A 20-year-old forest fertilization trial was used to investigate the effects of repeated P additions on P availability in the humus layer of a Norway spruce forest soil. N was supplied annually, and P, K, and micronutrients were supplied every 4th year. The last P application was made 2 years before the investigation started. Microbial P concentrations in the P + NK + micro-amended plots were about half as high as those in the control and the N-only treatment. In plots without P amendments, around 50% of the total P in the humus layer was found in microorganisms, whereas in P-amended plots the figure was around 25%. The block supporting more rapid tree growth, situated on the middle of a slope, showed a significantly higher microbial biomass P concentration than the less productive block at the bottom of the slope. Labile P concentrations did not vary between treatments and thus could not have directly contributed to the treatment-related differences in total microbial biomass P. Acid phosphatase activities were around three times lower in the sites treated with P + NK + micronutrients. Two sources are suggested for the acid phosphatases, active excretion by living roots and fungi and passive release from ruptured cells. For all eight plots investigated, there was a positive correlation (R = 0.83) between acid phosphatase activity and the microbial P concentration. The P concentration in current-year needles was the lowest in the N-only treatment at 1.13 mg g-1 dry weight, and the highest in the P + NK + micronutrients+lime treatment at 1.92 mg g-1 dry weight. The P:N ratio in needles varied from 0.115 in the P + NK + micronutrients + lime plots to 0.068 in the N-only plots. The latter value is at the level where P is considered to be the growth rate-determining nutrient.