Opportunity cost of postponed replacement (OC) is the income forfeited by keeping a cow for an extra day and is estimated by the income produced by an average replacement. The effect of adjusting a measure of net income, relative net income (RNI), for OC (RNIOC) by lactation was studied. After edits, the data set consisted of 2,982,001 Holstein cows. Prediction factors were developed for RNI and days of productive life (DPL) so that OC could be estimated from cows with shorter herd life opportunities. Within-herd correlations of RNI estimated from 84 month herdlife opportunity with that predicted from cows still alive at 36, 48, 60 and 72 months were .46, .59, .72 and .76, respectively and predictions reflected phenotypic trends of increased net income over time. Corresponding correlations for predicted 84 month DPL at the same ages were .28, .36, .41 and .47 and predictions conflicted with phenotypic trends of decreased herdlife over time. Total OC for cows with 84 month opportunity were raised by an average of $34 when OC was estimated specific to each lactation. The 433,116 cows with classification records and 84 month herdlife opportunity were used to estimate genetic and phenotypic (co)variances among type traits, production, and months in milk (MIM), RNI and RNIOC with a multiple trait Sire model. Production information from all cows in classified herds indicated that classified cows are not a random sample of cows in those herds. Heritability of RNIOC (.17) was higher than RNI (.12), but the genetic correlation between the traits was high (.97). Heritability of MIM was .06. Genetic correlations of MIM to the yield and linear type traits were less than -31 in absolute value. Evaluation of net merit using economic weights developed with RNIOC was more accurate than indirect prediction of MIM. Approximate reliability of a first crop AI sire evaluation for net merit is .65 compared to .42 for MIM. / Ph. D.