HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (11) Citing Articles via Google Scholar Google Scholar Articles by Bletsos, E. A. Articles by Goulas, C. K. Search for Related Content PubMed Articles by Bletsos, E. A. Articles by Goulas, C. K. Agricola Articles by Bletsos, E. A. Articles by Goulas, C. K.

CROP BREEDING, GENETICS & CYTOLOGY

Mass Selection for Improvement of Grain Yield and Protein in a Maize Population

^a National Agricultural Research Foundation (N.AG.R.E.F.), Cereal Institute of Thessaloniki, P.O. Box 312, 570 01 Thermi, Thessaloniki Greece
^b Univ. of Thessaly, School of Production Sciences, Faculty of Agriculture, Crop and Animal Production, Pedion Areos, 38334 Volos, Greece

chgoulas{at}uth.gr

	ABSTRACT

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 
Increasing grain protein concentration in maize (Zea mays L.) has not been a major focus of most breeding programs, which mainly focus on yield, maturity, and resistance to stress. The objective of this study was to evaluate the effectiveness of three cycles of mass selection to improve simultaneously yield and grain protein in the maize population GR-OP-319, derived from the F₂ generation of the single-cross hybrid PX-95. The two-step selection procedure involved a stratified mass selection system that included check plants for environmental control. In the first step, plants having grain yield higher than 80% of the check-plant mean and protein concentration greater than the check-plant mean were selected. In the second step, the four plants within each selection grid having the highest protein concentration were selected, which resulted in a final selection intensity of 5%. Three cycles of selection were completed during 1991 to 1995. Response to selection was evaluated during the selection process and by direct field evaluation of the C₀, C₁, C₂, and C₃ populations in 1995 and 1996. The average response during the selection was 5.1% cycle^-1 for yield, 0.8 g kg^-1 cycle^-1 for protein concentration, and 7.0% cycle^-1 for protein yield. No measurable differences among cycles were observed in the direct field evaluation even though the trends in the means of the cycles followed the trends shown during selection. After three cycles, the mass selection system studied did not appear to be particularly effective.

Abbreviations: _p, population mean • _s, mean of selected plants • S, selection differential • h²_r, realized heritability • R, response to selection

	INTRODUCTION

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 
ALTHOUGH MAIZE is mainly considered a carbohy drate source, it is also an important protein source because of its considerable total protein yield per hectare (Bjarnason and Vasal, 1992). Grain protein quantity in ordinary maize is low (80–110 g kg^-1) and of poor quality because it is low in the amino acids lysine and tryptophane. Grain protein quantity and quality have received relatively little attention from breeders although both traits can be manipulated by breeding and information on heritability has been accumulating for many years, particularly about protein quantity. This neglect is a consequence of focusing breeding objectives on attributes of immediate concern like yield, maturity, and resistance to stresses (Alexander, 1988). Duvick (1997) reported that while the yielding ability of the maize hybrids, adapted to Central Iowa, has improved at a linear rate of about 74 kg ha^-1yr^-1 during the period 1930 to 1990, protein concentration in grain decreased at a linear rate of 30 g kg^-1 for each decade.

Breeding projects specifically targeted to protein have been successful. CIMMYT's breeding effort in developing quality protein maize (QPM) succeeded in producing high quality germplasm adapted to tropical and subtropical countries (Bjarnason and Vasal, 1992; Vasal et al., 1993a, b). The classical selection experiment for high and low protein maize conducted at the University of Illinois has provided valuable information on the possibilities and limitations of recurrent selection for this quantitative trait (Dudley et al., 1974; Dudley and Lambert, 1992). After 90 generations of selection, the Illinois High Protein (IHP) strain reached 320 g kg^-1 protein and 173 genes were estimated affecting the trait indicating that additional progress should be possible (Dudley and Lambert, 1992) . Dudley et al. (1977) reported a negative correlation between grain yield and protein concentration and suggested that low protein was dominant to high protein concentration. On the basis of their data, they further proposed selection for intermediate levels of protein concentration and higher grain yield as an appropriate breeding strategy when the target is increased protein yield per hectare. Boyat et al. (1980) crossed the Illinois high protein strains with French germplasm and following pedigree selection developed high protein inbred lines having 20 to 90 g kg^-1 protein higher than checks. These lines when tested in hybrid combinations had low grain yield, confirming the negative association between protein and yield. In contrast, Kauffmann and Dudley (1979) and Pollmer et al. (1978a), using germplasm with protein quantity levels more nearly representative of standard maize, reported low or insignificant genetic and phenotypic correlations between the two traits (Pollmer et al., 1978a). These data suggest that simultaneous improvement of both traits should be feasible and the high negative correlation could pertain only to the particular Illinois high protein material. The quantitative nature of protein concentration in the grain, along with the evidence that additive genetic variance is of greater importance than non-additive variance, indicate that recurrent selection should be an efficient way to improve protein quantity and combining ability for yield (Pollmer et al., 1978b; Boyat et al., 1980; Kaan et al., 1980). Simultaneous selection for grain yield and protein concentration by means of half-sib family selection based on desired gain indices in two maize populations over two selection cycles was effective in improving both traits, whereas mass selection was as effective as half-sib family selection for increasing protein concentration (Kauffmann and Dudley, 1979). Six cycles of recurrent selection by means of half-sib family index selection for grain yield and protein concentration or stratified mass selection for grain yield accompanied by within block selection for protein concentration resulted in significant gains in both protein and protein yield per hectare with no change in yield (Alexander, 1988). Pollmer et al. (1980) reported that hybrids combining high grain yield with improved protein concentration and performance stability could be developed as was shown by the performance of hybrid combinations between high x high or high x low protein lines (Pollmer et al., 1978a; Boyat et al., 1980).

New elite populations with both high protein concentration and good combining ability for yield is needed and the appropriate germplasm to develop them exists (Dudley et al., 1996). Such improved populations could be used as sources for new inbred line development and/or improving elite hybrids. Simultaneous improvement of grain yield and protein concentration is feasible and a systematic breeding effort should lead to important new sources of productive and valuable germplasm.

The objective of this research work was to determine whether mass selection for the simultaneous improvement of grain yield and protein concentration in a maize composite population was effective.

	Materials and methods

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 
Population GR-OP-319 was the source material (C₀) used in this study. This population originated from the F₂ of the single cross PX-95, a proprietary hybrid of Northrup King, followed by one cycle of mass selection for yield. The hybrid PX-95 had high yield potential and above average protein under our conditions (unpublished data).

During 1991, the C₀ population was planted in isolation at the N. Zoe, Imathia, experimental farm. A stratified grid selection arrangement for mass selection was used. Planting was in 20.0-m rows spaced 0.80 m apart. Rows consisted of 40 hills spaced 0.5 m apart with one plant per hill at a final density of 25 000 plants ha^-1. Hills were over planted and thinned to one plant after stand establishment. A total of 24 rows was planted along with seven rows of PX-95, alternating every four C₀ rows, to be used as checks. The 960 C₀ population hills were arranged in 12 selection blocks. Each selection block had 80 hills in four 10.0-m-long rows bordered by 20 check plant hills. All check plants were detasseled prior to flowering to secure pollen control.

Planting time was mid-April and harvest was in early-October. Standard cultural practices were followed throughout the growing season. Stalk or root-lodged plants were discarded at harvesting time. Ears of all C₀ and hybrid check plants having competing neighboring plants were hand harvested. Ears were kept in the greenhouse for 40 d to dry and ears were then shelled for plant grain yield and protein concentration determination. Protein concentration was determined as Kjeldahl N x 6.25. A balanced seed composite from all harvested plants was formed to represent the C₀ population. To identify genotypes having both high grain yield and protein concentration, a two-step selection procedure was followed on a within-block basis. Grain yield 80% of the check average, combined with protein concentration above check average, was the culling level in the first step. A total of 560 plants were thereby identified (58% initial selection intensity). Then, within each block, the four plants having the highest protein concentration were finally selected resulting in a 5% selection intensity (48 plants out of 960). A balanced composite from the 48 selected plants was formed and used to initiate the next (C₁) selection cycle. The mean grain yield, protein concentration and protein yield of all plants evaluated (excluding checks) provided an estimate of the C₀ population performance. The mean grain yield, protein concentration, and protein yield of all 960 plants and the 48 selected plants was used to calculate the selection differentials . Selection in cycles C₁, C₂, and C₃ was carried out in the same manner during the 1992, 1993, and 1995 growing seasons. The only difference was that the number of plants exceeding the culling level in the first selection step corresponded to selection intensities of 59, 64, and 44% in cycles C₁, C₂, and C₃.

Population and selected plants means along with the selection differentials (S) in each cycle were adjusted for the selection year effect by expressing them as percentages of the appropriate means from the check plants. The relative values were converted to grams per plant or protein concentration on the basis of the appropriate average of the checks over all selection years. This allowed comparisons providing indirect estimates of response to selection (R), and the ratio R/S, which corresponds to realized heritability .

Response to selection was measured directly in field evaluation of cycles C₀, C₁, and C₂ during 1995 and 1996 in three locations each year in isolated fields. Population C₃ was evaluated only in 1996. Experiments were conducted at N. Zoe (lat. 40. 38°N, long. 22. 27°E), Palamas (lat. 39. 32°N, long. 21. 45°E), and Ioannina (lat. 39. 07°N, long. 20. 30°E). A randomized complete block experimental design with four replications was used. Planting was in two-row plots 4.8 m long, spaced 0.80 m apart with each row having 25 hills at a final planting density of 65 000 plants ha^-1. Hills were over planted and thinned to a single plant. An extra row with C₀ plants was interplanted every two plots to serve as common pollinator for the entries tested. Plants of all entries were detasseled prior to flowering to avoid any possible confounding of pollen source on grain protein concentration. Planting time was mid-April and harvesting early-October. Standard cultural practices were followed throughout growing season. Plots were machine harvested, and grain yield was adjusted to 155 g kg^-1 moisture. Protein concentration was determined on two samples per plot (Kjeldahl N x 6.25). Plant and ear height and days to silking were recorded on the four replications in each experiment. Plant and ear height was recorded as the average of measurements on five competitive plants per plot and measured as the distance from the soil surface to the base of tassel and to the highest earbearing node, respectively.

Randomized complete block design statistical analysis of each field experiment was made followed by a combined analysis. In the combined analysis, locations and years were equated to six (1995 and 1996 data) or three (1996 data only) environments. Environments were considered random effects while populations were considered fixed effects. Correlation coefficients were calculated during each cycle of selection for grain yield and protein concentration on a single plant basis . The average response to selection in the indirect evaluation was calculated as (C₃ - C₀)/3.

	Results and discussion

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 
The average hybrid check plant performance indicated a year effect on grain yield, protein concentration, and protein yield (Table 1) . Thus, the adjustment of the mean of the population and the mean of selected plants was justified in order to examine selection response. The grain yield of the C₀ population was 45% lower than the check hybrid (Table 2) . The mean protein concentration of C₀, 105 g kg^-1, was within the 89 to 147 g kg^-1 range reported for 175 European maize populations by Kaan et al. (1980), the 79 to 118 g kg^-1 reported for 36 hybrids from a diallel involving high protein inbreds by Motto et al. (1980), and the 92 to 110 g kg^-1 reported for hybrid combinations among 20 populations and three inbred lines by Dudley et al. (1996).

View this table: [in this window] [in a new window]   Table 2 Grain yield, protein concentration, and protein yield (± standard deviation) of maize populations and selected plants along with estimates of selection differentials, responses to selection, and realized heritability estimates in each of three cycles of mass selection in a maize composite population

Field evaluation of the C₀ base population and the corresponding C₁, C₂, and C₃, over a range of environments, provided a direct estimate of response to selection. As expected environmental effects were significant for protein yield per hectare and its components, grain yield and protein concentration. The cycles of selection were not significantly different for any trait. The cycles x environment interaction effect was not significant either. Thus, on the basis of data from the replicated field trials, no measurable improvement in grain yield, protein concentration, or protein yield was observed. The means of the cycles of selection followed the trends shown during the selection process (Table 3) but the field evaluation was probably too limited to discern the small differences among cycles.

The results are inconclusive about whether it is possible to simultaneously improve grain yield and protein concentration. Selecting for high protein concentration and intermediate to high grain yield may be an effective breeding strategy in improving protein yield per hectare, but our results showed that three cycles of mass selection were not particularly effective in improving any trait. Additional cycles of selection are needed to determine if selection would eventually prove more effective.

	ACKNOWLEDGMENTS

 
We express our appreciation to the anonymous reviewers for the comments and suggestions.

Received for publication July 16, 1998.

	REFERENCES

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 

• Alexander D.E. Breeding special nutritional and industrial types. In: Sprague G.F., Dudley J.W., eds. Corn and corn improvement. Madison, WI: Agron. Monogr. 18 ASA, CSSA, and SSSA, 1988:873-874 Protein quantity and quality..
• Bjarnason M., Vasal S.K. Breeding of quality protein maize (QPM). Plant Breed. Rev. 1992;9:181-216.
• Boyat A., Derieux M., Kaan F., Raton S. Maize breeding for improvement of kernel protein content using Illinois high protein population. In: Pollmer W.G., Phipps R.H., eds. Improvement of quality traits of maize for grain and silage use. Netherlands: Martinus Nijhoff Publ, 1980:173-183.
• Dudley J.W., Lambert R.J., Alexander D.E. Seventy generations of selection for oil and protein concentration in the maize kernel. In: Dudley J.W., ed. Seventy generations of selection for oil and protein in maize. Madison, WI: CSSA, 1974:181-211.
• Dudley J.W., Lambert R.G., De La Roche I.A. Genetic analysis of crosses among corn strains divergently selected for percent oil and protein. Crop Sci. 1977;17:111-117.
• Dudley J.W., Lambert R.G. Ninety generations of selection for oil and protein in maize. Maydica 1992;37:81-87.
• Dudley J.W., Lamkey K.R., Geadelmann J.L. Evaluation of populations for their potential to improve three maize hybrids. Crop Sci. 1996;36:1553-1559.[Abstract/Free Full Text]
• Duvick, D.N. (1997). What is yield? p. 332–335. In G.O. Edmeades et al. (ed.) Developing drought and low N tolerant maize. Proc. of a Symposium, El Batan, Mexico. 25–29 Mar. 1996. CIMMYT, El Batan, Mexico.
• Kaan F., Boyat A., Mavie R., Panonille A. Broadening the genetic bases of maize for grain protein content by use of European local populations and mutagenesis. In: Pollmer W.G., Phipps R.H., eds. Improvement of quality traits of maize for grain and silage use. Netherlands: Martinus Nijhoff Publ, 1980:185-197.
• Kauffmann K.D., Dudley J.W. Selection indices for corn grain yield, percent protein and kernel weight. Crop Sci. 1979;19:583-588.[Abstract/Free Full Text]
• Motto M., Difonzo N., Berenzin M., Gentinetta E., Maggiore T., Salamini F., Lorenzoni G. Genetic control of protein percentage in a diallel involving high protein inbreds. In: Pollmer W.G., Phipps R.H., eds. Improvement of quality traits of maize for grain and silage use. Netherlands: Martinus Nijhoff Publ, 1980:117-130.
• Pollmer W.G., Eberhart D., Klein D., Dhillon B.S. Studies on maize hybrids involving inbred lines with varying protein content. Z. Pflanzenzuchtg 1978;80:142-148 a.
• Pollmer W.G., Eberhart D., Klein D. Inheritance of protein and yield of grain and stover in maize. Crop Sci. 1978;18:757-759 b.[Abstract/Free Full Text]
• Pollmer W.G., Eberhart D., Klein D., Dhilon B.S. Stability performance of high protein maize (Zea mays L.) hybrids. In: Pollmer W.G., Phipps R.H., eds. Improvement of quality traits of maize for grain and silage use. Netherlands: Martinus Nijhoff Publ, 1980:257-272.
• Vasal S.K., Srinivasan G., Pandey S., Gonzales F., Crossa J., Beck D.L. Heterosis and combining ability of CIMMYT's quality protein maize germplasm. I. Lowland tropical. Crop Sci. 1993;33:46-51 a.[Abstract/Free Full Text]
• Vasal S.K., Srinivasan G., Gonzales F., Beck D.L., Crossa J. Heterosis and combining ability of CIMMYT's quality protein maize germplasm. II. Subtropical. Crop Sci. 1993;33:51-57 b.

This article has been cited by other articles:

				F. J. Kutka and M. E. Smith How Many Parents Give the Highest Yield in Predicted Synthetic and Composite Populations of Maize? Crop Sci., September 1, 2007; 47(5): 1905 - 1913. [Abstract] [Full Text] [PDF]

				B. Feil, S. B. Moser, S. Jampatong, and P. Stamp Mineral Composition of the Grains of Tropical Maize Varieties as Affected by Pre-Anthesis Drought and Rate of Nitrogen Fertilization Crop Sci., January 31, 2005; 45(2): 516 - 523. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (11) Citing Articles via Google Scholar Google Scholar Articles by Bletsos, E. A. Articles by Goulas, C. K. Search for Related Content PubMed Articles by Bletsos, E. A. Articles by Goulas, C. K. Agricola Articles by Bletsos, E. A. Articles by Goulas, C. K.