| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Agronomy, Univ. of Wisconsin-Madison, Madison, WI 53706
b Dep. of Animal Science, Michigan State Univ., East Lansing, MI 48824
* Corresponding author (deleonn{at}msu.edu)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: BLUP, best linear unbiased prediction • REML, restricted maximum likelihood
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Before the advent of mechanized harvesting starting in the late 1930s, prolificacy was an undesirable trait. Humans tended to select for the concentration of seed resources in a single large ear borne at the apex of the primary lateral branch for ease of hand-harvesting. It is likely that early maize was nearly as prolific as its ancestor, teosinte, and that systematic selection by humans led to the predominant phenotype of single-eared plants with few tillers. There is, however, considerable evidence that prolificacy is associated with higher yields in maize (Uhr and Goodman, 1995; Maita and Coors, 1996; Duvick, 1997)
A program of mass selection for prolificacy in the Golden Glow open-pollinated maize population [GG(MP)] was initiated in 1971 at the University of Wisconsin, Madison. Details of the selection process have been previously described (de Leon and Coors, 2002). The selection program has been successful, and the average number of ears plant–1 increased from 1.6 at Cycle 0 to 4.9 at Cycle 24, as measured under a low-planting density of 17000 plants ha–1 (de Leon and Coors, 2002). The morphology of prolific plants from advanced cycles of GG(MP) has diverged from that of a typical maize plant. Many of the nodes of the main stalk of these plants produce very long primary lateral branches, which produce secondary lateral branches terminating in multiple female inflorescences. Observed ear number has been as high as 25 ears plant–1 among S1 plants derived from extreme individuals. Over cycles of selection in GG(MP), the increase in number of ears plant–1 resulted from a decreased suppression of axillary meristems at multiple nodes of the main stalk. Other observed characteristics include an increase in lateral branch length and an increase in number of tillers plant–1 (de Leon and Coors, 2002). In recent cycles, the most extreme plants in this population, when grown under optimal conditions and low planting densities, have a highly tillered morphology, similar to that of teosinte.
The process of maize domestication involved an increase in apical dominance, which is associated with increased concentration of resources in the main stalk and subsequent suppression of lateral branches (Doebley et al., 1997). Selection for prolificacy in GG(MP) has resulted in a decrease in apical dominance, the opposite of the domestication process (de Leon and Coors, 2002). In fact, the morphological features of plants from advanced cycles of GG(MP) suggest that the changes accompanying selection for prolificacy may be associated with some of the same genetic regulatory systems involved in domestication, at least in regard to its branching patterns. Some of these genetic mechanisms have been extensively studied in maize (Doebley et al., 1995; Doebley and Wang, 1997).
Before a genetic mapping study was conducted testing this hypothesis, 194 F3 families developed from the cross of a single highly prolific S1 plant derived from GG(MP) Cycle 23 and inbred line A679 were phenotypically evaluated. The objectives of this study were to (i) assess phenotypic variation for prolificacy and a number of related morphological traits in this mapping population, and (ii) examine the manner and magnitude of associations among traits to determine the likelihood that some of these traits are controlled by similar genetic factors.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
In the summer of 1996, approximately 0.8 ha was planted with GG(MP) Cycle 23 at a density of approximately 28000 plants ha–1. Approximately 50 of the most prolific plants were self-pollinated to give rise to a series of S1 families. In the summer of 1997, one of the most prolific S1 families was selected, and the most prolific plant of that family was self-pollinated and also crossed to inbred line A679, using A679 as a female parent. The S2 family from the self-pollinated S1 plant used in the cross with A679 is henceforth referred to as the GG(C23) parent of the mapping population. Inbred line A679 was released in the late 1980s by the University of Minnesota, and it was derived from the cross between line A662 and B73, followed by four generations of backcrossing to B73 (Gerdes et al., 1993).
In the winter of 1997–1998, the F1 family produced from the cross between the prolific Golden Glow S1 plant and A679 was planted, and a single F1 plant was self-pollinated. The F2 (S0) population derived from that plant was used for this project. In the summer of 1998 and winter of 1998–1999, a large number of plants from this F2 population were self-pollinated, giving rise to the 194 F3 (S1) families used in this study.
Experimental Design
The experiment was conducted in three environments: Madison, WI, 1999; Arlington, WI, 1999; and Madison, WI, 2000. Soils at both locations were Plano silt loam (fine-silty, mixed, mesic Typic Argiudoll). There were two replications in each environment. The 194 F3 families and the parents were randomly assigned in a randomized complete block design at each environment. Parents A679 and GG(C23) were included in two plots each in all four replications in 1999. In 2000, a different seed source was used for A679 than that used in 1999, and it appeared that this source was contaminated or misclassified. Therefore, no observations were recorded for A679 in 2000 at Madison.
Plots consisted of two 6.1-m rows with 0.8 m between rows. Fifteen kernels were hand-planted plot–1 and later thinned to seven plants row–1, which gave rise to a final density of 15062 plants ha–1. This low planting density allowed the expression of morphological characteristics without substantial plant-to-plant competition. All trials were fertilized according to recommendations provided by the UW Agricultural Research Stations and the UW Soil and Plant Analysis Lab. At Madison in 1999 and 2000, the study was planted in a field in which alfalfa was the previous crop. In both years, approximately 90 kg ha–1 N was applied in the form of manure during the fall and winter before planting. In spring of 2000, an additional 100 kg ha–1 N was applied in the form of urea. At Arlington, the 1999 study was planted in a field in which soybeans was the previous crop, and 130 kg ha–1 N was applied in the form of anhydrous ammonia before planting. Weeds were controlled by application of cyanazine [1,3,5-triazine, 2-chloro-4-(ethylamino)-6-(cyano-1-methyl)-(ethylamino)-, 1.7 kg ha–1 a.i.] and alachlor [2-chloro-2',6'-diethyl-N-(methoxymethyl)acetanilide, 0.9 kg ha–1 a.i.] before plants emerged, and plots were also weeded by hand throughout the season.
For both locations in 1999, 5 plants plot–1, and in Madison in 2000, 10 plants plot–1, were measured for a number of morphological traits. All traits, except mid-pollen date, mid-silk date, and anthesis–silk interval were measured between the second and the fourth week after fertilization at the three environments. The flowering time traits were measured only at Madison in 1999 and 2000 during the period of silk elongation and pollen shedding. The mean of all plants measured in each plot was used as the experimental unit for statistical analyses.
Description of Phenotypic Traits
Variation of sixteen traits was analyzed in the 194 F3 families (Table 1). For ease of discussion, these traits were divided into two main groups, those tentatively associated with activation of axillary meristems and those tentatively associated with the activation of intercalary meristems. The assignment of traits to these two groups provided a convenient organizational framework, but, at the current time, the physiological basis for this framework remains open to challenge. A third group of traits represented the flowering dates of this population.
|
Within the second main group of traits, plants were characterized by the elongation of different plant parts. The main stalk total length (Trait 10) was measured as the distance from the first aboveground node to the uppermost node at the base of the flag leaf, and the main stalk average internode length (Trait 11) was calculated as the ratio between the total length and the number of aboveground number of nodes on the main stalk. The total length of the uppermost lateral branch (Trait 12) and the lowermost tiller (Trait 13) were measured from the branch point of the lateral branch on the main stalk (or, in the case of tillers, ground level) to the node at the base of the terminal inflorescence of that branch.
Traits related to flowering dates were mid-pollen (Trait 14) and mid-silk (Trait 15) dates, which were recorded as days from 1 July. The anthesis–silk interval (Trait 16) was calculated as the difference between the mid-silk and mid-pollen dates.
Statistical Analyses
Independent analyses were performed for each trait. A linear random-effects model was used for modeling progeny data. Analyses were performed using the MIXED procedure of SAS (Littell et al., 1996). Restricted maximum likelihood (REML) was considered for estimating variance components. Best linear unbiased prediction (BLUP) was used to predict progeny values (Bernardo, 2002). Whenever the interaction between families and environments was significant, Spearman (rank) correlation coefficients of predicted values of families in different environments were used to determine whether interactions were due to variation on the magnitude or the direction of the response. Broad-sense heritability on a family-mean basis was estimated for all traits. Exact confidence intervals (95%) for the heritabilities were calculated using the F distribution, as described by Knapp and coworkers (1985).
Histograms and kernel density estimations (Scott, 1992) for the distribution of progeny values were computed for each trait using the S-Plus software (Venaples and Ripley, 1996). Means and standard deviations were calculated for each trait, and the Shapiro-Wilks statistic from the CAPABILITY procedure of SAS (SAS Institute, 2000) was used to study each empirical distribution.
Confidence intervals (95%) of progeny means for each trait were used to compare progeny to the parental lines. A parental mean was considered to be significantly different from the average of the progeny families when their values were outside of the progeny confidence interval limits. A linear mixed-effects model was adopted for the comparison of the two parental lines, which included the fixed effects of each line, and the random effects of environment, replication, and interaction of lines and environment. The REML approach was considered for estimating variance components and best linear unbiased estimation (BLUE) was used for estimating parental means. To study associations between traits, phenotypic correlations were estimated among all traits using family BLUP predictions across environments. All statistical tests were performed using a nominal 0.05 level of significance. Correlations indicated with ** were significant at a nominal level of 0.01.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Variation among F3 families was highly significant for all 16 traits. None of the traits had a significant effect of environment. Environment-by-family interactions, however, were observed for all traits except number of aboveground nodes on the main stalk, main stalk internode length, and position of the uppermost ear on the main stalk. For the other 13 traits, highly significant Spearman correlations were found between the three environments (r values ranged from 0.82 to 0.99—data not provided). This suggested that the interactions between family and environment mostly involved changes in magnitude and not in the direction of the response in the different environments, and, therefore, all traits were analyzed across environments.
Heritability estimates on a family-mean basis were >0.6 for all traits (Table 1). These medium-to-high heritability estimates suggested that these traits had a relatively simple type of inheritance and/or were only moderately affected by environmental or other random, nongenetic factors.
The morphological traits described in this study were assigned to three major groups based on either the type of meristem (axillary or intercalary) that is activated during their development, or their association with flowering time. Six traits had normally distributed progeny values (total number of ears plant–1, number of ears on the main stalk, main stalk total length, number of aboveground nodes on the main stalk, main stalk internode length, and position of the uppermost ear on the main stalk). The nature of the distribution for the remaining traits will be discussed in the following sections.
Traits Associated with Axillary Meristem
Ears per Plant
Parent GG(C23), which had been intensively selected for increased number of ears plant–1, had more ears plant–1 than parent A679, and it was equal to the progeny mean (Table 2). Similarly, the two parents differed for number of ears on the main stalk and number of ears on the tillers (Table 2). In the case of number of ears on the main stalk, the progeny mean was also similar to parent GG(C23) and different from A679. For number of ears on tillers, the progeny mean was not different from either parent (Table 2).
|
|
|
Nodes on the Main Stalk
Parent A679 had, on average, approximately one more aboveground node on the main stalk than parent GG(C23), and the A679 value was similar to the progeny mean for this trait (Table 2). A small number of F3 families had fewer number of aboveground nodes on the main stalk than GG(C23), and >50% of them had a greater number of aboveground nodes on the main stalk than A679 (Table 3). Parent GG(C23) had fewer number of aboveground nodes on the main stalk than A679; nevertheless, a larger percentage of its nodes were active. This is apparent from the higher total number of ears plant–1 and the number of ears on the main stalk of GG(C23) compared with A679 (Table 2).
Tillers
A small number of A679 plants produced a single long lateral branch at the first aboveground node on the main stalk. The morphology of these lateral branches resembled that of a tiller in regard to length and proximity to the ground level, and, for the purpose of this study, that is how they were classified. These tiller-type lateral branches nevertheless always ended in either a mixture of female and male inflorescences or a purely female inflorescence, never a purely male inflorescence as is typical of tillers. For parent GG(C23), on the other hand, an average of 1.14 tillers plant–1 were produced, and they were terminated by male, female, or mixed inflorescences (Table 2).
The progeny distribution for number of tillers plant–1 was slightly skewed toward low number of tillers (Fig. 1). Parent GG(C23) had more tillers plant–1 than A679, and neither one of them was different from the progeny mean (Table 2). The interpretation of the comparison of parents vs. progeny for this trait might have been also confounded by the effect the interaction with the environment. For the two locations in 1999, plants of parent GG(C23) tillered frequently. Tillers were much less frequent for A679 individuals. At Madison in 2000, the frequency of this trait was significantly lower in GG(C23) (data not provided), and no information was available for A679.
The relatively high heritabilities (Table 1) and strong association found between total number of ears on the main stalk and tillers plant–1 (r = 0.69**) indicated that these two traits are possibly controlled by a small number of common genetic factors. The progeny values for total number of ears plant–1 and number of tillers plant–1 were similar to parent GG(C23) (Table 2). However, only 40% of the F3 families had more ears plant–1 than GG(C23), whereas 84% had more tillers plant–1 than the GG(C23) parent (Table 3, Fig. 1). This suggested that, even though both traits were probably controlled by a similar set of genetic factors, the progeny had a greater ability to produce vegetative primary lateral branches than reproductive organs.
None of the tillers produced by parent A679 terminated in a purely male inflorescence, whereas 31% of the GG(C23) tillers ended in tassels (Table 2). The progeny distribution for frequency of male inflorescence at the end of tillers was slightly skewed toward a lower percentage of purely male inflorescences. The progeny mean for this trait was similar to the midparent value. Approximately 57% of the progeny families had mean frequencies below the midparent value. And about 90% of the progeny values were found within the range of the parents (Table 3). The form of progeny distribution, the low frequency of transgressive segregants, and the relative medium-to-low heritability for this trait compared with other traits suggested that the genetic basis for frequency of male inflorescence at the end of tillers was either highly polygenic, or if conditioned by a relatively simple type of inheritance, a substantial amount of the progeny variation was nongenetic.
Uppermost and Lowermost Ears on the Main Stalk
No difference was found between parents A679 and GG(C23) and between either parent and the progeny mean for the position of the uppermost ear on the main stalk (Table 2). The average position of the lowermost ear on the main stalk of GG(C23) was below that of A679 (Table 2), and, even though a high percentage of the progeny families had their lowermost ear positioned at a lower position than GG(C23), their means did not differ (Tables 2 and 3). The progeny distribution for this trait was skewed toward lower values.
The total number of ears plant–1, as well as the number of ears on the main stalk, were positively correlated with position of the uppermost ear on the main stalk (r = 0.23** and r = 0.18, respectively) and negatively correlated with position of the lowermost ear on the main stalk (r = –0.76** and r = –0.78**, respectively). This indicated that the formation of more ears occurred in both directions, toward the top and the bottom of the plant. However, the correlation with the position of the lowermost ear was greater than with the uppermost one, indicating that the activation of nodal meristems toward the bottom of the plant was more prominent than in the opposite direction.
As previously mentioned, GG(C23) had one less node on the main stalk than A679, and since the position of the uppermost ear on the main stalk was similar for both parents, the uppermost active node of GG(C23) was one node closer to the main tassel than parent A679. The position of the uppermost ear on the main stalk for the progeny was similar to the parental average. The number of aboveground nodes on the main stalk for the progeny was similar to A679. The position of the lowermost ear on the main stalk for the progeny, on the other hand, was similar to GG(C23). Therefore, although the progeny had a mean number of ears on the main stalk similar to parent GG(C23), the distance from the uppermost active node on the main stalk to the main tassel for the progeny was one more node away from the tassel than for parent GG(C23).
Frequency of Plants with Completely Active Regions on the Main Stalk
On average, 12% of all the GG(C23) individuals measured had a so-called completely active region on the main stalk, whereas this region was completely absent in parent A679. The mean frequency of plants with completely active regions on the main stalk for all the F3 families was 11% (Table 2). A high percentage of the progeny had values that were intermediate between the two parents (Table 3). The distribution of progeny values for frequency of completely active regions on the main stalk was skewed toward a lower percentage of completely active main stalks. However, in a few F3 families, the frequency of these regions was up to three times higher than parent GG(C23) (Fig. 1).
The presence of this active region was a peculiar characteristic frequently seen in Golden Glow plants from the most advanced cycles of selection for prolificacy. The absence of apical dominance may permit activation of meristematic tissue at multiple nodes on the main stalk and formation of primary, secondary, and higher-order lateral branching, which is typical of teosinte. The activation of a large number of nodes on the main stalk in the GG(MP) resembles the morphology of teosinte.
In general, traits associated with axillary meristems, such as total number of ears and number of tillers plant–1 tended to be associated with each other. However, no correlation was observed between number of aboveground nodes on the main stalk and total number of ears plant–1, number of ears on the main stalk, and number of tillers plant–1. The lack of association between the number of aboveground nodes on the main stalk and these traits associated with axillary meristems suggested that the number of ears was more related to activation of axillary meristems at existing nodes, a trait most likely conferred by GG(C23) than to the presence of additional nodes on the main stalk, a trait most likely conferred by parent A679. This observation also indicated that the presence or absence of ear shoots and/or tillers is probably controlled by genetic factors that are somewhat independent from those responsible for the expression of the potential number of tillers and lateral branches, which depends on the number of nodes present in the main stalk.
In a previous study of selection response for prolificacy in Golden Glow, number of aboveground nodes on the main stalk was highly and positively associated with total number of ears plant–1 and number of tillers plant–1 (de Leon and Coors, 2002). However, this previous correlation was calculated across 24 cycles of selection and appeared to be confounded with the effect of unintentional selection for maturity during the first 12 cycles in GG(MP).
Traits Associated with Intercalary Meristems
Main Stalk
Parent A679 had longer main stalks than GG(C23), but there was no difference in main stalk internode length (Table 2). Many progeny families were taller and had numerically longer internodes than the tallest and longest-internode parent (A679), but their average values were similar (Table 3, Fig. 1).
Main stalk total length was positively correlated with number of aboveground nodes on the main stalk (r = 0.46**) and main stalk internode length (r = 0.79**). However, no association was found between main stalk internode length and number of aboveground nodes on the main stalk, and it appeared that these two traits contributed independently to the total elongation of the main stalk.
Length of Lowermost Tiller
No differences were found between parents or between the progeny and the parents for lowermost tiller length (Table 2). However, 90% of the progeny had longer tillers than GG(C23), the parent with the longest tillers (Table 3). The empirical distribution for this trait deviated from a normal distribution (Fig. 1).
Tillers are considered basal reiterations of the main stalk in maize (Moulia et al., 1999). It was expected that the essential morphology of tillers and main stalk would closely parallel each other, whereas lateral branches would be distinct from the main stalk, not only in the sex of the terminal inflorescence, but also in a number of other physiological traits. Expectations regarding similarities between main stalk and tillers were met in this study. Positive correlations were found between the lowermost tiller length and main stalk total length (r = 0.55**) and main stalk internode length (r = 0.42**). Positive correlations were also found between number of ears on the main stalk and the number of ears on tillers (r = 0.33**).
A correlation was also found between lowermost tiller length and the type of terminal inflorescence (r = 0.39**). The frequency of male inflorescences at the end of tillers increased with increasing tiller length. This same observation was reported by Doebley and Stec (1993) for progeny resulting from a cross of maize and teosinte. This association may be related to a sexual threshold zone that reflects the hormonal regulation of sexual differentiation in maize (Iltis, 2000).
Uppermost Lateral Branch
The progeny distribution for the uppermost lateral branch length was slightly skewed toward lower values. The uppermost lateral branch length of GG(C23) was approximately twice that of A679, and both parents differed from the progeny mean (Table 2). Approximately 90% of the progeny values fell within the range defined by the two parents, and about 50% of the progeny values were above the midparent value (Table 3). This pattern was contrary to the response observed for the lowermost tiller length for which >90% of the progeny values were above that of the high parent value (Table 3).
In general, traits such as the main stalk total length, main stalk internode length, uppermost lateral branch length, and lowermost tiller length, which reflect activity of the intercalary meristem, tended to be positively correlated with each other. No association, however, was found between the length of the uppermost lateral branch and the lowermost tiller, which, coupled with the difference in the progeny segregation patterns of these two traits, suggested that the developmental control of tillers is independent from the genetic control of lateral branch development.
In few cases, intercalary meristem activated traits were also associated with bud initiation traits. For example, number of tillers plant–1 was negatively correlated with the uppermost lateral branch length (r = –0.16) and positively correlated with lowermost tiller length (r = 0.34**). This indicated that families with more tillers plant–1 tended to have longer tillers and shorter uppermost lateral branches, suggesting that allocation of limited resources determined the relative size of the two types of lateral braches located near the top and bottom of the plant, and that in some instances, traits associated with axillary and intercalary meristems appear to be controlled by common factors.
Although some of the traits within the two groups (traits associated with axillary meristem and intercalary meristem) tended to behave in similar ways, a number of inconsistent correlations of traits within groups indicate that there is no clear genetic evidence supporting the hypothesis that common genetic factors affect all the traits in a group. Therefore, each trait has to be considered separately in the subsequent QTL study.
Traits Associated with Flowering Time
Mid-pollen date, mid-silk date, and anthesis–silk interval were used to assess variation in flowering time. The progeny empirical distribution of all three traits deviated from normality. No differences were found between the two parents and between parents and the progeny mean for mid-pollen and mid-silk dates (Table 2). The parents differed for anthesis–silk interval, but neither parent was different from the progeny mean for this trait (Table 2). Ninety-six percent of the F3 families had anthesis–silk interval values that were intermediate between the two parents (Table 3).
Increased prolificacy is usually associated with decreased anthesis–silk interval because prolificacy typically indicates a reduction in apical dominance leading to greater synchrony of male–female flowering (Maita and Coors, 1996). Our results were somewhat contradictory to this trend, but the stark differences in morphology between A679 and GG(C23) may have affected our observations. Parent GG(C23) had many more active ear shoots than A679 and, therefore, silk emergence occurred across a longer period of time.
There was a positive correlation between mid-pollen and mid-silk dates (r = 0.86**). A positive correlation was also found between mid-silk date and anthesis–silk interval (r = 0.39**). However, the correlation between anthesis–silk interval and mid-pollen date was not significant, indicating that anthesis–silk interval was more influenced by variation in silk emergence than pollen date.
A number of traits related to elongation of plant parts were associated with mid-pollen date and mid-silk date. For example, mid-pollen and mid-silk dates were negatively correlated with uppermost lateral branch length (r = –0.20** and r = –0.22**, respectively) but positively correlated with lowermost tiller length and main stalk total length (r = 0.27** and r = 0.27**; r = 0.25** and r = 0.15, respectively). This indicated that early flowering plants tended to have longer and more vigorous uppermost lateral branches, but they also tended to be shorter and have fewer tillers.
Mid-pollen and mid-silk dates were not correlated with main stalk internode length; whereas the two flowering traits were positively correlated with number of aboveground nodes on the main stalk (r = 0.42** and r = 0.29**, respectively). This suggested that plant maturity, as reflected by flowering dates, was associated with the production of nodes during development, but was not strongly associated with internode elongation on the main stalk.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
An outstanding plasticity was observed in the expression of a number of these morphological characteristics. A number of those traits appeared to be highly influenced by the level of heterozygosity of the plant. Parent A679 is an inbred line, and GG(C23) is a sib-pollinated S1 derived from a single plant from Cycle 23 of the Golden Glow selected population. The level of heterozygosity in the majority of F3 families was probably higher than either parent. Therefore, the progeny would be expected, on average, to exceed the parental mean for those traits influenced by heterosis.
As in a previous study of Golden Glow (de Leon and Coors, 2002), a number of morphologically related traits were highly correlated with each other. These involved traits mostly associated with either the activity of axillary meristems or those related with the activity of the intercalary meristem. In most instances, the correlations of traits across these two groupings were insignificant or of lesser magnitude than correlations within each group, although there are some exceptions to this generality.
For the most part, our results with the current mapping population were similar to previous results seen when evaluating progress through 24 cycles of selection in GG(MP) (de Leon and Coors, 2002). The differences between these two studies most likely involve the manner in which correlations were obtained. In the previous study, correlations represented morphogenetic associations caused by the effect of selection over time, and may have involved indirect or unintentional selection for several traits along with prolificacy. In the present study, however, the correlations better represent the genetic coordination of different traits derived from a single F1 genotype at a given point in time. Therefore, the current mapping study may present a more direct assessment of true genetic associations among traits.
|
ACKNOWLEDGMENTS |
|---|
Received for publication October 3, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  N. de Leon, J. G. Coors, and S. M. Kaeppler Genetic Control of Prolificacy and Related Traits in the Golden Glow Maize Population: II. Genotypic Analysis Crop Sci., May 27, 2005; 45(4): 1370 - 1378. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 | |||
