Identifying and describing children's play behaviours is an important component of evaluating child development. The Behaviour Mapping Schedule is a direct observational tool which aims to describe and quantify children's play behaviours but is yet to undergo reliability testing. This study aimed to determine the intra- and inter-rater reliability of the Behaviour Mapping Schedule. Twelve children aged 3-5 years were each video recorded for one 20-minute playtime period at a purposively selected Community Children's Centre in Adelaide, South Australia. The video recordings were coded independently by two raters against 23 behaviour codes. Intraclass correlation coefficients (ICCs) were calculated. Intra-rater ICCs for nearly 70% of the behaviour codes were considered 'excellent'; likewise, for inter-rater ICCs on more than 50% of the behaviour codes. Overall, the Behaviour Mapping Schedule is a reliable tool for observing children's play behaviour; however, additional training resources may be useful to further strengthen inter-rater reliability.