Ants at Work: How an Insect Society Is Organized

""The basic mystery about ant colonies,"" begins Gordon, who teaches at Stanford, ""is that there is no management."" How, then, do colonies exhibit such high degrees of organization? To answer that question, Gordon has spent 17 summers studying harvester ants in a ""small patch"" of the Arizona desert. This report on that research is an accessible but often dry mix of science writing, memoir and speculation. ""The first time I did this experiment, I used five sets of neighboring colonies. Each set included one enclosed colony and three or four neighbors...."" Thoreau this isn't, but neither is it pure number-crunching. Gordon invigorates her text through bone-clean prose and a welcome sense of humor (in long-sleeved shirt, curtained cap and big sunglasses, ""I look rather like an insect myself""). Gordon's experiments, which concerned numerous aspects of colony life, including their growth and functioning, and relations between colonies, have added greatly to our understanding of ants. Who knew, for instance, that, among ants that work outside the nest, a nest maintenance worker might switch tasks to patrol or forage, but that new maintenance workers come from inside the nest? Probably no one, until Gordon, as she recalls, was able to beat the desert heat and to mark ants, for observational tracking, by slowing them down using an ice cream-making machine. Gordon solved the opening mystery by finding that ant colonies exhibit behavior similar to that of other complex systems: ""Fairly simple units generate complicated global behavior."" She explains that it is the ""pattern of interactions"" among ants, ""not the signal in the interaction itself [that] produces the effect."" So, she concludes in this crystalline work, by studying ants, ""we see how the layers of a natural system fit together."" Drawings throughout. (Oct.)