Complexity in an obligate mutualism: do fungus-gardening ants know what makes their garden grow?

Fungus gardening ants make clear choices among fungal substrates (food for their fungus). It has been proposed, but never demonstrated, that these ants are collecting the best for their symbiotic fungus and the production of ant biomass (fitness). The goal of this study was to determine whether preferred substrates lead to higher fitness in the attine, Trachymyrmex septentrionalis. Preferences exhibited by foragers were established. Colonies were fed a single substrate or a mixture of substrates during the entire course of the experiment, which ended when sexual offspring appeared in the nest. The response variables were numbers and weights of ant offspring and the chitin content of fungus gardens. Preference was not strongly related to fitness. The preferred oak catkins produced the highest amounts of ant and fungal biomass, but the ants collected much more material than needed, which indicates that forager activity is decoupled from fitness. The preferred caterpillar feces were rejected shortly after the feedings began. The unpreferred oak leaves were just as effective at producing ant and fungal biomass as catkins. Leaves are possibly unpreferred because they are expensive to cut. The unpreferred huckleberry flowers were inferior but did not cause rejection behavior. The mixed diet was just as productive as catkins or leaves. This study indicates that foragers possess a default mechanism to prefer catkins and frass, which can be quickly changed if substrates are bad. In contrast, there does not appear to be a similar mechanism causing substrates to become preferred quickly.     

Mating frequency, colony size, polyethism and sex ratio in fungus-growing ants (Attini)

The mating frequency of queens was estimated for eight attine ant species, Myrmicocrypta ednaella, Apterostigma mayri, Cyphomyrmex costatus, C. rimosus (four lower attines), Trachymyrmex isthmicus, Serico-myrmex amabalis, Acromyrmex octospinosus and Atta colombica (four higher attines), and correlated to colony size, worker polyethism, and sex ratio. Mating frequency was calculated from within-colony relatedness estimated by CAP-PCR DNA fingerprinting. Most queens of lower attines and T. isthmicus mated with only one male, while those of the three higher attines mated with multiple males. Mating frequency was positively correlated with colony size. Polyethism among workers was dependent on worker age in lower attines but on body size in higher attines, suggesting some correlation between mating frequency (i.e., within-colony gene diversity) and caste complexity. The sex ratio was biased toward females in species where the mating frequency equaled one, but toward males in species where the mating frequency was greater than two. Changing in nest site from ground surface to deep underground may have facilitated the evolution of large colony size in Attini, and this may have resulted in the evolution of polyandry (a queen mates with multiple males). With the evolution of polyandry in higher attines, Atta and Acromyrmex in particular have generated high genetic diversity within their colonies and complex social structures. Received: 26 October 1999 / Revised: 25 May 2000 / Accepted: 24 June 2000