Somatic embryogenesis, which is still the method of choice for tissue culture, regeneration and transformation of maize, is largely considered highly genotype-dependent. The Hi II, a highly embryogenic genotype, has been extensively used in transformation protocols. However, this is not an inbred line; instead, it has a proportion of the undesirable A-188 background, and the progeny segregates for phenotypic characteristics and shows poor agronomic performance. In an effort to identify genotypes that combine a high somatic embryogenic response and good agronomic performance, we evaluated 48 advanced inbred lines developed at INTA. Callus development and somatic embryogenesis capacity were measured in 200 immature embryos per line. Embryogenic capacity [EC (mature somatic embryos/callus evaluated) x 100], Regeneration Capacity (RC) and Fertile Plant Recovery in greenhouse (FPR, fertile plants/regenerated plants) were recorded. A total of 17 lines reached an EC > 50%, and 14 out of those 17 lines regenerated seedlings. The FPR ranged between 50 and 100%. Also, we selected three promising lines with high agronomic performance, as alternatives to Hi II, in order to be included in a maize transformation scheme via somatic embryogenesis. In addition, we report the usefulness of Single Sequences Repeat (SSRs) in the determination of genetic diversity among 14 divergent lines for somatic embryogenesis response. The seven lines displaying good in vitro behaviour can be crossed to obtain hybrids combining desirable alleles for somatic embryogenesis response and different genetic backgrounds.

Regression logistic techniques were used to study the environmental variables associated with fumonisin content in maize grain. The dependent variable consisted of 29 values of fumonisins B1 and B2 (FB1 and FB2) from samples of susceptible hybrid sowed in several sites of the Pampas region from 2006/07 to 2008/09. Meteorological variables were recorded in relation to the silking stage. The Np2 variable (number of periods of two days with simultaneous recording of rainfall and relative humidity > or = 81 % (day 1) and relative humidity > 70 % at the second day) was the most strongly correlated with FB1 and FB2 (Kendall coefficient: τ: 0.63 and 0.52 respectively) and was included in simple and bivariate regression logistic models with Txm (mean maximum temperature). A trivariate model integrated for two interactive variables and Txm achieved the highest accuracy of prediction, classifying correctly 93 % of cases. The environmental variables identified in this study stimulate F. verticillioides infection and can explain the variability in fumonisin levels at harvest. These results could be useful for the rational management of both disease and mycotoxin.

Técnicas de regresión logistica fueron utilizadas para estudiar las variables meteorológicas asociadas al contenido de fumonisinas en grano de maíz. La variable dependiente consistió en 29 valores de contenido de fumonisina B1 y B2 (FB1 y FB2) provenientes de muestras de un híbrido susceptible sembrado en varios sitios de la región pampeana, desde 2006/07 a 2008/09. Las variables meteorológicas fueron procesadas alrededor de la aparición de estigmas. La variable Np2 (número de períodos de dos días con registro simultáneo de lluvia y humedad relativa > o = 81 %, día 1, y humedad relativa > 70% en el segundo día) resultó la más fuertemente correlacionada con el contenido de FB1 y FB2 (coeficiente de Kendall: τ: 0,63 y 0,52 respectivamente) y fue incluida en modelos de regresión logística simple y bivariado, junto a Txm (temperatura máxima media). Un modelo trivariado integrado por dos variables interactivas y Txm logró la mayor precisión de predicción al clasificar correctamente el 93% de los casos. Las variables ambientales identificadas favorecen la infección de F. verticillioides, de esta manera se puede explicar la variabilidad en los niveles de fumonisina a cosecha. Estos resultados podrían ser útiles para el manejo racional de ambos, la enfermedad y la micotoxina.

A diallel experiment (Method 4 of Griffing's model I) was performed among nine corn (Zea mays) lines in order to estimate general and specific combining ability (GCA & SCA) to Mal de Río Cuarto (MRC) disease resistance. The selected environments were: Pergamino, Junín, Ferré (corn region VI); Sampacho and Holmberg (corn region IV). The severity of the disease, evaluated as attack average degree (AAD), was null in Pergamino and Junín, and 0.14 in Ferré, with significant differences between treatments. In Holmberg, it was 0.59 and in Sampacho it was 1.57. The analysis of variance for this region showed significant differences between environments and treatments. It did not show interactions. Average AAD of experimental hybrids did not differentiate from resistant control and it was different from commercial control (more susceptible). Twenty-five experimental materials did not differentiate from the better resistant control. Pedigree selection was effective in order to improve resistance to MRC. The single cross hybrids showed characteristics that would give high competitiveness in the seed market. The most resistant lines to MRC according to their GCA (the average performance of a line in hybrid combination) were 2526 and 2378. According to their SCA (cases in which certain combinations do relatively better, or worse, than would be expected on the basis of the average performance of the lines involved) the most resistant to MRC single cross hybrids were: 2378 x 2600, 2335 x 2600, LP561 x 2568-2 and 2526 x 2600. In Sampacho 42% of yield loss was for MRC, in Holmberg: 8%.

Se realizó un experimento dialélico entre nueve líneas de maíz (Zea mays) (Método 4, modelo I de Griffing (1956)) con el objeto de determinar Aptitud combinatoria general y específica (ACG y ACE) para Resistencia a Mal de Río Cuarto (MRC). Los ambientes seleccionados fueron Pergamino, Junín, Ferré (región maicera VI) y Sampacho y Holmberg (región maicera IV). La severidad de la enfermedad, evaluada a través del Grado Medio de Ataque (GMA), resultó nula en Pergamino y Junín, y de 0,14 en Ferré, con diferencias significativas entre tratamientos. En Holmberg fue de 0,59 y en Sampacho 1,57. El análisis de variancia para esta región detectó diferencias significativas entre ambientes y entre tratamientos, no detectándose interacción. El GMA promedio de los híbridos experimentales no se diferenció del promedio los testigos resistentes, si lo hizo de los testigos comerciales (más susceptibles). Veinticinco materiales experimentales no se diferenciaron del mejor de los testigos resistentes. La selección por pedigrí fue efectiva para mejorar este carácter. Los materiales experimentales en general demostraron poseer características que les darían competitividad en el mercado. Por su ACG, las líneas más resistentes a MRC fueron la 2526 y la 2378 (con efectos significativos y negativos: resistentes) por los efectos de ACE, se destacaron los híbridos: 2378 x 2600, 2335 x 2600, LP561 x 2568-2 y 2526 x 2600. En Sampacho el 42% de la disminución en el rendimiento pudo atribuirse al MRC, en Holmberg el 8%.