ConclusionAlthough the full set of coffee agroforests examined in this studyappears to conserve comparable species richness to nearby nativeforests, the species that are being conserved are different. Coffeeagroforests had a lower proportion of trees of conservation con-cern, a higher proportion of pioneer trees, were dominated by Ingaspp., harbored lower tree species diversity at the plot level, andwere composed of different tree species compared to native forests.Farmers reported a sharp preference for Inga spp. and, as a result,this genus accounted for 45% of all trees in farms, a sharp increasefrom its occurrence of 11% in forests.Our study suggests that the promotion of coffee agroforestryshould include workshops in coffee farming communities that dis-courage the replacement of diverse canopies by Inga spp. and thatencourage tolerance for a greater number of adult tree species andtheir recruits. Strategies should also direct the establishment of cof-fee agroforests in replacement of land uses and land covers poorerin floristic diversity and structural complexity; ideally, coffee agro-forestry would not replace forests, but rather crop monoculturesand