Comparison of six extractants for determining available Phosphorus in soils from the Kingdom of Tonga


  • D.A. McGaveston Soil Bureau, DSIR, Private Bag. Lower Hull, New Zealand
  • J.P. Widdowson Soil Bureau, DSIR, Private Bag. Lower Hull, New Zealand


Six different extractants were evaluated as indices of the available phosphorus of topsoil and subsoil samples of nine Tongan soils. Regression and correlation analysis between percentage yield of green panic (Panicum maximum var. trichoglume) and phosphorus extracted by various method bowed that the Bondorff extractant (0·1 M H2SO4) and 0·5M H2SO4 provided good indices of available P. For all Tongan soils the phosphorus extracted by the Bondorff method gave the best correlation with percentage yield (Y=2·19 (Bondorff P) -0·37 where r = percentage yield of dry matter and P=mg phosphorus extracted from 100 g of oil, r=0·80). For soils derived from andesitic ash both extractants gave good correlations with plant yield (Y = 2·25 (Bondorff P) -1·22, r=0·81 and Y=0·61 (0·5M H2SO4 P)-3·80, r=0·83). From this work it is suggested that the Bondorff reagent and 0·5M H2SO4 extract different proportions of plant available P from the iron phosphate fraction of the andesitic soils and 0·5M H2SO4 also extracts P from calcium phosphates in soils derived from coralline sand.



Research Papers