FORAGER POLYMORPHISM, SIZE-MATCHING, AND LOAD DELIVERY IN THE LEAF-CUTTING ANT, ATTA-CEPHALOTES

1. This study examined the importance of forager polymorphism and divi sion of labour among foragers of different size for the economics of l oad delivery in a leaf-cutting ant, Atta cephalotes (L.). I collected A.cephalotes foragers coming down trees carrying leaf fragments to eva luate the degree of match between forager mass and the density (mass p er unit area) of leaves being cut, and to quantify how this match affe cts whether the mass of leaf fragments cut by the ants are within the range which maximizes the rate and efficiency of load delivery. 2. For agers ranged 23-fold in mass (1.4-32.1 mg). On average, larger workers cut at denser leaf sources. Leaf fragment area increased with ant mas s, but relative area (fragment area/ant mass) decreased with ant mass. The density of a leaf type had little or no effect on the area cut by ants of a given size. As a result, ants of a given mass cut heavier f ragments from the denser leaves. The effect of leaf density, however, was partly counteracted at the colony level by recruitment of larger a nts, which cut smaller area fragments relative to their body mass, to cut at denser leaf sources. 3. Despite a fairly high variance in the r elationship between ant mass and fragment mass, overall 87% of the lad en ants (74-100% for different trees) carried leaf fragments in the 1. 5-6 times body mass range. Earlier studies indicate that loads in this range yield the highest biomass transport rate and transport efficien cy. Thus, the variance falls within bounds such that it has little eff ect on load transport efficiency. Having a broad range in optimal load mass may be considered an adaptation to the expected variability in l oad masses. 4. If there were no correlation between ant mass and leaf density, mismatches between ant mass and load mass would be more commo n than observed. Thus, size-matching of larger workers to cut denser l eaves increases the rate and ergonomic efficiency of load delivery.