enables us to distinguish between adaptive and non-adaptive explanations? To answer this question, let us look at two illustrative examples that are more challenging than leafy coverage in insects: the evolution of tameness and religion. Sheep, goats, cows, cats, and dogs are all domesticated animals, created by the forces of artificial selection. All have been transformed from a wild type to an animal that not only lives with us, but sometimes lives for us as food. All are more relaxed, less fearful, and less stressed in the presence of humans than their wild ancestors. Many of these animals seek human companionship. These are the trademark features of tameness. They are also consistently associated with other features that never entered into the breeder’s calculations: floppier ears, curlier tails, more mottled fur, greater sensitivity to human communication, reduced response to predators, earlier sexual maturation, smaller brains, and higher levels of serotonin — a chemical messenger of the brain that regulates self-control. Some of these features appear directly relevant to tameness, whereas others appear entirely irrelevant. For example, serotonin is critically linked to self-control which is critically linked to an animal’s ability to suppress aggression when threatened, which is critically linked to building a life with humans. Mottled fur is not critically linked to any of these benefits. Domestication leads to a pastiche of characteristics, some indicative of the domesticator’s goals and others orthogonal to it. How does the process of domestication, and artificial selection in particular, generate both desired and unanticipated traits? In most cases of animal domestication, we know little about how the wild type changed because the only available information is either anecdotal or based on loose archaeological reconstructions. Consider the domestication of dogs from wolves, and especially the variability among dog breeds. Though it is clear that humans