Biochemical and hematological analytes were used to better understand some of the pathophysiological processes involved in the transition period and contribute to the prediction of risk of disease, particularly retained placenta (RP); three studies were carried out. In the first study, based on prepartum biochemical and hematological profiles of 1038 Holstein cows, a multivariable logistic model was used to predict the risk of RP. Cholesterol and fatty acids were associated with a higher risk of RP. In the second study, blood samples from 37 Holstein cows in the week before calving and 47 cows in the week after calving were obtained at 1 hour (-1h) before and 4(4h) and 10(10h) hours after the first feed offering. Fatty acids were determined in the precalving samples and [beta]-hydroxybutyrate (BHB) in the postcalving samples. Lipoprotein electrophoresis were carried out on 3 sera of 10 randomly selected cows in each group. Thirty two percent of cows had >=0.4 mmol/L of fatty acids at -1h, compared to 16% at 4 h and 10h in the precalving cows. In the postcalving samples, the proportions of cows with >=1400 [mu]mol/L of BHB were 8.5%, 10.6%, and 12.8% respectively. The low density lipoproteins were markedly reduced in the postcalving group. Cows were twice as likely to be above the suggested cut-point for increased disease risk when they were sampled before feeding to measure fatty acids. Lipoprotein electrophoresis results were of limited utility for assessing energy status. In the third study, serum and whole blood measurements from 1078 high producing cows from 20 herds from Southern Ontario were used to determine reference values in the week prepartum and the week postpartum. Cows that presented clinical disease were excluded. In sera, BHB, fatty acids, cholesterol, urea, glucose, calcium and phosphorus were determined. In whole blood, WBC, and differential leukocytes were counted. There were significant differences between the precalving and postcalving groups for all the biochemical analytes and eosinophils. Other hematological analytes were not significantly different. The results were also different to reference ranges of mid-lactation cows. Differential reference values of transition cows are required for accurate interpretation of results.