One hundred thirty-nine pigs, with an average initial weight of 15.9 kg, were used in three trials to evaluate plasma alkaline phosphatase (AP) as a criterion for estimating available P in feedstuffs for swine. The first two trials were designed to determine the effect of adding graded levels of P (as KH2 PO4) to a basal diet on the AP response pattern and degree of linear fit with time. Plasma AP increased (P<.05) with time and, when regressed on the concentration of total P (.26 to .50%), provided the best linear fit by d 14 (r = −.996). This high degree of fit was maintained through d 21 (r = −.983). The poor correlation observed at 7 d (r = −.828) indicated that the AP pattern was not fully established, whereas departure from the close, association by 28 d (r = −.913) may be related to a decline in the magnitude of P deficiency with increasing weight. In a third trial, available P was determined in high moisture corn (HMC, 24.6% moisture) using both AP and bone breaking strength (BKS) as response criteria. A standard curve was established by supplementing the basal diet (.26% P, .65% Ca) with graded levels of KH2 PO4. High moisture corn was substituted for dextrose on an equal dry matter basis. Blood samples were acquired on d 14 and 21 for AP analysis and pigs were killed on d 28 for determination of metatarsal BKS. Phosphorus from HMC was determined (slope-ratio technique) to be 41.3% available using BKS as the criterion (KH2 PO4, r = .961; HMC, r = .994) and 43.8% when AP at 14 d was used (KH2 PO4, r = −.988; HMC, r = -.999). The data indicate that AP is a sensitive index potentially useful in P availability studies.