| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Corn Breeding and Genetics, Dep. of Plant Sciences, North Dakota State Univ., Fargo, ND 58105-5051
* Corresponding author (marcelo.carena{at}ndsu.edu).
ABSTRACT
NDBS11(FR-M)C3 (Reg. No. GP-554, PI 650887), NDBS1011 (Reg. No. GP-555, PI 650888), and NDBSK(HI-M)C3 (Reg. No. GP-556, PI 650889), are three new maize (Zea mays L.) populations that serve as new sources for inbred line development targeted at 85-95RM hybrids. They were developed by the North Dakota State University maize breeding program and released by the North Dakota Agricultural Experiment Station (NDAES) in February 2007. NDBS11(FR-M)C3 derived from BS11(FR)C13 and three cycles of stratified mass selection for earliness in North Dakota. NDBS1011 was developed from BS10(FR)C13 x BS11(FR)C13 and four cycles of stratified mass selection for earliness in North Dakota. NDBSK(HI-M)C3 derived from BSK(HI)C11 and three cycles of stratified mass selection for earliness in North Dakota. Cycle populations were evaluated across nine North Dakota environments in 2005, 2006, and 2007. Stratified mass selection was confirmed to be an effective and inexpensive method for adapting populations (e.g., 2 d year–1) and also was demonstrated to produce significant positive correlated responses in economically important traits (e.g., 0.4 Mg ha–1 yr–1). Preliminary diallel studies suggested that NDBSK(HI-M)C3 shares the same heterotic group as BS22(R-T)C8 and NDSAB(MER-FS)C14 (e.g., early SSS), while NDBS11(FR-M)C3 shares the same group with BS21(R-T)C8, LEAMING(F-S)C6, and CGL(S1-S2)C5 (e.g., early non SSS). NDBS1011 seems to belong to an alternative heterotic group.
Abbreviations: GCA, general combining ability • NDAES, North Dakota Agricultural Experiment Station • NDSU, North Dakota State University • SCA, specific combining ability
NDBS11(FR-M)C3 (Reg. No. GP-554, PI 650887), NDBS1011 (Reg. No. GP-555, PI 650888), and NDBSK(HI-M)C3 (Reg. No. GP-556, PI 650889) maize (Zea mays L.) populations were coded as North Dakota populations once the process of adaptation to North Dakota was initiated. However, BS and K codes (B representing Iowa, S representing synthetic, and K representing landrace Krug Yellow Dent) and selection methodology codes (FR for reciprocal full-sib recurrent selection and HI for half-sib recurrent selection) were kept to recognize previous efforts of germplasm improvement at Iowa State University and at the Nebraska Agriculture Experiment Station. These populations were developed by the North Dakota State University (NDSU) maize breeding program and released by the North Dakota Agricultural Experiment Station (NDAES) on 5 Feb. 2007 (North Dakota State University, 2007). All three early versions of BS11(FR)C13, BS10(FR)C13 x BS11(FR)C13, and BSK(HI)C11 are now adapted to the northern U.S. Corn Belt. They were released due to their potential to produce early-maturing inbred lines with desirable quantitative traits and will serve as elite sources of new genetic diversity.
Materials and Methods
Development of Genetic Materials
BS11(FR)C13, BS10(FR)C13, and BSK(HI)C11 were acquired by the NDSU maize breeding program from Dr. Arnel R. Hallauer's corn breeding program at Iowa State University in 1999.
NDBS11(FR-M)C3 derived from BS11, originally designated as ‘Pioneer Two-Ear Composite’ and was developed by W.L. Brown at Pioneer Hi-Bred International, Inc. (Johnston, IA). It is a genetically broad-based population developed by crossing prolific germplasm with U.S. Corn Belt lines (Hallauer, 1967). Only BS11(FR)C2 (PI 550477) has been nationally registered. Thirteen cycles of reciprocal full-sib recurrent selection were performed for grain yield, grain moisture at harvest, and root and stalk lodging with BS10 (‘Iowa Two-Ear Synthetic’) as a tester across Iowa locations. The result was BS11(FR)C13. It was then brought to North Dakota for three cycles of stratified mass selection for days to silk emergence.
NDBS1011 was developed from a cross between BS10(FR)C13 and BS11(FR)C13 made in the 2000 Fargo, ND, breeding nursery. Only BS10(FR)C2 (PI550476) has been nationally registered. Four cycles of stratified mass selection for days to silk emergence were conducted on the population hybrid.
NDBSK(HI-M)C3 derived from open-pollinated variety Krug Yellow Dent developed at the Nebraska Agriculture Experiment Station. BSK was improved by 11 cycles of half-sib recurrent selection for stalk strength in Iowa [BSK(HI)C11] and three cycles of stratified mass selection for days to silk emergence in North Dakota. No registered germplasm was found for BSK or any of its derivations. Therefore, this germplasm is unique to the U.S. national collection.
Selection for Adaptation
Stratified mass selection for earliness was initiated in 2001 at the NDSU Research Centers located in Casselton and Prosper, ND. Each isolated field for stratified mass selection was planted in May each year, had approximately 20,000 plants at approximately 85,000 plants ha–1. To avoid pollen contamination, they were isolated at least 300 m away from any source of maize pollen, as recommended by Luna et al. (2001). Within all isolations, the main plot was divided into 20 subplots of approximately 1000 competitive plants with the grid system proposed by Gardner (1961). Ears were selected and harvested from subplots throughout the field. All competitive plants were included for selection except those that were stalk or root lodged. Within each subplot, the first 20 earliest silking plants were tagged at the ear node with color tags. Therefore, 400 plants were selected per population. The selected ears were harvested, dried at 45°C for 1 wk, shelled in balanced bulks of 20,000 kernels (ears were counted and shelled accordingly), and used for the next cycle of selection. In addition, two balanced bulks were saved and placed in cold storage. The process was repeated in 2002 and 2003. In 2004 NDBS1011(FR-M)C3 was grown in isolation, and cycle four was obtained as well. In addition, the different cycles (C0 to C3) were sown at the 2004 Fargo breeding nursery for seed production. This procedure assured same seed quality and age for evaluation. Cycle populations were evaluated together with checks across nine North Dakota locations in 2005, 2006, and 2007; this was identified as Exp. I.
Combining Ability
A diallel mating design without reciprocals was also performed at the 2005 Fargo breeding nursery. The mating design included nine populations for seed production: NDSAB(MER-FS)C14, LEAMING(S-FS)C6, BS21(R-T)C8, BS22(R-T)C8, CGSS(S1-S2)C5, CGL(S1-S2)C5, NDBSK(HI-M)C3, NDBS1011(FR-M)C4, and NDBS11(FR-M)C3. For more information on the populations used in this study and already adapted to North Dakota, see Carena and Hallauer (2001a,b), Carena (2005a,b), Melani and Carena (2005), and Carena and Wicks (2006). Pair-row crosses of six rows per population hybrid were used for a total of 216 rows. Detasselling was performed on male and female plants immediately after pollination to minimize number of pollinations and maximize sampling. At least 50 crosses were made from 100 parents for each population combination. The seed from each population hybrid was harvested in balanced bulks, dried at 45°C for 1 wk, shelled, and stored in cold storage. The diallel mating design produced a total of 36 population hybrids. These 36 hybrids together with four elite commercial hybrids representing a range of relative maturities (81RM Proseed 582Bt11, 84RM Wensman W5085Bt, 87RM Pioneer 39D82, and 90RM Monsanto DKC 40-05) were evaluated across five North Dakota environments in 2006 and 2007. This was identified as Exp. II.
Statistical Analyses
Plots were planted and harvested by machines adapted for small experimental plots. Individual ANOVAs were computed using SAS (SAS Institute, 1990) for traits within environments in both experiments. Data were collected and summarized on Excel files and then imported to SAS for analyses. Analyses of variance were performed for all traits at each location, as well as across locations using the SAS GLM and ANOVA procedures.
For Exp. I, analyses were performed in all locations and years based on a randomized complete block design with two replications per location. For 2005, 2006, and 2007, data were collected for emergence percentage, grain yield (adjusted on a 15.5% grain moisture basis), test weight, grain moisture at harvest, plant height, ear height, root lodging, stalk lodging, dropped ears, days to pollen shedding and silking, and ear traits (prolificacy, ear length, ear diameter, number of kernel rows per ear, and cob diameter). Weights, grain moisture at harvest, and test weight were measured at the time of harvest. Homogeneity of error mean squares was evaluated across environments and combined analyses of variance were computed across environments using individual observations for each trait. Expected mean squares were calculated following the rules of Schultz (1955) and were based on a mixed linear model that considered environments and replications as random effects and entries as fixed effects. Combined error mean squares (pooled error) were calculated by pooling the correspondent individual error mean squares weighed by their corresponding degrees of freedom. Mean comparisons were assessed by Fisher's protected least significant difference since it has been shown to be an adequate test for detecting differences (Carmer and Swanson, 1971). Establishing direct and correlated responses consisted of regressing the trait means of the selected C0 generations on cycles of selection. A direct response occurred when a selected trait changed, which in this case was the days to silking. A correlated response was a change in a trait that had not been selected; in this case, all the other traits except days to silking. Combined means of each cycle were calculated across environments and mean squares for cycles separated into linear, quadratic, and residual components. The mean squares were calculated using orthogonal polynomial contrasts (Carmer and Seif, 1963). The expected mean squares for location x cycle interaction were used to test the significance (P < 0.05) of cycles, linear and nonlinear (quadratic and residual). Selection response per year was estimated as the linear regression coefficient while for the nonlinear components (quadratic and residual), the response to selection was calculated as (mean of last cycle-mean of initial cycle)/number of cycles (years = cycles for this method since selection was made at a rate of one cycle per year). In this case, response was referred to as average response.
For Exp, II, ANOVAs were performed for all traits within and among environments as with Exp. I. Combined analyses of variance were performed after testing for error homogeneity. Analysis of variance by Gardner and Eberhart method IV was used for the estimation of general combining ability (GCA) and specific combining ability (SCA) effects as well as for the partition of the genetic and environmental effects (Gardner and Eberhart, 1966). The sources of variation genotypes and genotype x environment interaction were partitioned into their components using orthogonal polynomial contrasts, following the procedure described by Matzinger et al. (1959) for diallel mating designs across years and locations. Genotypes were considered as fixed effects, while environments and replications within each environment were considered as random.
Germplasm Characteristics
Field Evaluation
Experiment I
Across populations, highly significant (P 
0.01) differences were detected among selection cycles for most traits evaluated. Traits showing significant variation are shown in Tables 1
, 2
, and 3
. In addition, regression analysis for days to silking detected both linear and nonlinear responses across cycles. The average change in the number of days to silking over cycles of selection (direct selection) was close to –2.0 d yr–1 across populations with no unfavorable correlated selection responses of other economically important traits.
|
|
|
|
NDBSK, NDBS1011, NDBS11, and their derived selected cycles were tested after improvement to evaluate the direct and correlated responses to stratified mass selection for earliness. Significant responses to stratified mass selection for earliness were obtained at an approximate average rate of 2 d yr–1. Despite its limitations, mass selection was confirmed to be an effective and inexpensive method for adapting populations and was also demonstrated to produce significant positive correlated responses in economically important traits of quantitative inheritance such as grain yield (e.g., 0.4 Mg ha–1 yr–1). This research supports the cost-effective efforts to use stratified mass selection for adaptation in exotic improved populations.
Experiment II
This experiment was conducted to assess the combining ability effects of the advanced cycles (from the three populations of stratified mass selection) in relation to six elite North Dakota populations currently under genetic enhancement at the NDSU corn breeding program. Individual analyses were performed for each trait, and combined ANOVAs were computed across environments based on homogeneity of error mean squares data. Combined analyses of means showed significant differences (P 0.05) across genotypes for all traits studied except grain yield and root lodging. Even though the GCA and SCA values were both nonsignificant for grain yield, the sum of squares for SCA was higher than that of GCA, indicating that grain yield for these populations could be controlled largely by nonadditive (dominant and/or epistatic) genetic effects, which disagrees with most studies (Reif et al., 2005a). However, Matzinger et al. (1959) reported that the SCA was higher than GCA in highly selected material, which agrees with the type of genetic materials evaluated in this study. For most traits, GCA mean squares were highly significant, indicating the predominance of additive genetic effects. Preliminary data suggest NDBSK(HI-M)C3 seems to share the same heterotic group as BS22(R-T)C8, NDSAB(MER-FS)C14, and some Stiff Stalk–derived populations, while NDBS11(FR-M)C3 seems to share the same group with BS21(R-T)C8, LEAMING(F-S)C6, and CGL(S1-S2)C5. On the other hand, NDBS1011(FR-M)C4 seems to belong to an alternative heterotic group. However, assignment of these newly adapted populations to heterotic groups is still under investigation. Although NDBSK(HI-M)C3 could be the best of the three sources for improving early flowering, NDBS11(FR-M)C3 seems to carry most of the positive alleles for improving most of the desirable quantitative traits. Also, NDBS1011 seems to be a good source for improving and providing lodging resistance. We encourage further testing of NDBSK(HI-M)C3 x CGL(S1-S2)C5, NDBSK(HI-M)C3 x BS21(R-T)C8, NDBS11(FR-M)C3 x BS22(R-T)C8, NDBS11(FR-M)C3 x NDSAB(MER-FS)C14, NDBS1011 x CGSS(S1-S2)C5, and NDBS1011 x CGL(S1-S2)C5. The germplasm generated in this research have been helpful to increase the genetic diversity currently present on northern U.S. farms.
Population Maintenance and Distribution
NDBS11(FR-M)C3, NDBS1011, and NDBSK(HI-M)C3 maize germplasm were multiplied in the 2007 Fargo breeding nursery. A balanced bulk of 500 kernels per population was planted in 20 rows at a rate of 25 seeds per row for a total of 60 rows. Crosses within each population were produced among plants, avoiding full-sib family production. Also, tassels were removed from plants being used as males and females to reduce the number of pollinations needed and to maintain as much equal representation as possible from gametes. Lots of 200 kernels will be available for distribution, and populations will be multiplied by the breeder at NDSU and by the Plant Introduction Station in Ames, IA as needed.
Availability
North Dakota State University has transferred ownership of these materials to the NDSU Research Foundation. Requests should be made directly to Dale Zetocha (dale.zetocha{at}ndsu.edu). Material transfer, inbred research, and/or commercialization agreements will need to be signed before the breeder is authorized to send seed lots. Breeder seed will be maintained by the NDSU maize breeding program and will be distributed (200 kernels per request) from the corresponding author on approval by NDSU Research Foundation. After 20 yr from the date of publication, seed may be obtained from the National Plant Germplasm System.
Acknowledgments
The adaptation and improvement of early maturing maize germplasm is supported by the North Dakota State Board of Agricultural Research and Education (SBARE), the U.S. Germplasm Enhancement Maize Project, and the North Dakota Corn Council Utilization. This research formed part of the master's thesis of C. Eno.
Footnotes
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.
Received for publication November 23, 2007.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Soil Science Society of America Journal | Journal of Environmental Quality |
The Plant Genome | |||
