Weaning weights of calves of dams at ages in years of 2, 3, and older were modeled to be three separate traits. Fixed effects were sex of calf-year of birth combinations for nine pure breeds and sex of calf-year of birth-generation for three composite populations. Random effects fitted for each trait were correlated direct and maternal genetic, maternal permanent environmental, and temporary environmental. Direct and maternal effects were correlated across traits. A multiple-trait, derivative-free REML algorithm was used to estimate the 30 (co)variance components. Number of animals per breed group ranged from 1,244 to 4,326. For the three traits for pure breeds, average proportions of phenotypic variance were .34, .31, and .27 for direct genetic; .16, .15, and .12 for maternal genetic; and .18, .20, and .17 for maternal environmental effects. Average correlations among the three traits were .84 for direct genetic, .78 for maternal genetic, and .71 for maternal environmental effects. Average of direct-maternal genetic correlations for pure breeds was .05. For the composite breeds, average proportions of phenotypic variances were .44, .46, and .36 for direct genetic; .06, .06, and .05 for maternal genetic; and .16, .14, and .14 for maternal environmental effects. Average correlations among the three traits were .93 for direct genetic, .76 for maternal genetic, and .85 for maternal environmental effects. Average direct-maternal genetic correlation was .09 for composites. No evidence was found for greater direct-maternal genetic correlation for earlier than for later ages of dam. Sign and magnitudes of direct-maternal genetic correlations seemed to differ among pure breeds and were reflected in composites from those parent breeds.
