Abstract: Neighboring territory owners are often less aggressive toward each other than to strangers (“dear enemy” effect). There is, however, little evidence for territorial defense coalitions whereby a neighbor will temporarily leave his/her own territory, enter that of a neighbor, and cooperate in repelling a conspecific intruder. This is surprising, as theoreticians have long posited the existence of such coalitions and the circumstances under which they should evolve. Here we document territorial defense coalitions in the African fiddler crab Uca annulipes, which lives in large colonies wherein each male defends a burrow and its surrounding area against neighbors and “floaters” (burrowless males). Fights between a resident and a floater sometimes involve another male who has left his territory to fight the floater challenging his neighbor. Using simple experiments, we provide the first evidence of the rules determining when territorial coalitions form. Our results support recent models that suggest that these coalitions arise from by‐product mutualism.

Abstract: In both species of zebra, breeding males had higher urinary androgen concentrations (ng androgens/mg Cr) than did non-breeding bachelor males (30.0 +/- 5.0 (N = 9) versus 11.4 +/- 2.8, (N = 7) in the plains zebra; 19.0 +/- 2.2 (N = 17) versus 10.7 +/- 1.2 (N = 14) in the Grevy's zebra). In the more stable family structure of the plains zebra (single male non-territorial groups) variations in androgen concentrations could not be ascribed to any measured variable. In the Grevy's zebra, androgen values were significantly lower in samples taken from territorial (breeding) males which had temporarily abandoned their territories (N = 4) and the urinary androgen concentration for a male on his territory was negatively correlated with the time since females last visited the territory.