Maize varieties with improved nitrogen(N)‐use efficiency under low soil N conditions can contribute to sustainable agriculture. Tests were carried to see whether selection of European elite lines at low and high N supply would result in hybrids with differential adaptation to these contrasting N conditions. The objective was to analyze whether genotypic differences in N uptake and N‐utilization efficiency existed in this material and to what extent these factors contributed to adaptation to low N supply. Twenty‐four hybrids developed at low N supply (L × L) were compared with 25 hybrids developed at high N supply (H × H). The N uptake was determined as total above‐ground N in whole plants, and N‐utilization efficiency as the ratio between grain yield and N uptake in yield trials at four locations and at three N levels each. Highly significant variations as a result of hybrids and hybrids × N‐level interaction were observed for grain yield as well as for N uptake and N‐utilization efficiency in both hybrid types. Average yields of the L × L hybrids were higher than those of the H × H hybrids by 11.5% at low N supply and 5.4% at medium N level. There was no significant yield difference between the two hybrid types at high N supply. The L × L hybrids showed significantly higher N uptake at the low (12%) and medium (6%) N levels than the H × H hybrids. In contrast, no differences in N‐utilization efficiency were observed between the hybrid types. These results indicate that adaptation of hybrids from European elite breeding material to conditions with reduced nitrogen input was possible and was mainly the result of an increase in N‐uptake efficiency.

Maize (Zea mays L.) cultivars with improved N‐use efficiency would be beneficial for low‐input production systems. Our objective was to estimate quantitative genetic parameters to optimize breeding programs for improving productivity under low N levels. Results of 21 field experiments with European breeding materials belonging to the flint and dent gene pool are presented. The study was performed during 1989 and 1999 at several locations in typical maize growing regions of Germany and France. All experiments were conducted at high (HN) and low (LN, no N fertilizer applied) N levels. Average grain yield was reduced by 37% at LN compared with HN. Coefficients of genotypic correlation between HN and LN were variable with an average of rG = 0.74 for grain yield and generally high for grain dry matter content. For grain yield, analyses of variance were computed from relative data, where plot values were expressed as percentage of the trial mean. Variances caused by genotype (G), G × location (L) interaction, and error effects were higher at LN compared with HN, with similar heritabilities at both N levels. For the untransformed data, components of variance were higher at HN than at LN. Genotype × N as well as G × L × N level interaction variances were significant in most experiments. Efficiency of improvement of grain yield at LN through indirect selection at HN was 70% compared with direct selection at LN. A trend toward increased efficiency of direct selection under LN conditions was evident with decreasing grain yield at LN.

Three-way cross means were predicted with formulae involving linear functions of general (GCA) and specific combining ability (SCA) effects estimated from single-cross factorials between genetically divergent populations. Data from an experiment with 66 single-cross and 66 three-way cross forage maize (Zea mays L.) hybrids was used for comparing the prediction formulae. The genotypic correlation (r) between observed and predicted three-way crosses increased with increasing χ, the weighting factor of SCA effects, for plant height and ear dry matter (DM) content. It displayed slightly convex curves for total and stover DM yield, ear percentage, and metabolizable energy content of stover. For Jenkins' method B, r was considerably less than 1.0 for all traits, indicating the presence of epistasis. The square root of heritability (hĜ) of the predicted means decreased with increasing χ, the reduction being small with a greater number of test environments. Using the product r·hĜ as a criterion of efficiency, none of the prediction methods was consistently superior and the differences among them were rather small (< 7.5%) for all traits, irrespective of the number of test environments. We recommend evaluating the GCA of a greater number of lines from each parent population in testcrosses with a small number of elite lines from the opposite population. All possible three-way or double crosses between both sets of lines should be predicted by Jenkins's method C. This procedure allows one to select with a higher intensity among the predicted hybrids and thus should increase the genetic gain.

Selection indices were calculated for simultaneous improvement of relevant forage maize traits. Beside indices based on primary traits [metabolizable energy yield (MEY) and content (MEC), crude protein content (CPC)] determined either m vitro or by near-infrared reflectance spectroscopy (NIRS), we calculated indices based on secondary (five agronomic) traits. The required phenotypic and genotypic covariance matrices were estimated from a series with 120 maize hybrids tested at three sites. Base index, heritability index, optimum index, and restricted index showed only minor differences in the expected gain in total economic worth (AH) for the present agricultural conditions in Germany. CPC was the least important primary trait; selection for grain yield, the most suitable secondary trait, resulted in greater AH values than any other index. Harvesting the whole plant and subsequently determining MEG by NIRS was the most economic selection procedure with increased economic weighting for MEC at high animal performance levels. In the latter situation, we recommend grain yield trials as a pretest in breeding programmes and official performance trials for variety registration.

The importance ol epistasis for forage performance and quality traits was studied in early maturing European maize {Zea mays L.) breeding materials. Sixty-six three-way crosses {3W), 33 of the flint X (dent -dent) and 33 of the dent x (flint • flint) type, were compared with the mean of their respective non-parental single crosses (2W) using data obtained in six environmenis. For each of the 11 traits examined, at least 6 out of the 66 three-way crosses displaved significant (P < 0.05) epistatic deviations, calculated as 2W -3W. Averaged over crosses and environments, epistatic deviations were highly significant (P < 0.01) for 8 traits, but of minor importance except for grain yield, ln general, the variance ot epistatic deviations was appreciably smaller than the variance of genotype X environment interactions. Highly significant correlations occurred among epistatic deviations for silking date, dry matter content of ear. and dry matter yield ot grain and stover. .The reliability of predicting the performance of rhree-way cros.ses from 2W as opposed to assessing them directly was investigated as a function of the number of test environments. Disturbmg effects of epistasis on prediction only became important under intense testing (> 3 or 4 environments). With fewer test environments, both procedures were almost equivalent, due to a smaller influence of genotype X environment interaction on the 2W values. Epistasis is concluded to be of minor importance with regard to the optimum type of hybrid for forage use.Key words: 2ea mays -forage maize -epistasisgenotype X environment interaction -hybrid prediction -type oi hybrid During the last decade, the acreage of forage maize production in Centra) and Western Europe has increased dramatically. Most of the hybrids currently used for forage maize production in this area arc thrcc-way crosses, whereas the majority of hybrids for grain production are single crosses.Epistasis IS one of several important aspects when contrasting the advantages and disadvantages of the two types of hybrids {for detailed discussion, see HAELAUHR and MIRANDA 1981). Eirstly, epistasis might cause differences in the average performance level of single and threeway crosses. Secondly, epistasis might impair the precision of three-way cross prediction and, consequently, increase the time and cost requirements for breeding new varieties of this type compared with single crosses.Investigations about epistasis m grain maize comparing balanced sets ot different types of hybrids according co BAUMAN'S (1959) method have been reviewed by SCHNELL (1975) and more recently by MELCHINGER (1984 a). Concerning the average performance level, most experiments with selected hnes established epistatic effects to the advantage of single crosses for grain yield, but differences were generally small. As to the prediction of three-way crosses, several authors (STUBER et al. 1973, STUBER and MOLL 1974, MEECHINGER et al. 1986) have reported that epistasis becomes more important than genotype X environment interaction...