This study is predicated on the assumption that teaching is an intentional behaviour and that a study of teaching therefore requires attention to teachers’ thoughts as well as their observable actions. It is an attempt to describe and account for: (a) the cognitive activities involved in teacher planning; (b) the contrastive features of teachers’ attention to and thoughts about elements of lessons; and (c) the relationship between planned and actual lessons. Data concerning these aspects of teaching were gathered from three experienced classroom teachers and three student teachers who were each asked to teach three small-group discussion lessons. Several data collection and analysis procedures were developed to help the teachers report fully on the focus of their thoughts prior to and in the course of their lessons, and to allow the researcher to categorise accurately the content of their self-reports. Pre-lesson planning thoughts were captured by having teachers make a written ‘stream of consciousness’ record of emerging thoughts, and a ‘stimulated-recall’ procedure was used to help them identify incidents in the course of lessons during which their attention and thoughts had been consciously registered. Additional data about the planning undertaken for specific lessons were obtained from questionnaire responses, while a structured interview technique was used to identify teachers’ perceptions of ‘typical’ features of their conscious experience of teaching. A comprehensive system for categorizing these data was developed and, as a means of ensuring that teacher self-reports were accurately categorized, the researcher systematically discussed their meaning with the teachers who were asked to confirm, disconfirm and where necessary correct the researcher’s interpretations. The analysis of the data obtained revealed a number of significant similarities, differences, consistencies and inconsistencies in features of the teachers’ pre-lesson planning, their within-lesson attention and thoughts and plan-reality relationships. In accounting for these contrastive features, a number of explanatory propositions were formulated. These highlighted: (a) reasons why planning is likely to be undertaken before lessons; (b) the problem-solving nature of the planning process; (c) the contribution that factors internal and external to teachers may make to the nature of the planning process and the content of their plans (e.g. tolerance for uncertainty, teaching theories, information processing capabilities, availability of time and material resources); (d) factors that influence the focus of teachers’ within-lesson thoughts (e.g. the directive influence of plans, the ‘press’ of immediate events, significance and meaning attached to learner behaviours); and (e) the variables that influence teachers’ efforts to monitor learner behaviour (e.g. the availability, ambiguity and objectivity of behavioural cues). Educational implications that follow from these propositions centre on two recommendations. Firstly, those who wish to influence the actions of classroom teachers need to be aware of the manner in which the thoughts of teachers shape their overt behaviours, and to take account of factors that impose constraints on the capacity of teachers to attend to teaching situations in particular ways. Secondly, it is suggested that teachers will benefit from having opportunities to analyze, evaluate and experiment with a variety of ways of thinking about teaching situations.