The theol)' that informs this study is Systemic Functional linguistic theory, The major focus of this study is on the description of the register of play-by-play talk. The data for this study was collected by recording onto cassette tapes several Australian Rugby League commentaries from different radio stations. The data primarily used in this study are two Australian Rugby League radio commentaries: one from the commercial radio station, 2GB, and the other from the Australian national radio station, ABC. Both commentaries took place on Sunday afternoon on the 14th July, 1996, and they focus on the same event: the Rugby League professional match between the North Sydney Bears and the Sydney City Roosters. The research focuses on the language which is used to relay the game, 'play-by-play talk'. Although it is generdlly agreed that there is such a thing as 'play-by-play talk', there is scant research into the specific characteristics of this talk, or into the contextual features which motivate it. Thus, in order to study this specific variety of language, the study begins by making a broad-based description of the textual and material environment of play-by-play talk. It is argued that the textual environment consists of two language activities: the 'Sunday Afternoon Sports Broadcast' and the 'commentary'. The description focuses on the linguistic frames that distinguish these two language activities from those which precede and follow them. The study then constructs an operational definition of play-by-play talk and applies some linguistic criteria to the language of the commentary in order to extract language which satisfies this definition. To this end, the linguistic criterion which proves to be most useful is the successive-temporal conjunction and now. Play-by-play talk is found to be a monologic language variety which enters the talk of the commentary at different points in time and for varying time periods. Once play-by-play talk is extracted from the talk of the commentary, it is analysed from three different grammatical perspectives: the interpersonal, the experiential and the textual. The results of the analyses are quantified and displayed in a series of tables and diagrams, and the patterns of choices which emerge from the analyses are discussed in terms of a description of the register properties of play-by-play talk and in terms of the kind of contextual information construed. Chapters Five, Six and Seven end with a summary profile of the grammatical choices along with a summary of the kind of contextual information these choices construe. Some of the findings from this study are that the subject matter of play-by-play talk centres on a very limited set of activities and participants; that individuals are prioritised; that the activities of the game of Rugby League are construed as active, aggressive, and often violent in nature; that there are few grammatica1ised attitudes; and that the text is structured in a way that mirrors the actions which it is encoding. The approach and findings of this research highlight some ways in which a register may be located, analysed and profiled.