All farmers actively or inadvertently optimize their crops and farming practices to achieve the best possible harvest, or at least has a strong incentive to do so. Fungus farming ants are no exception as a more vigorous fungal symbiont would obviously benefit the ants. The ants rely on their symbiotic fungus capability to break down structural compounds in the plant material brought back by the ants to manure their fungus. The symbiotic relationship with a fungus has thus allowed the ants to explore a food source otherwise inaccessible to them. The conspicuous leaf-cutter ants in the genus Atta build huge nests displacing several cubic meters of soil, whereas lower attine genera such as Cyphomyrmex and Mycocepurus have small nests with a fungus garden the size of a table-tennis ball. Only the leaf-cutter ants are specialized on using fresh leaves as substrate for their fungus gardens, whereas the more basal attine genera use substrates such as flowers, plant debris, small twigs, insect feces and insect carcasses. This diverse array of fungal substrates across the attine lineage implies that the symbiotic fungus needs different enzymes to break down the plant material that the ants provide or different efficiencies of enzyme function. Fungal enzymes that degrade plant cell walls may have functionally co-evolved with the ants in this scenario. We explore this hypothesis with direct measurements of enzyme activity in fungus gardens in 12 species across 8 genera spanning the entire phylogeny and diversity of life-styles within the attine clade. We find significant differences in enzyme activity between different genera and life-styles of the ants. How these findings relate to attine ant coevolution and crop optimization are discussed.