Yield of arabica coffee (Coffea arabica) of grafted onto robusta coffee (Coffea canephora Pierre ex Froehner) Apoatã IAC 2258 was evaluated during six years of intercropping with five leguminous species: sunn hemp (Crotalaria juncea L.), Crotalaria spectabilis Roth., dwarf velvet bean (Stizolobium deeringeanum Bort.), soybean Glycine max (L.) Merryl] and pigeon pea [Cajanus cajan (L.) Millsp.] in the Western region of São Paulo State, Brazil, from 1989 to 1995. Leguminous species were sown 50 cm away from coffee plants and incorporated into the soil flowering stage. The treatments were arranged in randomized complete block design with five replicates. Coffee yield was not affected by Crotalaria spectabili Roth, sunn hemp, dwarf velvet bean and soybean IAC 9 as interrow crop. After stamping back the trees only the Crotalaria spectabilis Roth did not affect yield. Stem diameter and yield were significantly decreased by pigeon pea intercrop. The highest amount of legume dry biomass was produced by pigeon pea and sunn hemp. Linear correlation analysis showed that coffee yield was inversely correlated with leguminous dry biomass and positively correlated with coffee stem diameter at the harvest year.
coffee; Coffea arabica; C.canephora leguminous species; intercropping