Measuring Racial Differentiation in Maize Using Multivariate Distance Measures Standardized by Variation in F2 Populations1

Publisher Website
Google Scholar
Abstract
Thirty F2 maize (Zea mays L.) populations were studied to evaluate the use of various statistical distance measurement between parents, relative to their F2 populations, and to estimate racial differentiation between maize races. The parental and F2 populations, obtained from crossing 47 maize races, were grown at Raleigh, N.C., or Homestead, Fla., as appropriate to their adaptation. Five morphological characters of the ear were measured. Six statistical procedures were utilized to measure racial divergence: Euclidean distance, Mahalanobi's distance, generalized distance, modified generalized distance, approximate Dempster's distance, and Dempster's distance. The relationship among the measures, their advantages in geometrical representation, and their facility of computation were discussed. The degrees of relationship estimated were in general agreement with previous studies based on classical taxonomic methods and multivariate analyses. These results suggest that morphological studies of racial F2 populations can be useful in understanding the variability and relationship among maize races.