The goal of this study was to test the hypothesis that the ratio of liver to bone alkaline phosphatase in blood plasma reflects the ascorbate status in scurvy-prone teleost fish (rainbow trout [Oncorhynchus mykiss]). The studies focused on finding a method for distinguishing bone alkaline phosphatase present in blood plasma from other alkaline phosphatase isoforms. We tested temperature optima and thermostability of liver, kidney, gill cartilage, and intestinal alkaline phosphatases. We did not observe differences among liver, bone, and kidney enzymes with respect to temperature optima and thermostability. We partially purified alkaline phosphatase from juvenile rainbow trout vertebrae and liver using n-butanol solubilization and ammonium sulfate fractionation. We found a difference between bone alkaline phosphatase, which precipitated in 0%-20% ammonium sulfate saturation, and liver enzyme, which required 40%-50% ammonium sulfate saturation to precipitation. We conducted a series of urea inactivation studies on partially purified enzymes from liver and vertebrae. Urea differentially inhibited the enzymes with t 1/2 = 1.1 and 0.4 min, for bone and liver, respectively. Subsequently, we subjected blood plasma alkaline phosphatase to urea inhibition, and using regression analysis we calculated the ratio of liver to bone alkaline phosphatase. We found that thus obtained ratios of bone enzyme in blood plasma correlated with liver ascorbate concentration. Bone alkaline phosphatase declined in ascorbate deficiency 10-fold, whereas low ascorbate status resulted in a 3.5-fold decrease. In order to draw a general conclusion on the linearity of the response of blood plasma/bone alkaline phosphatase as an indicator of ascorbate deficiency in fish, further studies must include analysis of individual fish followed in the process of developing avitaminosis.