Crop Science Illumina
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 27 March 2006
Published in Crop Sci 46:1130-1136 (2006)
© 2006 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
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 Google Scholar
Google Scholar
Right arrow Articles by Pixley, K. V.
Right arrow Articles by Tongoona, P.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Pixley, K. V.
Right arrow Articles by Tongoona, P.
Agricola
Right arrow Articles by Pixley, K. V.
Right arrow Articles by Tongoona, P.
Related Collections
Right arrow Maize
Right arrow Plant Disease
Right arrow Crop Genetics

CROP BREEDING & GENETICS

Improvement of a Maize Population by Full-Sib Selection Alone versus Full-Sib with Selection during Inbreeding

Kevin V. Pixleya,*, Thanda Dhliwayob and Pangirayi Tongoonac

a International Maize and Wheat Improvement Center (CIMMYT), Apdo Postal 6-641, 06600, Mexico D.F., Mexico
b Department of Agronomy, Iowa State University, Ames, Iowa, IA 50011, USA
c Department of Plant Pathology, African Centre for Crop Improvement, University of KwaZulu-Natal, P Bag X01, Scottsville, Pietermaritzburg 3209, Rep. South Africa

* Corresponding author (k.pixley{at}cgiar.org)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Maize (Zea mays L.) improvement involves formation, evaluation, selection, and recombination of genetically variable families or inbred lines, and because cultivars must combine many desirable traits, the process can be complicated and lengthy. This study compared 12 experimental open-pollinated varieties (OPVs) developed from maize population Pool 9A by two approaches, full-sib (FS) selection alone and FS selection combined with inbreeding with selection for Maize streakvirus (MSV) resistance. The experimental OPVs were evaluated in 13 environments in Africa, including three sites with artificial MSV inoculation. Full-sib selection combined with inbreeding and selection for MSV resistance [with mild selection for gray leaf spot resistance (GLS, caused by Cercospora zeae maydis Tehon & Daniels)], reduced ear height and reduced days to flowering] resulted in OPVs with similar grain yield but superior MSV resistance (P < 0.01), fewer days to anthesis (P < 0.01), and lower ear height (P < 0.05) than OPVs developed by FS selection alone. Each generation (from S0 to S3) of inbreeding with selection resulted in OPVs that were 16, 8, 2, and 1% improved for MSV and GLS resistance, ear height, and days to anthesis, respectively. Our results demonstrate improvement of a maize population for MSV resistance and other traits by selection during inbreeding (from S0 to S3), without negative impact on gains for grain yield achieved by evaluation and selection among the progenitor FS families.

Abbreviations: CIMMYT, International Maize and Wheat Improvement Center • ET, turcicum or northern leaf blight • FS, full-sib • GLS, gray leaf spot • masl, meters above sea level • IITA, International Institute of Tropical Agriculture • MSV, Maize streakvirus • OPV, open-pollinated variety


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
DEMAND for maize grain is projected to increase from 1995 levels by 50% globally by 2020, including 93% in sub-Saharan Africa, 92% in South Asia, 62% in Latin America, and 46% in East and Southeast Asia (IFPRI, as cited by Pingali and Pandey, 2001). It is therefore alarming that, although average maize yield in USA and other high-income countries is 8.3 Mg ha–1, most of Africa and large areas of Asia and South America achieve yield below 2.0 Mg ha–1 (Aquino et al., 2001). Part of this yield difference is explained by use of unimproved varieties in much of the nontemperate maize growing world, where about 50% of all maize is planted with farm-saved grain (not purchased seed), while 33% is planted with hybrids, and 15% with improved open-pollinated varieties (OPVs) (Morris, 2001). Many public breeding programs, including CIMMYT and IITA, invest considerable resources to develop and promote improved OPVs as an intermediate technology between landraces and hybrid varieties. Improved OPVs offer more yield potential than landraces, with less risk than hybrids regarding availability or access to seed by resource-poor or marginalized farmers (Pixley and Bänziger, 2004; Pixley, 2006).

Maize OPVs can be products of intra- or interpopulation improvement within broad- or narrow-based populations or existing OPVs. For example, an international progeny testing trial (IPTT) scheme has been successfully used by CIMMYT to improve tropical maize populations while generating experimental OPVs with good grain yield and stability-conferring biotic and abiotic stress tolerances (Vasal et al., 1982; Pandey et al., 1986, 1987; Byrne et al., 1995). A comprehensive review of published results for recurrent selection programs indicates an average grain yield gain of 79 kg ha–1 yr–1 for 80 intrapopulation studies (87 kg ha–1 yr–1 for 27 FS studies), and average gains of 98, 93, and 43 kg ha–1 yr–1 for interpopulation, S1 and S2 recurrent selection studies, respectively (Coors, 1999). If more than one selection method is applied simultaneously or in tandem, theory suggests that expected progress will be the sum of expected progress for each individual method (Hallauer and Miranda, 1981); however, if selection is for different traits, progress achieved for one trait may reduce, negate, or reverse progress for the other.

Successful OPVs can also be developed as products of hybrid-oriented, pedigree breeding research in which (i) synthetic varieties are formed among lines with good general combining ability or (ii) variety crosses are made between synthetic varieties representing complementary heterotic groups (Pandey and Gardner, 1992). Most new experimental OPVs developed at CIMMYT now arise as products from such pedigree breeding research.

Heritability estimates for MSV resistance have been high, e.g., 83% (Kim et al., 1989), 62 and 93% (Welz et al., 1998), 73 to 98% (Pernet et al., 1999a), and 71 to 98% (Pernet et al., 1999b), with one major QTL (Kyetere et al., 1995; Kyetere et al., 1999) and two, three, or "few" total genes implicated (Efron et al., 1989; Kim et al., 1989). Many maize OPVs or populations have been successfully improved for MSV resistance by a variety of methods (see review by Efron et al., 1989). Tang and Bjarnason (1993) improved the maize population ‘La Posta’ using backcross and S1 family evaluations for MSV resistance while conducting multilocation yield trials of the parental full-sib families. We had intended to use a similar scheme in the study reported herein (recombine S1 lines derived from full-sib families selected through international yield trials), but instead, while awaiting yield trial data for the full-sib families, the S1s were screened for MSV resistance and advanced to S2, and the S2s were further selected for MSV resistance and advanced to S3. As happened to us, it is not unusual for time to complete a cycle of recurrent selection to be delayed beyond the breeder's expectation. For trials evaluated in several countries, for example, delays may be caused by differences in cropping seasons, protocols to satisfy plant quarantine regulations, or slow communication of or about data.

The objective of this study was to compare grain yield, MSV resistance and several agronomic traits for 12 experimental varieties formed by full-sib selection in a maize population, with zero, one, two, or three generations of inbreeding under artificial selection for MSV resistance. Our hypothesis was that additional generations of selection pressure for MSV resistance and good general plant type (e.g., less lodging, less ear rot, etc.) would compromise the gains for grain yield possible by simply recombining the best full-sib families from international yield trials.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Germplasm and Selection Methods
CIMMYT's maize ‘Pool 9A’ has white and dent grain and is adapted to tropical transition zone to highland environments [{approx}1600–2200 m above sea level (masl)]. Its constituent germplasm includes materials from highlands of western Kenya (Kitale synthetics), Ecuador (Ecuador 573), and Mexico (Tuxpeño de Altura), plus germplasm adapted to mid-altitudes ({approx}1500 masl) of Zimbabwe (SR52) (Beck, 2001). Pool 9A is particularly suited to the highlands of Ethiopia, where it is a released variety known as ‘Kuleni’. Because of concerns that MSV is increasingly encroaching into highland environments, Pool 9A has been improved by CIMMYT by crossing it with a MSV-resistant IITA mid-altitude maize population, ‘TZMSR-W’, followed by three cycles of intrapopulation S1 recurrent selection under artificially induced MSV epidemics. To take advantage of yield and agronomic gains from continued work in Mexico with the original, MSV-susceptible Pool 9A, the MSV-resistant version of Pool 9A, ‘PL9A-SR’, was backcrossed at CIMMYT (Zimbabwe) to the most recent improved cycle (cycle seven) of Pool 9A from CIMMYT's highland research program in Mexico (recurrent parent). This backcross population, PL9A-SR(BC1), was then subjected to full-sib intrapopulation improvement for grain yield, MSV resistance, and other agronomic traits (see below). One hundred fifty full-sib families from PL9A-SR(BC1) were evaluated in replicated yield trials at three locations (Namulonge, Uganda; Mbeya, Tanzania; and Bako, Ethiopia) in 1996.

While the yield trials were grown and data were returned from the three locations in eastern Africa, three seasons of disease nurseries at Harare (Zimbabwe) evaluated the 150 full-sib families, S1, and finally S2 lines under artificially induced MSV epidemics under which best plants were self-pollinated and advanced ear to row to finally reach S3. During the inbreeding process, both intra- and mild interfamily selection were practiced, resulting in a few families being eliminated because of susceptibility to MSV (symptom scores >3.5; see description of 1–5 scale, below), or other serious defects (e.g., susceptibility to gray leaf spot or severe root lodging). Thus, the number of and selection intensity among inbred families varied for the 150 parental full-sibs.

A weighted selection index was used to select 15 full-sib families (i.e., 10% selection intensity) to constitute (using remnant seed of the FS families) an experimental OPV, ‘AC969A(FS)-SR’, that optimized performance of full-sib (FS) families across (AC) all 1996 (96) sites of the PL9A-SR (9A-SR) trial in eastern Africa and 12 full-sib families (i.e., 8% selection intensity) to form three experimental OPVs, ‘TZ969A(FS)-SR’, ‘ET969A(FS)-SR’, and ‘UG969A(FS)-SR’, that optimized performance at each of the individual trial sites in Tanzania (TZ), Ethiopia (ET), and Uganda (UG), respectively (Table 1). The index was based on grain yield rank across sites, plus grain yield at the individual site for each of the three site-based experimental OPVs. In addition to increasing grain yield, selection was to avoid lengthening maturity (measured as days to flowering), decrease ear rot severity, lower ear height, decrease turcicum [caused by Exserohilum turcicum (Passerini) Leonard & Suggs, ET] leaf blight severity and lower percentage root lodging. Several iterations of the index were performed, varying the relative importance of the traits above, until good consensus emerged about which families were selected by the index.


View this table:
[in this window]
[in a new window]
  		Table 1. Pedigrees and description of the 14 experimental varieties formed from Pool 9A-SR.

 
A second experimental OPV was formed for each of the four cases (i.e., Across, Tanzania, Ethiopia, and Uganda) by recombining the most advanced inbred lines for each selected full-sib family. For each of the full-sib families selected to form the four full-sib-based OPVs described above, remnant seed of the most advanced surviving inbred generation (S1 to S3; Table 2) from selections in MSV nurseries at Harare was used to represent the selected full-sib family for recombination to form an inbred-based experimental OPV (i.e., ‘AC969A(Sn)-SR’, ‘TZ969A(Sn)-SR’, etc.; Table 1). The Syn1 generation of each experimental OPV was formed at Harare in summer 1997–1898, using 34 plants for each of the selected FS families, dividing the families into two groups, and making half-sib pollinations among the two groups on at least three dates. A balanced bulk, with equal representation for each FS family, was formed for each Syn1 after harvest. Because three FS-derived inbred lines failed to produce usable seed in the recombination nurseries, two FS families were not represented in AC969A(Sn)-SR and one FS family was not represented in UG969A(Sn)-SR (Table 2).


View this table:
[in this window]
[in a new window]
  		Table 2. Number of full-sib (FS) or inbred lines recombined to form 12 experimental maize varieties.

 
To allow more precise comparison of OPVs formed from selected full-sib families with OPVs formed from S1, S2, or S3 lines derived from selected full-sib families, we formed four additional experimental OPVs. For this purpose, the 14 best full-sib families (i.e., 10% selection intensity) were selected on the basis of an informal index (as described above) with the across-location data but considering only full-sib families for which a complete set of full-sib, S1, S2, and S3 was available (there were about 140 such complete sets). The resultant OPVs were named ‘AC969A-SR(FS of best Sn)’, ‘AC969A-SR(S1 of best Sn)’, ‘AC969A-SR(S2 of best Sn)’ and ‘AC969A-SR(S3 of best Sn)’ (Table 1). The Syn1 generation for each experimental OPV was formed at Harare in summer 1997–1998, using 34 plants for each FS, or 17–34 plants for each of one to four inbred lines derived from each of the 14 selected FS (Table 2), dividing the families into two groups, and making half-sib pollinations among the two groups on at least three dates. A balanced bulk, with equal representation for each FS family, was formed for each Syn1 after harvest; all 14 FS families were successfully and equally represented in each experimental OPV (Table 2).

Seed of the two previous cycles of improvement for PL9A-SR, PL9A-SRc2, and PL9A-SRc3 was produced during the same season (winter 1998) as the Syn2 generation was formed for the 12 new experimental OPVs. Seed was produced by hand pollinating (plant-to-plant, full-sib pollinations) more than 100 plants in a grow-out of breeder's seed for each of the 14 experimental OPVs. The yield trials evaluated Syn2 generation for the 12 new experimental OPVs and Syn4 generation for PL9A-SRc2 and PL9A-SRc3.

Evaluation of Progress from Selection
The 14 experimental varieties (Table 1) were evaluated for disease resistance and grain yield at 13 environments (Table 3) from 1998 to 2001. One of the 13 environments was a green house for which only MSV data were collected. A CIMMYT experimental double-cross hybrid (CML202/CML395//CML312/CML206) was included in the trial as a reference entry across all environments. The final entry was selected by the collaborating scientists and generally was the most popular variety at the site or region conducting the trial (different genotype at each environment). The experimental design for the 16 genotypes was an {alpha}-lattice (0,1) (Patterson et al., 1978) with three replications of two-row plots. Final plant density and crop management practices varied among sites, but all trials followed recommended practices for the respective sites.


View this table:
[in this window]
[in a new window]
  		Table 3. Locations where the 14 experimental varieties were evaluated during 1998 to 2001.

 
MSV data were collected only at Harare from two field environments (1998–1999 and 2000–2001 seasons) and one greenhouse environment (1999). Inoculation of plants with MSV was done by infestation with viruliferous leafhoppers 10 d after planting. Leafhoppers [Cicadulina mbila (Naude)] were anaesthetized with CO2 before dispensing them into plant whorls. MSV symptom severity was rated about 7 and 8 wk (two scores) after infestation in 1998–1999 season on a whole-plot basis by a 1 (no symptoms) to 5 (severe streaking, plants stunted and almost dead) scale with half-point increments as described by Welz et al. (1998). In the 2000–2001 season, plants were rated only at 8 wk after infestation.

For greenhouse evaluation of MSV resistance, 10 seeds were planted and seedlings were thinned to 8 per pot just before infestation with viruliferous leafhoppers. MSV infestation was done 10 d after planting with the pots placed in a cage covered with a net cloth to prevent insects from escaping. Viruliferous insects (anaesthetized with CO2) were then released into the cage by placing an average of three leafhoppers per plant whorl. The cage remained covered for a 7 d insect feeding and virus transmission period. Streak disease symptoms were scored on an individual plant basis 11, 18, and 25 d after infestation by the same scale as for the field experiments. MSV scores were averaged for all eight plants and the three scores (from three scoring dates) before analysis.

A whole-plot visual assessment using 1-to-5 scale (as for MSV, described above) was also used to score GLS and ET disease severity at all environments where either or both of the diseases occurred naturally during the trial seasons. Data on a plot basis were also collected for days to anthesis (number of days from planting to 50% pollen shed), plant height (distance from soil surface to the base of the flag leaf), ear height (distance from soil surface to the highest ear-bearing node), root lodging (% of plants leaning more than 30° from the vertical), stem lodging (% plants with broken stalks at or below the primary ear node), and grain yield in Mg ha–1 adjusted to 125 g kg–1 moisture content.

Statistical Analyses
Individual location analyses of variance were done by PROC MIXED of SAS (SAS, Inc., 2001). Across-environments analyses of variance for each trait used lattice-adjusted means with the local check deleted and only included data for locations where there was significant variance for the trait. In all analyses, variety effects and their subpartitions were considered fixed, while replication and environmental effects were considered random effects. In the across-environment analysis of variance, fixed effects and their subpartitions were tested for significance using their respective interactions with environments as the error terms while all interactions with environments were tested for significance using the pooled error. Progress per generation for three generations (S0 to S3) inbreeding with selection for the experimental variety AC969A-SR(FS of best Sn) was estimated by simple linear regression of each trait on inbreeding and selection cycles.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
There were significant differences (P < 0.01) for MSV, GLS, plant height, ear height, days to anthesis, and grain yield among the 14 experimental varieties across trial locations (Table 4). Differences for stem and root lodging, and ET were not significant (data not shown). Mean squares for the contrast FS vs. Sn indicated significant differences between FS- and inbred-derived experimental varieties for MSV and days to anthesis (P < 0.01), and for ear height (P < 0.05), but no significant difference for grain yield, GLS, or plant height (Table 4). Mean squares for linear regression indicated a significant (P < 0.05) decrease for MSV symptom score and number of days to pollen-shed, and a trend (P < 0.10) toward improved GLS disease resistance and lower ear placement for varieties formed from increasingly inbred families, confirming that selection during inbreeding was effective for these traits. Importantly, grain yield was not affected by selection for resistance to MSV and improvement of these agronomic traits during inbreeding.


View this table:
[in this window]
[in a new window]
  		Table 4. Analysis of variance for disease resistance scores and agronomic traits of 14 experimental maize varieties evaluated during 1998–2001.

 
Because farmers generally save seed of OPVs for subsequent planting, it is noteworthy that average variance for grain yield for FS varieties was slightly larger than for Sn varieties (standard deviation 1.69 and 1.52 Mg ha–1, respectively, data not shown); however, all of the new experimental OPVs had larger variance for grain yield than PL9A-SRc2 (1.33 Mg ha–1), which suggests that recycling seed is unlikely to lead to inbreeding depression for any of these experimental OPVs.

Regression of means on inbreeding-with-selection generations indicated a decreasing (favorable) trend (negative regression coefficient values; Table 4) for MSV symptom score and additional traits under mild selection. MSV disease symptom score and anthesis date decreased an average of 16 and 1%, whereas GLS symptom score and ear height showed decreasing trends (P < 0.10) of 8 and 2% per generation of inbreeding with selection, respectively (regression slope, b, expressed as a percentage of FS-derived variety; Tables 4 and 5). The desirable trends realized for all traits indicate that inbreeding with selection was effective for improving the traits of interest, without compromising grain yield (Table 5). We did not estimate the added cost of conducting the three generations of disease nurseries, but from Table 5 can conclude that the gains achieved for MSV resistance were similar (–0.71 vs. –0.75 MSV symptom score) to those achieved by one cycle (requiring 1 yr and two seasons of nurseries) of S1 recurrent selection for PL9A-SR (cycle 2 to cycle 3). On the other hand, because the financial and time costs for each cycle of selection involving multilocation yield trials are greater than the costs of one or more disease nursery evaluations, and because improvement of MSV resistance was of high priority in addition to improving grain yield, the "tandem" selection scheme described herein seems to have been appropriate.


View this table:
[in this window]
[in a new window]
  		Table 5. Means of disease resistance scores and agronomic traits for 14 experimental varieties formed from Pool 9A.

 
Although differences in genetic variance for MSV and GLS resistance in Pool 9A might explain the differences in selection gains for resistance to the two diseases, we expected to make more progress for MSV than GLS because we used artificial inoculation for MSV but relied on natural disease occurrence for GLS screening. The mean MSV score for varieties formed from inbred families was (2.50), significantly lower (P < 0.01) than that of the varieties formed from the full-sib families (2.99). The corresponding means for GLS were 1.62 and 1.86, respectively, which differed only at P < 0.10, but suggests selection response in the desired direction.

The interaction of environment with variety effects was significant for grain yield, anthesis date, and MSV symptom score (Table 4). However, environments generally did not affect the average differences between FS- and Sn–derived varieties and the linear changes in traits associated with generations of inbreeding with selection. This consistency across environments strengthens the conclusions outlined above.

Comparison of cycles of selection for PL9A-SR indicated that the newly formed cycle 4 [AC969A(FS)-SR] produced more grain yield and had better GLS resistance than cycles 2 and 3 (Table 5). The MSV disease symptom score for cycle 4 (AC969A(FS)-SR) was similar to cycle 2 and significantly larger than cycle 3, confirming that some MSV resistance was sacrificed by backcrossing PL9A-SRc3 to the MSV-susceptible Mexican Pool 9Ac7. Nevertheless, MSV symptom score of 3.02 is not likely to result in significant yield reduction and reflects a "worst case scenario" because plots were artificially infested in this study. Grain yield, MSV, and GLS scores of the double-cross hybrid check were similar to those for AC969A(FS)-SR, but the hybrid was significantly taller (230 cm) and later to flower (86.5 d) than this experimental OPV (data not shown). In general, although selection gains were realized for most traits when comparing cycle 4 formed by recombining best FS families [AC969A(FS)-SR] with PL9A-SRc3, better gains were achieved when inbreeding with selection was performed before recombination [e.g., AC969A-SR(S3 of best Sn)] (Table 5).

The improvement of MSV resistance achieved by inbreeding with selection, i.e., superiority of Sn vs. FS varieties and significant linear improvement of MSV resistance for S1–, S2– and S3–constituted varieties (Tables 4 and 5), is consistent with high heritability of MSV resistance as reported by Efron et al. (1989), Kim et al. (1989), Kyetere et al. (1999), and others. Furthermore, the fact that gains for MSV resistance were achieved without affecting selection gains for grain yield agrees with theory that predicts additive gains from tandem selection for uncorrelated traits (Hallauer and Miranda, 1981). The Pearson phenotypic correlation coefficient for MSV score with grain yield was not significant (r = 0.24, 12 df) for sites where MSV was not inoculated, whereas it was highly significant (r = –0.73, 12 df, P < 0.01) for sites where MSV was inoculated. Incidentally, experience indicates that the negative correlation between MSV score and grain yield at MSV-inoculated sites would be larger than –0.73 if a wider range of genotypes were evaluated, particularly if highly susceptible genotypes were included.

Jenkins (1935) found no difference among hybrids formed from lines at different levels of inbreeding and concluded that no change in combining ability for yield was expected because of inbreeding alone, mainly because selection during inbreeding was not sufficient to modify allele frequencies controlling grain yield. Sprague and Miller (1952) predicted that if selection practiced during inbreeding is effective in modifying gene frequency for combining ability, then yields of hybrids made after successive generations of inbreeding should exhibit a linear trend, but they reported no such response for grain yield. Sprague and Eberhart (1977) reviewed results from 50 yr of published studies and concluded that "selection practiced during inbreeding has less effect on combining ability for yield than was once thought;" however, selection "is highly effective in modifying lines with respect to general vigor, maturity, and insect or disease resistance." This conclusion is at the heart of common maize breeding practices (Bauman, 1981, as cited by Hallauer, 1990): Most breeders rely on visual selection (including for disease resistance, inbred vigor, and producibility), followed by testing combining ability for grain yield; 60% of breeders in USA delay first testcross evaluation of grain yield to S3 (33%) or S4 (27%) generations. Although Lamkey (personal communication, 2005) suggests that several private sector maize breeders in USA currently rely heavily on testing of very early generation lines (e.g., S1 or S2), we speculate that this strategy is only advisable for very elite, narrow-based populations for which little variation is expected for traits that might otherwise be readily improved during early generations of inbreeding.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Two sets of experimental OPVs lead to the same conclusion that FS selection combined with inbreeding with selection for MSV resistance (with additional mild selection for GLS resistance, reduced ear height and reduced days to flowering) resulted in superior varieties than FS selection alone. These results provide an example of increased overall breeding progress per year by conducting inbreeding-with-selection for highly heritable traits simultaneously with and while awaiting results from multilocation yield trial evaluations of experimental hybrids. Our decision to delay conclusion of the selection cycle (i.e., recombination of selected families to form new experimental OPVs) to conduct one, two, or three generations of inbreeding with selection for MSV resistance was driven by logistical constraints, namely, delays in return of yield data; however, our results indicate that this tandem selection scheme was effective, and similar strategies seem advisable for breeding programs where, in addition to grain yield, highly heritable traits other than and uncorrelated with grain yield are important.


   ACKNOWLEDGMENTS
 
We are grateful to the scientists, research staff and the organizations that supported them to conduct the trials and return valuable data, especially, but not only: D. Kyetere, J. Imanywoha and G. Bigirwa, NARO, Uganda; S. Twumasi-Afriyie, CIMMYT, and EARO staff at Ambo, Ethiopia; N. Lyimo and Z. Mduruma, Tanzania; G. Ombakho and KARI staff at Kitale, Kenya; B. Cowley, Pannar Ltd., Zimbabwe and Rep. South Africa. Technical assistance by the research staff at CIMMYT, Zimbabwe was invaluable: especially S. Mawere and S. Nsingo. Financial support of this work by the United Kingdom's Department for International Development (DFID) is gratefully acknowledged.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 





This Article
Right arrow Abstract Freely available
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 Google Scholar
Google Scholar
Right arrow Articles by Pixley, K. V.
Right arrow Articles by Tongoona, P.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Pixley, K. V.
Right arrow Articles by Tongoona, P.
Agricola
Right arrow Articles by Pixley, K. V.
Right arrow Articles by Tongoona, P.
Related Collections
Right arrow Maize
Right arrow Plant Disease
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