The use of incomplete block designs has gained wider acceptability among researchers in developing countries as a tool to control random variation, particularly for evaluation trials of many genotypes. In the CIMMYT Maize Program, alpha (0,1) lattice designs have been used since 1986 for evaluating groups of more than 200 genotypes, generally planted in single row plots. Relative efficiencies of 1.0-2.0 have been observed, but the use of the alpha-lattice designs has thus far been restricted mainly to genotype evaluation trials. There is little information on the use of these designs in agronomic experiments, although they could be useful where there is no factorial type relationship among treatments, where the number of treatments is relatively large (e.g., > 16), and where soil heterogeneity at the test site is high. Among the most important constraints to the use of these designs are: the lack of an appropriate language-independent software interface for entering and transforming data files, the difficulty of handling information in a common format, and the lack of integration with other analytical tools (graphing applications, selection algorithms, regression, etc.). A software module in English and Spanish that facilitates and integrates the use of the ALPHANAL and LATANOVA programs (Scottish Agricultural Statistics Service, Edinburgh)for the design and analysis of alpha (0,1) lattices is described. The software also allows for the processing of lattice-adjusted means using selection software developed by the CIMMYT Maize Program. This type of easy-to-use software offers an additional resource for improving the efficiency of agricultural research in national agricultural programs in developing countries.

Genetic Resources Program

Text in English

9802|AGRIS 9702|anterior|R97-98PROCE|FINAL9798

CCJL01