Full text
| |
| Author(s): |
Carlos Mauricio Basso
Total Authors: 1
|
| Document type: | Doctoral Thesis |
| Press: | Piracicaba. |
| Institution: | Universidade de São Paulo (USP). Escola Superior de Agricultura Luiz de Queiroz (ESALA/BC) |
| Defense date: | 2000-02-15 |
| Advisor: | Jose Branco de Miranda Filho |
| Abstract | |
Com stunt is a vírus and virus-like disease complex affecting maize crops in a wide area of American tropical and sub-tropical environments. In Brazil, according with several reports, its incidence has been increasing in the Iast years due to some new agronomic practices that allow the development of favorable epidemiological conditions. Com yield losses around 30% were estimated for susceptible genotypes. Because genetic resistance is the most efficient way of control, searching for new resistance sources to be incorporated in elite materiais or to be used as basic populations in recurrent selection schemes, is an important goal in Brazilian maize breeding programs. The objective of this assay was to predict and synthesize maize composites with good disease resistance and high yield and to estimate genetic parameters useful to breeding programs. For that purpose, ten maize improved populations from several public institutions and twenty two entries from germplasm bank, both groups previously selected for disease resistance, were crossed in two mating schemes: intergroups partial diallel scheme and intragroup top-cross scheme. Five maize synthetics, derived from five commercial hybrids with different leveIs of disease resistance, were used as testers for the "intergroup" scheme. In addition, thirteen maize exotic populations derived from temperate and temperate by tropical germplasm and not previously tested for com stunt resistance, were crossed with two well adapted composites in a partial diallel scheme. The experiments were evaIuated in randomized complete block experiments with four replications in three environments at São Paulo State (Brazil), under high natural com stunt occurrence in two of them. In general, the improved populations and bank entries showed better leveIs of disease resistance than the exotic material. Crosses among the adapted germplasm (top-cross scheme) were used to predict composite means for the identification of the best potential yield, for different number of parental varieties. In the partial diallel schemes, high leveIs of average heterosis were detected for yield in crosses from public and private sources of germplasm indicating a high degree of genetic divergence between them, so they were considered good sources to start a breeding programo Comparing the reI ative importance of sum of squares for com stunt resistance, varietal effects were more important than heterosis effects, indicating additive effects as the most important in determining disease resistance. However, no-additive effects can not be neglected. The last ones probably derive from the fact that hybrid vigor diminished disease severity symptoms in the most vigorous crosses. On the other hand, it was detected a high and negative phenotypic correlation between com stunt symptoms and yield. From such correlation, the importance of additive effects and the presence of hybrid vigor affecting disease symptoms, it is possible to suggest that the composites to be formed can be used as sources of germplasm for an intrapopulation recurrent selection programo The former should be based on the evaluation of yield in progenies with some leveI of inbreeding and under high disease pressure conditions, similar to that reported in this study. In addition, selections should emphasize the reduction of plant height, considered high in most of the predicted composites. (AU) | |