This dissertation asks what young children do with style at a time when their social and linguistic worlds begin to expand beyond the home, into the peer group. Grounded in a yearlong ethnography of a preschool classroom in the San Francisco Bay Area, I show how play moved gradually from indoors to outdoors as children got older. This shift reflected increasing social and physical autonomy, bringing about new kinds of relationships--like having a best friend--and new ways of interacting with the peer group, like excluding someone, being private, and withholding information. Given expressivity's role in play, I first consider what two children did with expressive vocalizations, direct representations of internal states and moods. I find that vocalization type differed by play practice, getting us towards how speaking styles might be emerging in play at this age. From this interspeaker analysis, I then ask whether children also displayed intraspeaker variation, focusing on the role of laughter in interaction. Considering not just laughter's function but its many forms, I find that one child engaged in the most acoustically extreme types only with her best friend, compared to her interactions with adults. I argue that laughter serves an important function in peer interaction as a way of performing best friend intimacy and resisting adult authority. As this child had access to two geographic varieties of English, I expand this analysis to ask whether she showed any segmental variation alongside the already established variation in laughter. Notably, I find that she followed the same pattern, only shifting to a California dialect style with her best friend, maintaining a British English style with both an American and a British adult. This provides evidence that segmental features are implicated in structured and meaningful expressive variation at this age and calls into question the relationship between affect, style and dialect acquisition. I ask what this might tell us about the more basic social meanings underpinning linguistic variation by considering how one girl varied degree of engagement in the prenasal split-- a local California feature--to take up a "playful" voice in contrast to a "bossy" voice. From here, I stay with the prenasal split to ask what a larger cohort of children did with this feature. I find that variation was conditioned not only by age and gender, but also by children's play practices. Pulling all these analyses together, I propose that development is an indexical process. Through accessing new types of affect and attitude in play, children engage in new types of relationships, which, in turn, (re)structures the emerging peer group. This expansion in the stylistic landscape brings about increasingly complex and nuanced social meanings. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://bibliotheek.ehb.be:2222/en-US/products/dissertations/individuals.shtml.]