiopia (2003–2004).

 
The pooled analysis of variance over the three locations indicated significant differences between locations, genotypes, parents, and crosses for all traits (data not presented). The comparison of parents vs. F1s, which indicates the presence of heterosis, was also highly significant (P ≤ 0.01) for all traits. Genotype x location and F1s x location interaction effects were significant (P ≤ 0.01) for all traits. Parents x location interaction effect was significant (P ≤ 0.01) to all traits except for primary branches. Similarly, parent vs. F1s x location interaction effect was significant (P ≤ 0.01) for all traits except for primary branches and pod length.

Estimates of Combining Ability and Other Genetic Parameters
Results from the partitioning of the sum of squares of the F1s into GCA and SCA and their interaction effect with location are presented in Table 3. Mean squares of GCA were significant for all traits except secondary branches and pods per plant. Mean squares of SCA were not significant in five of the traits: primary branches, plant height, days to maturity, seed yields, and oil yields. For all traits with significant GCA and SCA mean squares, GCA variance was higher than SCA, and mostly GCA variance was more than triple the SCA variance component (details not given). All traits except days to flowering, days to maturity, and oil content had significant mean square for GCA x location interaction. Mean squares for SCA x location interaction were significant for all traits. In some traits like secondary branches, pods per plant, seed yields, and oil yields, the interaction variances of GCA and SCA x location were substantially high and sometimes the magnitude was even higher than the main GCA and SCA variances (details not given). Estimates of H2 varied from 0.21 for secondary branches to 0.98 for days to flowering (Table 3). Seed yield had intermediate H2, 0.51. Except secondary branches, pods per plant, and oil yield, estimates for H2 were generally high (>0.83) for the other traits.


View this table:
[in this window]
[in a new window]
  		Table 3. Mean squares from analysis of variance and broad sense heritability (H2) for 12 traits in 36 F1s of Ethiopian mustard grown at three locations in Ethiopia (2003–2004).

 
Variation in Heterosis and Correlations between Heterosis, General Combining Ability, Mid-Parental Value, and F1 Performance
Minimum, maximum, and mean values for absolute mid-parent heterosis (AMPH), relative mid-parent heterosis (RMPH), absolute high-parent heterosis (AHPH), and relative high-parent heterosis (RHPH) are given in Table 4. The magnitude of heterosis was variable for the different traits and cross combinations. The largest heterosis was observed for seed yield, followed by oil yield. All the F1s RMPH and RHPH for seed yield that varied from 25.1 to 145.4 with a mean of 67% (Fig. 1) , and from 16.2 to 123.6 with a mean of 52.8%, respectively. Oil yield and pods plant–1 showed a mean of 65.9 and 37.2% RMPH, and 51.8 and 25.1% RHPH, respectively. Secondary branches and pods plant–1 had intermediate RMPH and RHPH with a mean of 27.4 and 37.2% and 17.7 and 25.1%, respectively. Expression of both RMPH and RHPH for other traits were low. Mean RMPH was negative only for days to flowering. But in the case of RHPH, six traits (days to flowering, primary branches, days to maturity, pod length, seeds per pod, and 1000-seed weight) had negative mean values.


View this table:
[in this window]
[in a new window]
  		Table 4. Minimum (Min.), maximum (Max.), mean, and standard error (SE) for absolute mid-parent (AMPH), high-parent (AHPH), relative mid-parent (RMPH), and high-parent (RHPH) heterosis for 12 traits in 36 Ethiopian mustard F1s grown at three locations in Ethiopia (2003–2004).

 


View larger version (54K):
[in this window]
[in a new window]
  		Fig. 1. The relative contribution of mid-parental values (MPV) and absolute mid-parent heterosis (AMPH) to the observed seed yield in 36 Ethiopian mustard F1s grown at three locations in Ethiopia (2003–2004).

 
The correlation between GCA and absolute mid-parent heterosis (AMPH) was positive and highly significant for seed yield, seeds per pod, and oil yield, but negative for 1000-seed weight (Table 5). The highest correlation between GCA and AMPH (r = 0.67) was observed for seed yield (Fig. 2) . GCA and MPV were significantly and positively correlated for all traits. For all traits, correlation coefficients between GCA and F1 performance were positive and significant, ranging from r = 0.42 for secondary branches to r = 0.99 for days to flowering. As shown in Fig. 3 , the correlation between GCA and F1 performance was quite high (r = 0.77, P ≤ 0.01) for seed yield. The correlation coefficients between MPV and AMPH were either significantly negative or nonsignificant except for pod length. MPV and F1 performance were correlated significantly and negatively for secondary branches and pods per plant, but significantly and positively for pod length and 1000-seed weight. The correlation between AMPH and F1 performance was positive and statistically significant (P ≤ 0.01) for all the traits except days to flowering, primary branches, plant height, days to maturity, pod length, and 1000-seed weight. This association was very strong in some cases like seed yield (r = 0.93), oil yield (r = 0.89), and seeds per pod (r = 0.83).


View this table:
[in this window]
[in a new window]
  		Table 5. Phenotypic correlation coefficients between different parameters for 12 traits in 36 Ethiopian mustard F1s grown at three locations in Ethiopia (2003–2004).

 


View larger version (13K):
[in this window]
[in a new window]
  		Fig. 2. Relationship between GCA effect (sum of the two parents) and absolute mid-parent heterosis (AMPH) for seed yield per plant in 36 Ethiopian mustard F1s grown at three locations in Ethiopia (2003–2004). ** significant at P ≤ 0.01.

 


View larger version (12K):
[in this window]
[in a new window]
  		Fig. 3. Relationship between GCA effect (sum of the two parents) and F1 performance for seed yield per plant in 36 Ethiopian mustard F1s grown at three locations in Ethiopia (2003–2004). ** significant at P ≤ 0.01.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
The study indicated hybrid vigor to be an important factor in increasing seed yield, and on the average, F1s showed 67% higher yield than the mean of their parents. All the F1s had positive RMPH and RHPH for seed yield, and 67 and 41% of the F1s showed more than 50% RMPH and RHPH, respectively. Four and two F1s gave higher than 100% RMPH and RHPH, respectively. Heterotic response in this study will likely be maximized because: (i) the F1 seed was obtained by hand pollination that ensure 100% purity in seed; (ii) mid-parental values were relatively low, as none of the parents were preselected for their per se performance; and (iii) the parental divergence was high as the parents were originated from seven different geographic regions in Ethiopia. Heterosis was estimated from only 10 plants per plot, but the average heterosis for each of the 36 F1s is estimated from three replications at three locations. In B. juncea, Pradhan et al. (1993) pointed out that the level of yield obtained at single plant basis is higher than that obtained from field level.

The 67% mean RMPH for seed yield observed in this study is comparable to the 77% in B. juncea (Pradhan et al., 1993) and 53% in B. napus (Lee et al., 1980), but was less than the 81% of Diers et al. (1996) and 89% of Riaz et al. (2001) in B. napus. The present RMPH, however, was higher than the level reported for non-inbred (varietal/cultivar) parents derived F1s of 43% (Lefort-Buson et al., 1985), 15% (Leon, 1991); and 42% (Diers et al., 1996) in B. napus and 19% (Banga and Labana, 1984) in B. juncea. The mean RHPH (53%) reported here is in the range of 50% (Pradhan et al., 1993) in B. juncea, and 69% (Brandle and McVetty, 1989) and 67% (Riaz et al., 2001) in B. napus. In B. rapa, Schuler et al. (1992) and Falk et al. (1994) reported RHPH for seed yield that ranged from 0 to 64%.

Large-scale use of heterosis requires the production of a large quantity of seed and a sufficient level of heterosis. The level of mid- and high-parent heterosis observed in this study could make heterosis breeding an attractive option for Ethiopian mustard yield improvement. However, the immediate exploitation of heterosis by developing hybrid varieties is limited because of the unavailability of suitable pollination control mechanisms (sterility systems) that ensure cross pollination in the crop.

The preponderance of GCA variance for all traits except secondary branches and pods per plant demonstrates the predominance of additive gene effects. Hence, the best performing F1 may be produced by crossing parents with the highest GCA. The predominance of additive genetic variance for the traits also means that, besides hybrid and synthetic breeding, opportunity exists for genetic improvement by accumulating favorable alleles from the inter-regional variability through selection. The predominance of SCA variance for secondary branches and pods per plant denotes that nonadditive gene effects were largely influencing the expression of these two traits; hence, selection will bring no or slow genetic improvement. The lack of significant GCA variance for secondary branches and pods per plant may be partly explained by the very high GCA x location interaction variance.

The primary criteria for choosing parents that might have high heterotic response and subsequently produce superior yielding F1 would be the GCA effect of the parents. This relationship could be used to reduce cost when GCA is estimated from a large number of lines with a limited number of testers. Although GCA variance was predominant for seed yield, no evidence was found that high-yielding parents tended to produce high-yielding F1s. But there existed a weak trend of high-yielding parents being less heterotic. This relationship, however, does not exclude the occurrence of a high level of heterosis in high-yielding parents (Fig. 1). For example, the highest AMPH for seed yield was observed from parents with low MPV, while the second highest from parents with average MPV. Moreover, among parental combinations with the highest five MPV, two gave 84 and 97% RMPH. The strong correlation between AMPH and F1 performance observed in this experiment implies that heterosis was a significant component of variation among the F1s.

In the absence of an established pollination control mechanism, the observed heterosis could at least partly be utilized through development of synthetic cultivars. When a strict uniformity of product is not a requisite, seed of synthetics are simpler to produce than hybrids and can be multiplied by smallholder farmers. The heterogeneous and heterozygous genetic constitution of synthetics contributes to their yield stability over environments (Becker et al., 1998). As synthetics mainly exploit that part of heterosis contributed by GCA, the predominance of GCA variance observed for seed yield and the other traits also encourage synthetic varieties development. For a practical utilization of heterosis in synthetic varieties, a high out-crossing rate is an essential factor (Becker et al., 1998). Like its amphidiploid relatives, B. napus and B. juncea, Ethiopian mustard possesses on average about 29% cross-pollination, but has been observed to possess as high as 39% cross-pollination (Teklewold, unpublished, 2005). To ensure a maximum level of intercrossing during the multiplication of synthetics, lines with a high level of out-crossing could be selected. Becker et al. (1998) reported a high genetic variation for out-crossing rate in B. napus double haploid lines.

The average RHPH for oil content was virtually zero, although some F1s expressed positive RMPH as high as 8.5% (data not presented). Negative or absence of heterosis for oil content is a common phenomenon in oil seed Brassicas (Banga and Labana, 1984; Brandle and McVetty, 1990; Schuler et al., 1992; Falk et al., 1994). Heterosis for oil content could be much appealing, but the available experience in B. napus indicates that it is not an essential prerequisite for the success of hybrids. Heterosis for plant height is a common phenomenon in B. napus hybrids (Leon and Becker, 1995) and was also observed in this experiment, but the relative 12% of RMPH and 4.8% of RHPH is not a major concern toward the higher lodging susceptibility associated with increased plant height, as most of the hybrids were within the range of the parents.

As the ecological settings of the trial sites are different (refer to the Materials and Methods section), the performance of parent and F1s was variable across locations. This presupposes a multilocation genotype performance assessment in the initial stages of evaluation. A higher genotype x location interaction variance for F1s than for parents observed in this experiment disagrees with the contemporary stability philosophy of hybrids. In B. napus, hybrids were found to be relatively homeostatic (Brandle and McVetty, 1989; Leon, 1991; Frauen et al., 2003) and buffered more against environmental variations than their inbred parents. The preponderance interaction effect of parents with location requires growing test crosses in various locations to select potential parents for hybrid combination. Similarly, the significant location x F1s interaction effect indicates lack of consistency in expression of heterosis across locations. Ultimately, the development of different F1s would increase genetic gain for each location. However, the extra gains should overweigh the extra cost of having different hybrid varieties for each location.

In conclusion, the levels of heterosis in Ethiopian mustard observed in this experiment are comparable to what was reported for the other oilseed Brassicas, indicating a considerable potential to embark on hybrid or synthetics breeding in this species too. The maximum of 145% RMPH and 124% of RHPH indicates the potential for increasing yield by a systematic search for heterotic groups and testing parents for their combining ability.


   ACKNOWLEDGMENTS
 
The Catholic Academic Exchange Service (KAAD), Bonn, Germany, granted the first author a scholarship. We are grateful for the technical help given by Tadesse Debele, Kassahun Kumssa, Tadesse Deme, Wondemagne Weldesemyat, and the Kulumssa Research Center oil crops research staffs during the fieldwork in Ethiopia.

Received for publication January 26, 2005.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Teklewold, A.
Right arrow Articles by Becker, H. C.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Teklewold, A.
Right arrow Articles by Becker, H. C.
Agricola
Right arrow Articles by Teklewold, A.
Right arrow Articles by Becker, H. C.
Related Collections
Right arrow Other Oil Crops
Right arrow Crop Genetics

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Vadose Zone Journal
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome