Patch Size and Composition of Social Groups in Wintering Tufted Titmice

Abstract

Formation of small, non-kin flocks in winter is typical of many resident passerines of the temperate zone (e.g. Ekman 1989). The adaptive significance of such aggregations is usually explained in terms of better predator detection and improved foraging success for flock members (Thorpe 1963, Pulliam 1973). If kin-based aggregations in winter are formed by means of delayed dispersal of young, additional benefits may arise for both adult and first-year group members (Emlen 1997). Ekman and Rosander (1992) suggested that parental control of natal dispersal could be the main factor determining the size and composition of winter groups. According to their model, in the face of limited resources, a territorial mated pair will increase their level of aggression toward non-kin flock members, forcing them to leave. By contrast, subordinate kin (e.g. offspring) will be tolerated to a greater extent by the territorial pair as long as resources are sufficient. When adults cannot afford to share food even with their own offspring, they will exclude such offspring from the flock as well (see Ekman et al. 1994). This model has been supported by studies of Siberian Jays (Perisoreus infaustus; Ekman et al. 1994, 1996) and Gray Jays (Perisoreus canadensis; Waite and Strickland 1997). The Tufted Titmouse (Baeolophus bicolor) is one of few non-corvid, permanent-resident, temperatezone passerines in which offspring are known to spend the winter with their parents (Grubb and Pravosudov 1994). Because cohesive winter units of titmice may include up to eight members (Nice 1930), which may or may not be related, it remains unclear how often offspring are retained in this species and what factors determine the size and composition of winter groups. In our study sites within the agricultural landscape of central Ohio, titmice are permanent residents of forest fragments and often form small wintering groups of three or four birds. We predicted from Ekman and Rosander's (1992) model for offspring retention that under conditions of low resources, a wintering pair of adult titmice would be more likely to tolerate an offspring than a non-offspring flock member. When a pair of titmice is the only territorial pair in a very small woodlot, which presumably affords low levels of food resources, we expected that any juvenile in the same woodlot