Neto, José Félix de BritoBüll, Leonardo Theodoro [UNESP]da Silva, André Luiz Pereira [UNESP]do Vale, Leandro SilvaRigon, João Paulo Gonsiorkiewicz [UNESP]2018-12-112018-12-112017-04-01Australian Journal of Crop Science, v. 11, n. 4, p. 491-495, 2017.1835-27071835-2693http://hdl.handle.net/11449/179721The current study aims to determine critical P levels in different soil classes and their availability for castor bean initial growth. Extraction was done using Mehlich-1 and ion-exchange resin. The experiment was conducted in a greenhouse, in a randomized blocks design with the application of five P doses, which were previously set for each of the four experimental soils (TCo: Chromic Orthic Luvisol; CXbe: Eutrophic Haplic Cambisol; CXve: Typical Eutrophic Cambisol; RY: Fluvic), with four repetitions for 40 days. In the end of the experiment, plants were harvested and shoot dry mass was weighted. At the time plants were collected, samples were gathered to determine soil P availability through Mehlich-1 and ion-exchange resin. Mehlich-1 was more efficient in phosphorus recovery in the following soils: TCo, CXbe and CXve, regardless of applied dose. Critical P levels ranged from 21.33 to 416.64 mg dm-3 due to the clay content in Mehlich-1, and from 30.70 to 177.55 mg dm-3 in resin. MPAC presented positive coefficients regarding soil clay content. The highest castor bean shoot dry mass production was found in CXve and TCo soils. It ranged from 34.65 to 33.55 g vessel-1 in the treatment containing the lowest P dose applications (87.94 mg dm-3 and 104.33 mg dm-3, respectively).491-495engAdsorptionAluminumExtractorsNutritionPrecipitatesArtigo10.21475/ajcs.17.11.04.7812Acesso restrito2-s2.0-850445921281090072947808223