Content uploaded by Paul Scott
Author content
All content in this area was uploaded by Paul Scott
Content may be subject to copyright.
ABSTRACT - In order to determine how modern hybrids
have impacted grain composition and amino acid content
of the corn crop, we characterized a set of cultivars that
were widely grown in different eras from the 1920s
through 2001. Grain composition exhibited clear trends
with time, with protein decreasing and starch increasing.
The effects of different plant densities were examined.
The grain protein content of modern hybrids responds to
plant density and environment differently than the protein
content of older varieties. These differences are consistent
with a model in which protein content is modulated by
different growth conditions. These differences may ex-
plain, in part, the mechanism by which modern hybrids
maintain yield in different environments, i.e. reduction of
protein content in stressful environments frees resources
that are used to maintain yield. We examined the content
of the nutritionally limiting essential amino acids lysine,
methionine and tryptophan in grain of these cultivars. On
a per tissue mass basis, the levels of these amino acids
dropped with time while on a per protein basis, their lev-
els were not significantly changed. We conclude that the
development of modern hybrids has resulted in maize
with reduced protein content, but the nutritional quality
of this protein has not changed.
KEY WORDS: Protein quality; Plant densities; Nutritional
quality; Essential amino acids.
INTRODUCTION
Maize production in the U.S. has increased at a
rate of 1-2% per year on a per acre basis since
about 1930 (TRACY et al., 2004). In order to deter-
mine the reasons for this gain, experiments have
been conducted using public and private (RUSSELL,
1974; DUVICK, 1977) cultivars that were widely
grown in different decades. By evaluating these “era
cultivars” in common environments, it has been es-
timated that greater than 70% of the yield gain is
due to genetic improvement (DUVICK,1984; RUSSELL,
1984). These studies suggest a mechanism for the
observed increase in production through breeding.
By evaluating the era cultivars in low plant densities
commonly used in the 1930s and in the high plant
densities used today, it became clear that part of the
gain in productivity is due to adaptation to higher
plant densities. At low plant densities, grain yield is
not significantly different in modern hybrids than in
older ones, while at higher plant densities, modern
hybrids yield considerably more than older cultivars
(DUVICK, 1977).
The set of largely private germplasm used in
these studies consists of two widely grown open
pollinated cultivars and a series of hybrids released
by Pioneer Hi-bred International that were widely
grown between 1930 and 2001. This set of cultivars
is called the era hybrids and has been characterized
with respect to many factors involved in determin-
ing agronomic performance (DUVICK et al., 2004).
While changes in agronomic performance in the
course of developing modern hybrids have been
extensively studied, less attention has been given to
changes in grain quality during this process. The
majority of the effort spent in developing maize cul-
tivars has been devoted to improving the agronomic
characters leading to grain yield. Modifications to
grain quality are essentially unintended effects;
however, it is desirable to know the magnitude and
direction of these effects as they are likely to be
magnified in future breeding efforts. Nutritional
quality is particularly important because most of the
maize grain that is produced is used for food or
feed. It has been reported that grain protein de-
creased on average 0.3% per 10 years, while grain
starch increased on average 0.3% per 10 years and
Maydica 51 (2006): 417-423
GRAIN COMPOSITION AND AMINO ACID CONTENT IN MAIZE CULTIVARS
REPRESENTING 80 YEARS OF COMMERCIAL MAIZE VARIETIES
M.P. Scott1,*, J.W. Edwards1, C.P. Bell2, J.R. Schussler3, J.S. Smith3
1
USDA-ARS, Corn Insects and Crop Genetics Research Unit, Ames, Iowa 50011, USA
2
Interdepartmental Genetics Program, Iowa State University, Ames, Iowa, 50011, USA
3
Pioneer Hi-Bred International, Johnston, Iowa, 50131, USA
Received February 22, 2006
* For correspondence (e.mail: pscott@iastate.edu).
grain oil did not change significantly in the era hy-
brids (DUVICK et al., 2004). Among the main nutri-
tional limitations of maize grain are contents of the
essential amino acids lysine, methionine and trypto-
phan. While it has been reported that levels of
these amino acids are correlated with protein con-
tent (MILLER et al., 1950), the correlation of trypto-
phan with non-zein protein is higher that with total
protein (FREY et al., 1949). Thus, increasing total
protein may lead to increasing levels of essential
amino acids but decreased quality of protein as de-
fined as levels of essential amino acids on a per to-
tal protein basis. Consistent with this are several
analyses of grain from the Illinois long-term protein
selection experiment that conclude high protein
populations have a greater proportion of zeins than
low-protein populations (reviewed in (BELOW et al.,
2004). It is therefore interesting to examine the pro-
tein quality of the era hybrids to determine if the re-
ported decrease in protein content over the devel-
opment of modern hybrids has been accompanied
by a change in protein quality.
The objectives of this study were to characterize
differences in cultivars that were grown widely in
different eras with respect to grain composition. We
expanded on the earlier study (DUVICK et al., 2004)
by examining the constituents protein, oil, and
starch in different plant densities. We also report
levels of nutritionally limiting amino acids trypto-
phan, lysine, and methionine to identify changes
that have occurred to the amino acid balance of
grain.
MATERIALS AND METHODS
Experimental design
The 45 cultivars listed in Table 1 were produced in 2003 at
two locations (Woodland, CA and Johnston, IA). At each loca-
tion, each hybrid was produced at two planting densities: 90,000
plants/Ha (high density) and 44,500 plants/Ha (low density).
Each plot consisted of 2 rows which were allowed to open polli-
nate. The Iowa location was produced with normal rainfall. The
California location receives essentially no rainfall during the
growing season and is irrigated. At this location, drought stress
treatments were imposed by withholding irrigation prior to and
during flowering or during the grain fill period. The three envi-
ronments were managed as separate experiments, due to the
need to block drip irrigation plumbing. Thus, the two locations
provide three separate environments in which the experiment
was conducted. Within each experiment, 2 replicates of the den-
sity main plots, each containing all genotype entries as split
plots, were established. Density main plots were randomized
within each experiment and hybrid split plots were independent-
ly randomized within each replicate of the density main plots.
Grain analysis
Plots in Iowa were harvested by hand and grain from several
ears from each plot was bulked and used for analysis. Bulk grain
representing each plot in California was harvested with a com-
bine equipped for grain sampling. Samples were dried to about
10% moisture. Each sample was analyzed with NIR to predict
protein, oil and starch content using a Perstorp 6500 NIR spec-
trometer (Foss North America, Eden Prairie, Minnesota) equipped
with a sample transport module and natural product cell. Sam-
ples are reported on a zero percent moisture basis.
Amino acid content was determined on a 40 kernel sub-sam-
ple using a microbial assay (SCOTT et al., 2004) to measure the
content of tryptophan, methionine and lysine. Assays were car-
ried out in 96-well plates, with each rep randomized as a block
and analyzed on three plates. Thus, six 96-well plates were used
for the experiment.
Data analysis
The amino acid data were analyzed using a mixed linear
model as follows:
yijklmn =
l
i+
r
ij +
g
jk +
a
jkl +
b
jkm +
d
n+
ld
in +
yl
i+
yd
n
+
yld
in +
h
p+
hd
np +
hl
ip +
e
ijklmn
yijklmn =response
l
ieffect environment i
r
ij =effect of replicate jwithin environment i
g
jk =effect of plate kwithin replicate j
a
jkl =effect of row lon plate jk
b
jkm =effect of column mon plate jk
d
n=effect of density n
ld
in =environment by density interaction
y
=year of hybrid release (continuous)
yl
i=year by environment interaction
yd
n=year by density interaction
yld
in =year by environment by density interaction
h
p=effect of hybrid p
hd
np =hybrid by density interaction
hl
ip =hybrid by environment interaction
e
ijklmn =residual
Plates, rows, and columns are laboratory variables for the 96-
well plates used in the microbial amino acid analyses only. Rows
and columns represented the rows and columns of individual
cells on the plates. The environment variable was for the two en-
vironments in California (two different water treatments) and the
Iowa environment.
Year was fit as a continuous variable (covariate) along with
year by environment, year by density, and year by environment
by density interactions. The analysis of protein, oil, and starch
data was done with by NIR, and did not involve 96-well plates,
thus the model for those traits did not have plate, row within
plate, or column within plate. When samples were assigned to
plates for amino acid analysis, all the entries in replicates labeled
“one” in all three environments were randomly ordered and as-
signed to three plates. Likewise, all samples assigned to repli-
cates labeled “two” across the three environments were random-
ly assigned to a second set of three plates. Therefore, in the nest-
ing of effects in the model, replicates are nested within environ-
ments due to the field design, but due to the lab design, plates
are nested within the replicate variable but not within environ-
418 M.P. SCOTT, J.W. EDWARDS, C.P. BELL, J.R. SCHUSSLER, J.S. SMITH
ments (plate effects were specific to the laboratory). All effects in
the model were considered fixed except hybrids, hybrid by envi-
ronment interaction, and hybrid by density interaction. Variances
were estimated by restricted maximum likelihood (SAS proc
mixed). Hybrid by environment by density interaction was not
significant, and was dropped for all traits. Hybrid by environ-
ment and hybrid by density interaction was not significant for
amino acid traits and was dropped for those traits. All main ef-
fects and interactions not involving the covariate effect of year
were tested for significance with type-I F-tests. Effect of years
and interactions of year with densities and environments were
tested with type III F-tests. Degrees of freedom were computed
according to Satterthwaite approximations as implemented in
SAS proc mixed. Generalized least squares estimators of regres-
sion coefficients and their standard errors for the change in trait
value per year and their standard errors were presented for ap-
propriate effects based on significance of the type III hypothesis
tests. The magnitudes of the residual terms in the models were
between 35 and 71% of the magnitude of the total variance for
each trait. These F-test results are summarized in Table 2.
RESULTS AND DISCUSSION
A set of 45 cultivars that were each grown wide-
ly at some time between 1920 and 2001 were used
in this study (Table 1). The majority of these culti-
vars are commercial hybrids released by Pioneer Hi-
bred International. Each cultivar was grown in three
different environments differing in their water avail-
ability, and in two different plant densities. Within
each water experiment, a split plot treatment design
GRAIN COMPOSITION IN MAIZE VARIETIES 419
TABLE 1 - Hybrids used in this study and a year that they were
widely grown.
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
Cultivar Year Cultivar Year
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
Krug YD 1920 3571 1968
Reid YD 1920 3334 1969
351 1934 3388 1970
307 1936 3517 1971
322 1936 3366 1972
317 1937 3301A 1974
330 1939 3541 1975
336 1940 3382 1976
340 1941 3377 1982
339 1942 3378 1983
344 1945 3475 1984
352 1946 3379 1988
350B 1948 3417 1990
347 1950 3394 1991
301B 1952 3489 1994
354 1953 3335 1995
329 1954 33G26 1998
354A 1958 33P67 1999
3618 1961 34B23 1999
3206 1962 34M95 2001
3306 1963 34N44 2002
3376 1965 34H31 2002
3390 1967
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
TABLE 2 - Hypothesis tests.
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
Composition (%) Amino acids (g/100g tissue) Amino acids/Protein
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
Effect1Oil Protein Starch Lysine Methionine Tryptophan Lysine Methionine Tryptophan
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
Rep n.s.2n.s. * n.s. ** ** n.s. ** **
Plate(Rep) ** ** ** ** ** **
Row(Plate*Rep) ** ** ** ** ** **
Colm(Plate*Rep) ** ** ** ** ** **
Environment n.s. ** ** ** ** ** ** ** **
Density n.s. ** ** ** ** n.s. ** n.s. **
Environment*Density n.s. ** n.s. n.s. * n.s. * * n.s.
Year ** ** ** ** * ** n.s. n.s. n.s.
Year*Environment n.s. * n.s. n.s. n.s. n.s. n.s. n.s. n.s.
Year*Density n.s. * n.s. n.s. n.s. n.s. n.s. n.s. n.s.
Year*Environment*Density * n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
1Row, Colm and Plate are laboratory variables used for the measurement of amino acid levels and not in the analyses of protein, oil or
starch.
2** P<0.01; * P<0.05.
was used with density as the main plot and entry as
the split plot. Two replicates of each treatment and
cultivar were grown in each experiment. Each water
treatment was established as a separate experiment.
The combination of imposed drought stress and dif-
ferent plant densities resulted in production envi-
ronments that were quite different. Because drought
stress has been shown to result in decreased yields
with increased protein content (LILBURN et al., 1991),
we anticipated that these environments should pro-
vide an opportunity to observe environmental ef-
fects on the traits examined. Data were collected on
protein, oil and starch content as well as the con-
tent of the nutritionally limiting amino acids methio-
nine, tryptophan and lysine. Amino acid levels were
expressed both on the basis of tissue mass and total
protein content to provide information about the to-
tal amino acid content and the protein quality, re-
spectively.
Effect of environment on grain composition
Environment had a significant effect on all traits
except oil (Table 2). The magnitudes of these
changes were up to .03% of the tissue mass. A
study of commercial hybrids revealed that total pro-
tein content and total sulfur amino acids were ele-
vated with a concomitant decrease in grain yield
when these hybrids were produced in drought con-
ditions (LILBURN et al., 1991). This result is consistent
with our study. The well-watered environment had
lower mean protein values than the two drought
stress environments (Fig. 2B). The significant envi-
ronmental effects observed in our study may be a
manifestation of the extreme differences in produc-
tion environments imposed during plant growth
and the fact that two of our environments involved
drought stress.
Plant density had a significant effect on all traits
except oil, methionine/per protein, and tryptophan
420 M.P. SCOTT, J.W. EDWARDS, C.P. BELL, J.R. SCHUSSLER, J.S. SMITH
0.4
0.3
0.2
0.1
0.0
200019601920
14.0
13.5
13.0
12.5
12.0
20001980196019401920
5.0
4.5
4.0
3.5
3.0
200019601920
71.0
70.5
70.0
69.5
69.0
20001980196019401920
A
B
C
D
Year of Hybrid
Percentage
FIGURE 1 - Grain composition of the era hybrids produced averaged across environments and plant densities. Lines are fit to data plotted
as percentage of the tissue mass of each component. A. Starch; B. Protein; C. Oil; D. Amino acids; dotted line, lysine; solid line, trypto-
phan; dashed line, methionine. Slopes and standard errors for these data are presented in Table 3.L.
TABLE 3 - Slopes of regression of trait values on years.
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
Trait Slope ± Standard Error
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
Oil -0.006 ± 0.002
Protein -0.014 ± 0.003
Starch 0.015 ± 0.003
Lysine -0.00042 ± 0.00013
Methionine -0.00012 ± 0.00005
Tryptophan -0.00027 ± 0.00006
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
(Table 2). Protein content averaged 0.6% higher and
starch content averaged 0.3% lower in low plant
densities. The effect of plant density on amino acid
content was mixed, with methionine being higher
and lysine being lower at low plant density. Protein
quality was decreased at low plant density in the
two significant cases lysine/protein and tryptophan/
protein. Thus, while low plant density allows accu-
mulation of more protein, lysine and tryptophan are
underrepresented in this protein, resulting in lower
quality.
Changes in grain composition with year
of hybrid release
The year that the hybrids were widely grown
had a significant effect on protein, oil and starch
content (Table 2). Protein decreased and oil de-
creased with time, and starch increased with time
(Table 3, Fig. 1). This is similar to a previous report
(DUVICK et al., 2004), in which protein decreased
with hybrid year while starch went up. Our values
for protein and starch are slightly lower than the
values of 0.03% per year previously reported. In
contrast to our results, the previous study reported
no change in oil content.
Lysine, tryptophan, and methionine all had sig-
nificant year effects when expressed as acid content
per tissue mass, while the same amino acids when
expressed on a per protein basis did not (Table 1).
The results for amino acid per tissue mass are illus-
trated in Fig. 1 and Table 3. Like total protein con-
tent, the content of amino acid per tissue mass de-
GRAIN COMPOSITION IN MAIZE VARIETIES 421
TABLE 4 - Slopes ± standard errors of regression of protein content on year of hybrid.
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
Averaged across environments Averaged across plant densities
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
Density High -0.016 ± 0.0030
Low -0.013 ± 0.0030
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
Environments IA -0.012 ± 0.0034
CA Stress at grain fill -0.019 ± 0.0034
CA Stress at flowering -0.012 ± 0.0034
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
TABLE 5 - Slopes ± standard errors of regressions of oil content on year for different tr eatments.
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
Environments
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
IA CA stress at fill CA stress at flowering
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
High Density -0.0079 ± 0.0024 -0.0070 ± 0.0024 -0.0062 ± 0.0023
Low Density -0.0014 ± 0.0024 -0.0044 ± 0.0024 -0.0093 ± 0.0023
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
13.0
12.0
11.0
20001980196019401920
13.0
12.0
11.0
20001980196019401920
Percentage
Year of Hybrid
A
B
FIGURE 2 - Change in protein con-
tent with time. A. Dashed line, low
plant density; solid line, high plant
density. B. Dashed line, drought
stress at flowering; dotted line, well
watered; Solid line, drought stress at
grain fill. Slopes and standard errors
for these data are presented in
Table 4.
creased with year of hybrid release. The fact that
amino acid per protein content did not have a sig-
nificant release year effect (in contrast to amino
acid per tissue mass) suggests that composition of
hybrids has changed over time, while the quality of
the protein (defined as methionine, lysine or trypto-
phan per protein) has not changed in a statistically
detectable way.
The test of significance of the Year*Density effect
can be used to identify significant differences be-
tween high and low plant density in the change in
trait values with time. These effects were only sig-
nificant for protein content (Table 2). The signifi-
cant Year*Density effect for protein content indi-
cates that the protein content of modern hybrids is
reduced more in high plant density than the protein
content of older hybrids (Table 4, Fig. 2). It has
been reported that the yield of modern hybrids is
reduced less by increased plant density (DUVICK et
al., 2004). This greater reduction in protein content
in modern hybrids may be a compensating mecha-
nism that allows these varieties to maintain their
yield in high density conditions. Reduced protein
content may free resources that can be used by the
plant to maintain yield.
The test of significance of the Year*Environment
effect can be used to identify significant differences
between environments in the change in trait values
with time. Like Year*Density, the Year*Environment
effect was only significant for protein content (Table
2, 4, Fig. 2). This indicates that the protein content
of modern hybrids responds differently to different
environments than that of older hybrids.
The test of Environment*Density can be used to
identify significant differences in response across all
release years for the interaction between the envi-
ronment and density treatments. This effect was on-
ly significant for oil content (Table 1). The effect of
plant density on oil content varied with environ-
ment. The mean oil content was higher in high
plant densities. The Year*Environment*Density ef-
fect can be used to determine if the slopes of trait
vs. time regression lines are effected differently in
different environment-density combinations. This ef-
fect was only significant for oil content. Some envi-
ronments had larger plant density effects than oth-
ers (Table 5, Fig. 3). This can be seen by comparing
the difference in slopes of the two lines in the well
watered case to the difference in slopes of the lines
in the stress cases.
While clear trends are observable in grain com-
position in over the course of development of the
era hybrids, the magnitude of the changes are small
and on the order of magnitude of changes attrib-
uted to environmental effects. At plant densities
used today and with low environmental stresses,
these changes have resulted in grain with higher
starch content and lower content of protein with
unchanged quality. These changes may be favor-
able for some end uses and unfavorable for others.
ACKNOWLEDGEMENTS - This manuscript is dedicated to Don-
ald Duvick whose experiments with the Era Hybrids and grain
quality inspired this work. We thank Merinda Struthers for techni-
cal assistances. Names are necessary to report factually on the
available data; however, the USDA neither guarantees nor war-
rants the standard of the product, and the use of the name by
the USDA implies no approval of the product to the exclusion of
others that may be suitable.
422 M.P. SCOTT, J.W. EDWARDS, C.P. BELL, J.R. SCHUSSLER, J.S. SMITH
4.0
3.8
3.6
3.4
3.2
200019601920
4.0
3.8
3.6
3.4
3.2
200019601920
4.0
3.8
3.6
3.4
3.2
200019601920
Percentage
Year of Hybrid
A
B
C
FIGURE 3 - Change in Oil content of the era hybrids over time. Dashed line, low plant density, Solid line, high plant density. A. Drought
stress at grain fill. B. Drought stress at flowering. C. Well watered. Slopes and standard errors for these data are presented in Table 5.
REFERENCES
BELOW F.E., J.R. SEEBAUER, M. URIBELARREA, M.C. SCHNEERMAN, S.P.
MOOSE, 2004 Physiological changes accompanying long-term
selection for grain protein in maize. Plant Breed. Rev. 24: 133-
151.
DUVICK D.N., 1977 Genetic rates of gain in hybrid maize yields
during the past 40 years. Maydica 22: 187-196.
DUVICK D.N., 1984 Genetic contributions to yield gains of U.S.
hybrid maize, 1930 to 1980. pp. 15-47. In: W.R. Fehr (Ed.),
Genetic contributions to yield gains of five major crop plants.
Proc. Symposium Crop Sci. Soc. America and Am. Society
Agronomy, Atlanta, Georgia.
DUVICK D.N., J.S.C. SMITH, M. COOPER, 2004 Long-term selection
in a commercial hybrid maize breeding program. Plant
Breed. Rev. 24: 109-151.
FREY K.J., B. BRIMHALL, G.F. SPRAGUE, 1949 The Effects of Selec-
tion Upon Protein Quality in the Corn Kernel. Agronomy J.
41: 399-403.
LILBURN M.S., E.M. NGIDI, N.E. WARD, C. LLAMES,1991 The in-
flucence of severe drought on selected nutritional charac-
teristics of commercial corn hybrids. Poultry Sci. 70: 2329-
2334.
MILLER R.C., L.W. AURAND, W.R. FLACH, 1950 Amino Acids in
High and Low Protein Corn. Science 112: 57-58.
RUSSELL W.A., 1974 Comparative performance for maize hybrids
representing different eras of maize breeding. 29th Corn and
Sorghum Res. Conf. pp. 81.
RUSSELL W.A., 1984 Agronomic performance of maize cultivars
representing different eras of breeding. Maydica 29: 375-390.
SCOTT M.P., S. BHATNAGER, J. BETRAN,2004 Tryptophan and me-
thionine levels in quality protein maize breeding germplasm.
Maydica 49: 303-311.
TRACY W.F., I.L. GOLDMAN, A.E. TIEFENTHALER, M.A. SCHABER, 2004
Trends in Productivity of U.S. Crops and Long-term Selec-
tion. Plant Breed. Rev. 24: 89-108.
GRAIN COMPOSITION IN MAIZE VARIETIES 423
