The P status of the calcareous soils described by Russell' was evaluated in two greenhouse experiments using ryegrass as a test crop. Soils were arranged by yields into very responsive, moderately responsive and slightly responsive groups and mean values for all methods of soil analysis1 increased in the same order. P uptake by ryegrass, 40 days after sowing, from soils given superphosphate for 5 to 7 years in the field correlated most closely (r = o.ggo-o.g5g) with o.g~-sodium bicarbonate-soluble P and least closely (r = 0.822-0.697) with 1% citric acid-soluble P.
P uptake during the first weeks of growth depended partly on the monocalcium phosphate potential measured on soils before cropping, whereas total P uptake in 4 months was closely correlated (r = 0.979) with the labile P the soils contained but was almost independent of the initial phosphate potential. The soils from different experiments differed more in the total quantity of P in the labile pool than in their capacity, defined as the change in the chemical potential of monocalcium phosphate per unit change in total labile P. Ammonium sulphate applied at 3 cwt. per acre for 5 years in the field to Lulsgate series soils significantly increased P soluble in o.orM-calcium chloride and o.g~-sodium bicarbonate, labile P, yield and P uptake in the greenhouse.