COMBINACIÓN DE LAS TÉCNICAS ESTADÍSTICAS MULTIVARIADAS Y EL DISEÑO AUMENTADO MODIFICADO (DAM) EN LA SELECCIÓN DE LÍNEAS DE PRUEBA EN EL PROGRAMA DE MEJORAMIENTO GENÉTICO DEL ARROZ (Oryza sativa L.).

The current development of researchers in the field of genetic improvement in Cuba and the necessity of economizing experimental area and resources outlines the necessity to value experimental designs that facilitate an appropriate efficiency with a maximum of economy. This work intends to combine the kindness of DAM, with the adjustment method for line-column to control the soil heterogeneity, and the statistical multivariate techniques, offering the investigator a mathematical tool in the selection of test lines in rice breeding programs. DAM was structured by means of a Latin Square (4x4), with four controls lines (INCA LP-5, INCA LP-2, J-104 and Reforma) and 120 test lines distributed at random in the corresponding subplot. The use of DAM that allows to compare a considerable number of test lines, overcoming the limitations of a not replied experiment, what presupposes an economic benefit for the area reduction, the economy of experimental material and control of the environmental heterogeneity, in combination with statistical multivariate techniques provides a methodology more efficient in the selection of promissory lines in rice breeding program. [ABSTRACT FROM AUTHOR]